The Oxford Handbook of Ethics at the End of Life 9780199974412, 9780199976263, 9780190628642, 9780199983643, 0199974411

Citation preview

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

E T H IC S AT  T H E E N D OF  L I F E

The Oxford Handbook of

ETH ICS AT THE

END OF LIFE Edited by

STUART J. YOUNGNER and

ROBERT M. ARNOLD

1

3 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America. © Oxford University Press 2016 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. Library of Congress Cataloging-​in-​Publication Data Names: Youngner, Stuart J., editor. | Arnold, Robert M., editor. Title: The Oxford handbook of ethics at the end of life / edited by Stuart J. Youngner and Robert M. Arnold. Description: New York : Oxford University Press, 2016. | Series: Oxford handbooks | Includes bibliographical references and index. Identifiers: LCCN 2016003939 | ISBN 978–0–19–997441–2 (hardcover : alk. paper) | ISBN 978–0–19–997626–3 (ebook (updf)) | ISBN 978–0–19–062864–2 (ebook (epub)) | ISBN 978–0–19–998364–3 (online content) Subjects: LCSH: Terminal care—Moral and ethical aspects. | Death—Moral and ethical aspects. | Medical ethics. Classification: LCC R726.8 .O895 2016 | DDC 179.7—dc23 LC record available at http://lccn.loc.gov/2016003939 1 3 5 7 9 8 6 4 2 Printed by Sheridan Books, Inc., United States of America

Contents

Acknowledgment  Contributors Introduction  Stuart J. Youngner and Robert M. Arnold

ix xi 1

SE C T ION I   C L I N IC A L A N D L E G A L I S SU E S  1. Legal Issues in Death and Dying: How Rights and Autonomy Have Shaped Clinical Practice  Alan Meisel 2. “So What Do You Want Us to Do?” Patients’ Rights, Unintended Consequences, and the Surrogate’s Role  Mark P. Aulisio

7

27

3. Death at the Beginning: The Neonatal Intensive Care Unit  Renee D. Boss

42

4. Dying Children and the Kindness of Strangers  John D. Lantos

55

5. Medical Futility and Potentially Inappropriate Treatment  Douglas B. White and Thaddeus M. Pope

65

6. Conscientious Objection  Mark R. Wicclair

87

7. Continuous Sedation at the End of Life  Sigrid Sterckx and Kasper Raus 8. The Ethics of Medically Assisted Nutrition and Hydration at the End of Life: Separating the Wheat from the Chaff   Daniel P. Sulmasy

109

126

vi   Contents

9. Disorders of Consciousness and Neuro-​Palliative Care: Toward an Expanded Scope of Practice for the Field  Joseph J. Fins and Maria G. Master 10. Ethical Issues in Prognosis and Prognostication  Alexander K. Smith and Paul Glare

154 170

SE C T ION I I   T H E OR E T IC A L , C U LT U R A L , A N D P SYC HO S O C IA L I S SU E S  11. The Smell of Chlorine: Coming to Terms with Death  Stuart J. Youngner

193

12. Talking and Working with Dying Patients: True Grief and Loss  Lisa Humphrey

202

13. The Nature of Suffering  Eric J. Cassell

216

14. On Our Difficulties Speaking to and About the Dying  David Barnard

227

15. The Cost of Dying Among the Elderly in the United States:   Ethical Issues  Susannah L. Rose and Janelle Highland

242

16. Death, Dying, and the Disabled  Anita Silvers and Leslie P. Francis

260

17. The Effect of Social Media on End-​of-​Life Decision Making  Jessica Berg

279

18. Cultural Factors  Megan Crowley-​Matoka

291

19. Ethnicity as a Factor  Kimberly S. Johnson and Ramona L. Rhodes

308

20. Reframing Care in End-​of-​Life Care: Helpful Themes from a Catholic-​Christian Understanding of Death  Michael McCarthy and Mark Kuczewski

330

Contents   vii

SE C T ION I I I   P H YSIC IA N - ​A S SI S T E D DE AT H  21. Physician-​Assisted Death in the Netherlands  Gerrit Kimsma

343

22. The Case Against Physician-​Assisted Suicide and Euthanasia  Ira Byock

366

23. Goodbye, Thomas: The Case for Physician-​Assisted Dying  Margaret P. Battin

383

24. Depression and the Desire to Die Near the End of Life  Nathan Fairman and Scott A. Irwin

389

SE C T ION I V   T H E E M E RG E N C E OF PA L L IAT I V E C A R E A N D HO SP IC E  25. Hospice and Palliative Care: Developments, Differences, and Challenges  David Clark 26. Potential Perils to the Promise of Specialty Palliative Care  Robert  M. Arnold

409 425

27. Marketing Palliative Care  439 Bridget Tracy and Rolfe Sean Morrison Index  447

Acknowledgment

The editors gratefully acknowledge the assistance provided by Barbara Juknialis in preparing this volume.

Contributors

Robert M. Arnold is the Leo H. Criep Chair in Patient Care and Professor of Medicine at the University of Pittsburgh Medical Center Montefiore Hospital. Mark P. Aulisio is the Susan E.  Watson Professor and Chair of the Department of Bioethics at Case Western Reserve University. David Barnard is the Miles J. Edwards Chair in Professionalism and Comfort Care and the Assistant Vice Provost for Interprofessional Education at Oregon Health & Science University’s Center for Ethics in Health Care. Margaret P. Battin is Distinguished Professor of Philosophy at the University of Utah. Jessica Berg is Professor of Law and Co-​Dean of the School of Law at Case Western Reserve University. Renee D. Boss is Assistant Professor of Neonatology at The Johns Hopkins Berman Institute of Bioethics. Ira Byock is Professor of Medicine at the Geisel School of Medicine at Dartmouth and Director of Palliative Medicine at Dartmouth-​Hitchcock Medical Center. Eric J. Cassell is Emeritus Professor of Public Health at Weill Medical College of Cornell University, Adjunct Professor of Medicine at McGill University, and an attending physician at New York Presbyterian Hospital. David Clark is Director of Interdisciplinary Studies at the University of Glasgow. Megan Crowley-​Matoka is Assistant Professor of Medical Humanities and Bioethics at Northwestern University Feinberg School of Medicine. Nathan Fairman is Assistant Professor in the Department of Psychiatry and Behavioral Sciences at UC Davis Health System. Joseph J. Fins is Chief of the Division of Medical Ethics and the E.W. Davis Professor of Medical Ethics at Weill Cornell Medical College. Leslie P. Francis is Professor and Chair of the Department of Philosophy at the University of Utah. Paul Glare is Chief of the Pain and Palliative Care Service at Memorial Sloan-​Kettering Cancer Center.

xii   Contributors Janelle Highland is a graduate student in the Harris School of Public Policy and the Graduate Program in Health Administration and Policy at the University of Chicago. Lisa Humphrey is the Director of Inpatient Services and Co-​Director of the Pediatric Palliative Care Program at Nationwide Children’s Hospital in Ohio. Scott A. Irwin is Director of Psychiatry & Psychosocial Services and Patient & Family Support Services at the UC San Diego Moores Cancer Center. He is also Director of Palliative Care Psychiatry at the UC San Diego Health System and Associate Clinical Professor of Psychiatry at the UC San Diego School of Medicine. Kimberly S. Johnson is Assistant Professor of Geriatrics at Duke University. Gerrit Kimsma is a Professor in the Department of Metamedicine at Vrije Universiteit in The Netherlands. Mark Kuczewski is The Fr. Michael I.  English, SJ, Professor of Medical Ethics and Director of the Neiswanger Institute for Bioethics & Health Policy at Loyola University Chicago. John D. Lantos is Director of the Children’s Mercy Bioethics Center at the University of Missouri—​Kansas City School of Medicine. Maria G. Master is a psychiatrist affiliated with New York-​Presbyterian Hospital/​Weill Cornell Medical Specialties. Michael McCarthy is Assistant Professor in the Neiswanger Institute for Bioethics & Health Policy at Loyola University Chicago. Alan Meisel is the Dickie, McCamey & Chilcote Professor of Bioethics, and Professor of Law and Psychiatry at the University of Pittsburgh School of Law. Rolfe Sean Morrison is Professor of Geriatrics and Palliative Medicine, Medicine, and Anesthesiology at The Mt. Sinai Hospital in New York. Thaddeus M. Pope is Director of the Health Law Institute at Hamline University School of Law. Kasper Raus has a Master’s degree in philosophy from Ghent University. His current research focuses on the ethics of end-​of-​life decisions in general, and, more specifically, on the practice of terminal (or palliative) sedation. Ramona L. Rhodes is Assistant Professor of Medicine at UT Southwestern Medical Center. Susannah L. Rose is a member of the Center for Ethics, Humanities and Spiritual Care at the Cleveland Clinic Foundation. She is also an Assistant Professor at the Cleveland Clinic Lerner College of Medicine of Case Western Reserve University.

Contributors   xiii Anita Silvers is Professor and Chair of the Department of Philosophy at San Francisco State University. Alexander K. Smith is Assistant Professor of Medicine in the Department of Medicine, Division of Geriatrics, at the University of California—​San Francisco. Sigrid Sterckx is a Professor of Ethics and Political and Social Philosophy at Ghent University. Daniel P. Sulmasy is Professor of Medicine at the University of Chicago School of Medicine MacLean Center for Clinical Medical Ethics. Bridget Tracy is Associate Professor of Palliative Medicine at Mt. Sinai Hospital in New York. Douglas B. White is Associate Professor of Critical Care Medicine at the University of Pittsburgh Medical Center. Mark R. Wicclair is Professor of Philosophy at West Virginia University. Stuart J. Youngner is Professor of Bioethics and Psychiatry at Case Western Reserve University School of Medicine.

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

E T H IC S AT  T H E E N D OF  L I F E

I n t rodu ction Stuart J. Youngner and Robert M. Arnold

This volume explores the topic of death and dying during the fifty to sixty years spanning the late twentieth and early twenty-​first centuries, primarily in the United States. This has been a time of great change in our society. An explosion of technology radically changed the nature of medical practice and the way we die. Simultaneously, traditional structures of power were successfully challenged by rights claims—​from African Americans, women, gays, students, and, most relevant here, patients. Respecting patients’ values increasingly has been recognized as the most essential moral component of clinical decision-​making. We soon discovered that technology’s promise to save and improve life had a dark side. Too often, it only served to prolong the dying process, making death appear to many as the least worst alternative. For the first time in their history, human beings had the ability to control the timing of death in almost all instances. With this ability came a responsibility that is both awesome and inescapable. How we understand and manage this responsibility is a theme that runs through every chapter in this book. In Section I, Alan Meisel examines how the law has helped shape clinical practice, emphasizing the roles of rights and patient autonomy. He illustrates how key court cases in the second half of the past century have led to a general consensus about the legality of withholding and withdrawing life-​sustaining treatment. Meisel touches briefly on the recent legal recognition of physician-​assisted suicide in some jurisdictions in our country. Mark Aulisio then points out that many, if not most, decisions at the end of life are made when the patients are no longer capable of exercising their autonomy. This reality has spawned the notion of advance directives, whose inadequacies have led to the inevitable reliance on surrogate decision-​making—​which is itself fraught with problems. Mark Wicclair addresses the controversial issue of conscientious objection by health care professionals. What do we do when patients’ values conflict with those of health professionals? Wicclair explores the possibilities, calling for a “reasonable accommodation” approach. Section II addresses specific clinical issues. Renee Boss examines the problems that arise for families and health professionals when technological advances permit the survival of infants who are born prematurely or have severe congenital anomalies. Despite

2    Stuart J. Youngner and Robert M. Arnold many instances of success, neonatal complications sometimes result in chronic illness, serious disability, and a shortened life span, thus posing difficult choices for families, health professionals, and the larger society. John Lantos explores the tragic situation of death and dying in children. Drawing on fiction, poetry, film, and memoir, he describes a “kindness of strangers” approach to caring for dying children and their families in tertiary care medical centers. Sigrid Sterckx and Kasper Raus explain and analyze continuous sedation as a way to relieve suffering at the end of life. They discuss the problem of terminology and definition and the complicated moral relationship between continuous sedation and the restriction of medically provided nutrition and hydration. Next, Daniel Salmasy discusses in detail the medical provision of fluids and nutrition. He discusses the technical aspects of providing this intervention, the clinical effects of its application or limitation, and the moral problems engendered by different courses of action. Alexander Smith and Paul Glare write about the problem of prognostication in patients who are thought to be dying. Despite an inevitable level of uncertainty in most cases, they highlight the importance of communicating with patients about prognosis and how prognostic information should be used. Finally, Douglas White and Thaddeus Pope address the ongoing problem of futility and medical power (a chapter that is also related to issues of surrogate decision-​making and conscientious objection). They summarize the circumstances and arguments that have been proposed to justify the limitation of life-​sustaining interventions without patient or family consent, emphasizing the value-​laden nature of most of these judgments. They then offer suggestions for how physicians might respond to requests for treatment they deem to be medically or ethically inappropriate. Section III examines psychosocial and cultural issues. Stuart Youngner and Lisa Humphrey share their personal experiences as clinicians providing care for dying patients and their families. Youngner tells his story as a psychiatrist coming of age who is trying to be more honest with dying patients and their families as they face death together. Youngner blinks and turns his career path from emotionally draining confrontations with death and dying to a more intellectual pursuit—​the consideration of the morality of limiting life-​sustaining treatment in the age of medical technology. Humphrey also tells a personal story about her training and practice as a pediatric palliative care specialist. She shares daily confrontations with tragedy and, unlike Youngner, how she successfully developed emotional coping tools. She poignantly labels one such tool as “tagging your baggage.” David Barnard and Megan Crowley-​Matoka look at our death-​denying society at a less personal level. Barnard examines how society has struggled with the death problem, attributing much of it to a manifestation of shame “induced by exposure to, or reminders of, the physical body—​its functions and disintegration—​that the processes of civilization have progressively kept out of sight and out of consciousness.” Crowley-​Matoka uses two classic ethnographies of dying in American hospitals, spaced four decades apart, as a lens to compare how key cultural features of death and dying have and have not shifted over “a particularly critical period in the history of US health care.” Next, Eric Cassell examines the problem of suffering. He argues that there are not different types

INTRODUCTION   3 of suffering, such as physical, psychological, or existential. There is only suffering, which is something that happens to persons and not to bodies. He concludes that suffering is so pernicious because it attacks one’s integrity as a person and inevitably involves self-​ conflict. Finally, Jessica Berg explores the ethical and legal issues involved in the use of social media to make end-​of-​life decisions. After providing an overview of the various types of social media, she considers the many different ways social media might play a role as we catapult into our electronic future. In Section IV, the authors examine death and dying among various vulnerable populations. Susannah Rose and Janelle Highland consider the elderly. After an examination of the data, they observe that end-​of-​life care preferences in the elderly are varied and that we need to allow for a wide conception of autonomy and choice that is potentially very expensive. They propose a soft paternalism and a system that provides palliative care as the default choice for people with serious and chronic diseases. Anita Silvers and Leslie Francis address persons with disabilities, another vulnerable population. They consider both the intellectually and physically disabled. The problem of free choice is complex and often controversial in a group that has suffered considerable social stigmatization and discrimination. Evaluating depression and decision-​making capacity in the dying is a difficult and often contentious issue. Is depression a natural part of dying? Can and should it be treated? Are all depressed patients incapable of making decisions about their end-​of-​life care? Nathan Fairman and Scott Irwin carefully review these issues and offer suggestions to help guide decision-​making. Kimberly Johnson and Ramona Rhodes examine how race or ethnicity affects access to and the quality of palliative care by focusing on the experiences of African Americans and Hispanics. They suggest strategies for health care providers, organizations, policy-​makers, and researchers on how to improve care for these vulnerable populations. Joseph Fins explains how neuropalliative care can be provided to patients with severe, chronic brain injuries. He addresses the delicate balance between the right to die and the right to care for these unfortunate patients. Michael McCarthy and Mark Kuczewski propose an approach to care of the dying that is critical of the overemphasis on autonomy. They emphasize three aspects of care—​embodied, compassionate, and communal—​that have strong roots in the Catholic-​Christian theological tradition. Section V, on physician-​assisted death (PAD), defined here as physician-​assisted suicide (PAS) and active euthanasia (lethal injection), remains the most controversial choice in end-​of-​life-​care in the United States, despite Alan Meisel’s report in Section I that PAS is now legal in several US jurisdictions. The section begins with a description of PAD in the Netherlands, where it has been practiced, regulated, studied, modified, and eventually legalized. Even if one accepts that PAD is sometimes morally justified, it is questionable whether it can be successfully implemented in a complex political entity such as an entire modern nation. Critics say no, but Gerrrit Kimsma presents a generally optimistic view of what has happened in the Netherlands. Physicians have been at the forefront of the social acceptance of PAD in the Netherlands, and PAD is widely accepted among the public. Ira Byock strenuously disagrees with the Dutch physicians,

4    Stuart J. Youngner and Robert M. Arnold arguing that physicians must never kill patients because doing so would weaken the moral grounding and structural integrity of the medical profession and its service to society. Margaret Battin disagrees with Byock. She reviews and rejects three common arguments against PAD, including Byock’s and two others—​that is, that killing is always wrong and that, if allowed, PAD would lead down an inevitable slippery slope to abuse so dire that it has been called a moral holocaust. Battin defends PAD with two arguments: categorically forbidding PAD is an unconscionable abridgment of one’s liberty in a society that celebrates freedom, and our celebrated right to pursue happiness implies a right to avoid the unhappiness occasioned too often by death in the modern era. The final section, Section VI, explores hospice and palliative care, which many have seen as an answer, if not the answer, to the psychosocical and ethical problems of death and dying in our modern society. David Clark starts the section by discussing the development, similarities, and differences between hospice and palliative care. He describes the emergent clinical evidence produced by the field and its growing professionalization. Robert Arnold expresses a cautionary note about this growing professionalization. He argues that increasing specialist palliative care is not the best way to improve the care of most dying patients. Increasing specialist palliative care may have the unintended consequences of allowing the health care system to continue to deny death and de-​skilling the nonpalliative physicians who provide the bulk of care for dying patients. Sean Morrison suggests a different approach. He argues that the underutilization of specialist palliative care is due to a failure to consider basic marketing principles. His marketing recommendation is that palliative care needs to disassociate itself from the words “death” and “dying.” Have we moved full circle, back to the late 1960s, where encouraging denial of death is again considered the best way to cope with it? This irony speaks to the power of death, high-​tech or otherwise, to frighten us—​perhaps with the result that we act against our own best interests.

Section I

C L I N IC A L A N D L E G A L  I S SU E S

Chapter 1

Legal Issu es i n De at h and Dy i ng  How Rights and Autonomy Have Shaped Clinical Practice Alan Meisel

The development of the law governing end-​of-​life decisions owes much to other social trends that somewhat predated its origins and paralleled its development—​most notably the civil rights and consumer rights movements of the 1950s through the 1970s. Yet without one other factor—​the development of medical technology—​it is unlikely that there would be any significant body of end-​of-​life law today. Without ventilators, dialysis, cardiopulmonary resuscitation, feeding tubes, and all the other death-​postponing devices, technology, techniques, and medications, decisions about whether to keep people alive or allow nature to take its course would not have to be made; instead, nature would take its course as it always had. The confluence of these larger social trends virtually guaranteed that some social institution for resolving disputes would need to be invoked to address those disputes arising from making decisions about end-​of-​life medical treatment. Several social institutions existed to address this need, including the organized medical profession (in the form of the American Medical Association and other medical and health-​care specialty organizations) and organized religion, but both of these lacked the finality—​especially concerning the possibility of legal liability—​that the courts possessed, and thus it largely fell to the latter to define the rights, roles, and responsibilities of those involved in the end-​of-​life decision-​making process, with state legislatures increasingly playing a role. For almost four decades, American courts have been the final arbiter in this process. Other civic, professional, and governmental institutions continue to play a role, but it is ultimately to the courts that we look for definitive guidance. Key players in end-​of-​life dilemmas—​usually the family members of patients whose doctors declined to withhold or withdraw life-​sustaining medical treatment, but occasionally physicians or health-​care managers—​turned to the courts to break impasses that could not otherwise be resolved.

8   Alan Meisel This is not difficult to understand, for courts are able to grant immunity from legal liability that no other social institution (except for another law-​making entity, legislatures) is able to do, and fear of legal liability has been one of the primary drivers, if not the primary one, of physicians’ uncertainty about permitting patients to forgo life-​sustaining treatment. There have not been large numbers of end-​of-​life cases reaching the courts. In the fifteen-​year period between the Karen Ann Quinlan case in 1975 (In re Quinlan 1976)—​ the first true end-​of-​life case—​and the Nancy Cruzan case in 1990 (Cruzan v. Director 1990)—​the first U.S. Supreme Court case to address end-​of-​life issues—​there were fewer than 100 such cases. However, almost all of them were legal landmarks that captured the public’s attention too. Many made the national news—​the covers of the newsweekly magazines, the evening news on the national television networks—​and those that did not, without exception, made local or statewide headlines. They were also the grist for articles in an array of medical journals read by physicians generally and those in the specialties and subspecialties most likely to be involved in end-​of-​life controversies. And, of course, the lawyers who counseled health-​care institutions were well aware of these decisions. The lessons of these court decisions quickly became incorporated into clinical practice, especially in those states in which the cases were decided. But as time went by, legislatures in other states and medical professional associations adopted, respectively, statutes and policies that incorporated the principles established in the court decisions. A Presidential Commission issued a report in 1983 that also reflected the emerging consensus about how end-​of-​life decisions should be made (President’s Commission 1983). Taken together, the cases, statutes, policies, and the commission report gradually affected clinical practice, and, as clinical practice changed, it began to influence the decisions in new cases, thus creating a feedback loop further influencing the development of the law. The first major clinical case to become a legal case was that involving Karen Ann Quinlan, beginning in 1975. Prior to that time, there had been a handful of legal cases in the 1950s and 1960s raising similar—​though not quite the same—​issues, but for one reason or another each case was somewhat idiosyncratic and had little overall influence. The early 1970s witnessed an explosion of legal cases promoting autonomy in medical decision making even before the end-​of-​life cases arose. Gradually building steam for fifteen years, courts decided three important cases in 1972 involving “informed consent” to medical treatment. And at the beginning of 1973, the U.S. Supreme Court issued its opinion in Roe v. Wade—​following in the wake of earlier decisions removing state prohibitions on the sale and use of contraceptives—​recognizing a woman’s right to control her own reproductive matters.

The Karen Quinlan Case The facts of the Quinlan case are well known to many. Those who were alive at the time remember it from television news shows and from Karen Quinlan’s picture on the cover

Legal Issues in Death and Dying    9 of national newsmagazines. It was not merely a legal case; it was a cultural phenomenon. Never before had such a series of events occurred, and they held the country in their grip. Karen Quinlan was a twenty-​one-​year-​old woman who was discovered by friends at a party that they were all attending to be unconscious and not breathing. They summoned emergency medical services personnel, who resuscitated her, put her on a ventilator, and took her to a nearby hospital, to which she was admitted. After she was diagnosed as being in a persistent vegetative state and after her parents came to understand the implications of that diagnosis—​namely, that she was incapable of perceiving her external environment, that her cerebral cortex was no longer functioning and thus she was incapable of thought or feeling, and that she would never regain any of these human capacities and would remain forever unconscious until she died—​they concluded that it was best that her ventilatory support be discontinued and that nature be allowed to take its course. As devout Roman Catholics, they consulted with their priest, who assured them that there was no religious imperative for them to maintain the medical life support, and they asked Karen’s physicians to discontinue it. The physicians, however, would not accede to this request because of their concern that medical ethics required them to do what they could to keep their patient alive and because of their concern about possible criminal liability for homicide if they discontinued life support and Karen died, as her parents intended. Quinlan’s parents had only two choices: to acquiesce in their doctors’ position or to seek to overturn it through judicial proceedings, and they chose the latter. Although the trial court sided with the physicians and refused to order the termination of life support on the ground that to do so would subject the physicians to liability for criminal homicide, the New Jersey Supreme Court, to which the trial court’s decision was appealed, held otherwise. It found that Karen Quinlan had a right to privacy, protected by the U.S. Constitution, which encompassed the right to be free from unwanted medical treatment, and because she was unable to exercise this right personally because of her medical condition, her father, as her legal guardian, was empowered to do so. Additionally, the court mandated that in cases such as this, “physicians should consult with [the] hospital ethics committee and if [the] committee should agree with physicians’ prognosis, the life-​support systems may be withdrawn” (In re Quinlan, 670). And finally, the court made clear that decisions to forgo life-​sustaining medical treatment did not automatically need judicial approval. Instead, the ethics committee’s approval would suffice. The Quinlan case revolutionized legal thinking and clinical practice regarding decisions near the end of life. Before the Quinlan case, a great cloud of legal uncertainty cast a pall over decisions to forgo life-​sustaining medical treatment. Doctors were simply afraid that forgoing treatment, which would result in a patient’s death, would subject them to criminal liability. Although the cloud did not entirely dissipate with the Quinlan decision, it was now clear that there were no simple black-​and-​white rules for making end-​of-​life decisions. And it was equally clear that forgoing life-​sustaining medical treatment would not subject physicians to legal liability if they did it in accordance with the rules that the court established. End-​of-​life decision making was far more

10   Alan Meisel complex a matter than it had previously seemed, and it would become more complex and confused before a consensus emerged and the rules were sorted out. Finally, the New Jersey Supreme Court assumed that hospitals had ethics committees; in fact, most did not. But the mandate to consult with an ethics committee was a strong incentive, first in New Jersey and then elsewhere, for hospitals to establish such committees, and eventually the Joint Commission, which accredits health-​care institutions, established standards for such committees. Thus, although the law never required ethics committees, it played an extremely influential role in their establishment.

To Go to Court or Not The year following the Quinlan decision, another state supreme court had the opportunity to consider a similar matter. This case, Superintendent of Belchertown State School v. Saikewicz (1977) involved a sixty-​seven-​year old man who was a resident of a Massachusetts state institution for the developmentally disabled and who was diagnosed with acute myeloblastic monocytic leukemia. Ordinarily, people with this medical condition would be offered chemotherapy, but Mr. Saikewicz’s doctors recommended against its administration, ostensibly on the ground that because of his profound mental retardation, he would not be able to understand what was being done to him and would have to be physically restrained for treatment to be administered. The most significant aspect of the Saikewicz case for medical decision-​making purposes was the court’s requirement—​directly contrary to that of the New Jersey Supreme Court—​that end-​of-​ life decisions required judicial oversight. In other words, before life-​sustaining medical treatment could be withheld or withdraw, a judge had to approve the decision. Thus within a year of what had seemed to be a clarifying decision by the New Jersey Supreme Court, the situation was once again thrown into confusion by the conflicting aspects of the Massachusetts decision. Only through a series of decisions by both of these courts—​and by state courts in many other jurisdictions, strongly influenced by the President’s Commission Report—​were these conflicts ironed out and a consensus ultimately achieved over the next decade or more. The consensus that evolved was that, ordinarily, decisions to forgo life-​sustaining treatment did not need the approval of a judge, and thus they could be made in the clinical setting between physicians and patients possessing decision-​making capacity or between physicians and close families members of patients who did not possess decision-​making capacity. Beginning in the 1990s, state legislatures began to step into this aspect of end-​of-​life decision making by enacting statutes (referred to as either surrogate decision-​making or family decision-​making statutes) recognizing the legitimacy of this practice and establishing the order of family members who had the authority to make such decisions. These statutes also contained provisions for resolving disputes among family members. By 2012, thirty-​nine states and the District of Columbia had enacted such laws.

Legal Issues in Death and Dying    11

Standards for Surrogate Decision Making Another significant aspect of the Saikewicz case was the court’s holding that decisions to forgo life-​sustaining treatment must be based on the patient’s own wishes. In the case of patients who currently lack decision-​making capacity, the court must find that the patient would have wanted treatment forgone. This is fairly uncontroversial and consistent with the Quinlan decision. However, the Massachusetts court went further by applying this standard even to a person who—​like Mr. Saikewicz—​never possessed decision-​making capacity and thus could never have formed any preferences about treatment. The Massachusetts Supreme Judicial Court persisted in this illogical position in a number of similar cases throughout the 1980s, including at least one involving a patient who was an infant and thus, like Saikewicz, could never have formed any treatment preferences. Eventually, however, as was the case with the question of judicial review, the courts in most states coalesced around a point of view closer to that taken by the New Jersey Supreme Court, namely that in making decisions for patients who lack decision-​making capacity, the surrogate should attempt to discern what the patient would have decided if he or she now possessed it—​what is referred to as the substituted judgment standard. If a patient had never possessed decision-​making capacity, the surrogate should, instead, make a decision reflecting the “best interests” of the patient. Although courts in two states (New York and Missouri) at first took a more extreme position—​that there must be clear and convincing evidence of the patient’s actual wish to forgo life-​sustaining treatment rather than the patient’s probable wish to do so as the substituted judgment standard allows—​the law in both states has subsequently gravitated closer to the mainstream substituted judgment standard, in New York by statute and in Missouri by later judicial decision.

Rejection of Limitations on Forgoing Life-​Sustaining Treatment The New Jersey Supreme Court turned out to be a leader in shaping end-​of-​life law. Each of the court’s efforts to resolve one end-​of-​life dilemma gave rise to others, so that by the mid-​1980s the court had become the national leader in defining end-​of-​life law, which in turn shaped clinical practice. The next New Jersey case after Quinlan was the Conroy case in 1985 (In re Conroy 1985) For present purposes, the significance of Conroy was that it lent judicial support to the rejection of two modes of analysis of end-​of-​life decision making that

12   Alan Meisel had long been employed in the clinical setting to limit the authority of patients or their surrogates to forgo life-​sustaining treatment. The influence of the Conroy opinion cannot be understated. It has been cited by dozens of other judicial end-​of-​life opinions, and it was written about extensively in medical journals. Its direct effect on clinical practice has been substantial, in no small part because of its ratification of the President’s Commission Report by incorporating large parts of its reasoning into the opinion.

Limitation of the State Interests Analysis The first of these is what might be called the “state interests” approach. Prior cases had established a constellation of four state interests to be weighed against an individual’s right to refuse medical treatment and that, in theory at least, could overcome that right. These are the state’s interest in (1) the preservation of life, (2) the prevention of suicide, (3) safeguarding the integrity of the medical profession, and (4) “protecting innocent third parties who may be harmed by the patient’s treatment decision” (In re Conroy 1985, 1223–​1225). Although acknowledging the validity of these state interests in the abstract, the court concluded that they could not overcome the decision of a competent patient (nor, implicitly, the decision of a surrogate of an incompetent patient acting on the basis of the patient’s wishes) in the end-​of-​life context because to allow them to do so would overcome a patient’s fundamental legal right to refuse treatment.

Rejection of the Categorical Approach Another mode of analyzing the propriety of end-​of-​life decisions that had arisen mostly from judicial opinions was what might be termed the “categorical” approach to decision making. Since long before the Quinlan case, theologians (especially Roman Catholic theologians), clinicians, and some courts had employed a number of categories for analyzing the propriety of medical decisions that would have the likely effect of ending a patient’s life. The President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, in its 1983 report titled “Deciding to Forego Life-​Sustaining Treatment,” had thoroughly analyzed and rejected these categories as a useful means for analyzing the legitimacy of end-​of-​life decisions, but the report carried no legal weight. However, its reasoning proved to be extremely influential on the New Jersey Supreme Court, which cited it repeatedly in the Conroy opinion.

Legal Issues in Death and Dying    13 These categories are (1)  active versus passive, (2)  withholding versus withdrawing, and (3) ordinary versus extraordinary.1 The court held that they are irrelevant to end-​of-​life decision making, where the primary focus should be the patient’s wishes and “not the type of treatment involved” (In re Conroy 1985, 1233). “Thus, we reject the distinction that some have made between actively hastening (In re Conroy 1985, 1233–​1234) death by terminating treatment and passively allowing a person to die of a disease as one of limited use in a legal analysis of such a decision-​making situation. Characterizing conduct as active or passive is often an elusive notion, even outside the context of medical decision-​making.” Similarly, the court “reject[ed] any distinction between withholding and withdrawing life-​sustaining treatment” (In re Conroy 1985, 1234), holding that it was a variant of the active/​passive distinction and that while the distinction might be psychologically important, it was not logically sound. Finally, the court found “unpersuasive the distinction relied upon by some courts, commentators, and theologians between ‘ordinary’ treatment, which they would always require, and ‘extraordinary’ treatment, which they deem optional” (In re Conroy 1985, 1234) because “the terms ‘ordinary’ and ‘extraordinary’ have assumed too many conflicting meanings to remain useful. To draw a line on this basis for determining whether treatment should be given leads to a semantical [sic] milieu that does not advance the analysis” (In re Conroy 1985, 1235). The judicial rejection of this form of reasoning has probably had a significant effect on clinical practice. Whereas at one time doctors might have hesitated to begin a patient on dialysis or a ventilator or a feeding tube for fear that once they began they could not stop—​ because although it was acceptable to withhold treatment, it was unacceptable to withdraw it—​now they felt free to begin a trial of treatment to see if it will benefit the patient. If it did not, the treatment can then be withdrawn without fear of legal consequences—​assuming of course that there is valid authorization from the patient or surrogate to do so. Patients who in the past might have benefited from a trial of treatment are no longer deprived of that opportunity for fear that they will be forever bound to continue. The rejection of the ordinary/​extraordinary distinction means that doctors no longer have to engage in ultimately futile discussions about whether a treatment is extraordinary because it is new, or expensive, or technologically sophisticated, and what the meaning of those terms are as applied to a particular treatment. Treatments may be withheld or withdrawn based, instead, on patients’ wishes and the physician’s judgment of their medical efficacy—​criteria of actual relevance to a patient’s well-​being rather than abstruse criteria bearing no such relationship.

Feeding Tubes After Karen Quinlan was weaned from the ventilator in accordance with the New Jersey Supreme Court’s decision, she continued to breathe nonetheless. Although she had

14   Alan Meisel initially been dependent on the ventilator to keep her alive, her respiratory drive had recovered to the point where it no longer was needed. Thereafter, the only medical treatment she required to keep her alive was a surgically implanted feeding tube and, occasionally, antibiotics for the treatment of pulmonary and urinary tract infections. Karen Quinlan died in 1986 while still being fed and hydrated by a feeding tube, and if her parents ever considered asking to have her feeding tube removed, there is no record of that request. However, the families of other patients did seek to have feeding tubes withheld or withdrawn during the period when Quinlan was being kept alive by one (1975–​1986), and considering how many elderly nursing home patients are on feeding tubes, it is probable that the number is quite large. However, there is no judicial record of such a request prior to the Barber case in California in 1982 (Barber v. Superior Court 1983). That case was a criminal prosecution—​to date, the only criminal prosecution for the withdrawal of life-​ sustaining medical treatment. The decision in that case finding that there was no criminal liability on the part of the physician who ordered the tube removed (with the permission of the patient’s next-​of-​kin) stands for the proposition that feeding tubes are medical treatment that may be removed on the same basis as any other medical treatment. This conclusion was reaffirmed in the Conroy case in 1986 and in a number of cases thereafter. Whether or not the provision of nutrition and hydration by medical means is—​for legal purposes—​to be considered medical treatment and whether it may be forgone on the same terms as any other medical treatment was not definitively settled until the decision by the U.S. Supreme Court in the Cruzan case in 1990. Although the Supreme Court did not directly address this issue, the Court assumed that it was a medical treatment (Cruzan v. Director 1990, 279). Furthermore, in a concurring opinion, Justice O’Connor got to the heart of the matter when she commented that whether or not a feeding tube constituted a medical treatment was, for constitutional purposes, irrelevant because tube feeding, where consent had never been given or had been withdrawn, constituted a restraint and intrusion on that individual’s liberty, which violated the due process clause of the Fourteenth Amendment to the U.S. Constitution (Cruzan v. Director 1990, 288). Since the Supreme Court’s decision in the Cruzan case, there has been very little litigation about the legality of withholding or withdrawing feeding tubes (the major exception being the Schiavo case; In re Schiavo 2001). Although withholding and withdrawing tube feeding is now well-​accepted clinical practice among some segments of the medical profession, it is still met with significant resistance by those who hold the position that forgoing tube feeding in the case of those who can no longer be fed orally constitutes “starving the patient to death.”

The Acceptance of Do-​Not-​R esuscitate Orders In the atmosphere of uncertainty about the legal status of forgoing life-​sustaining treatment before, during, and even after the Quinlan case—​that is, from the mid-​1970s

Legal Issues in Death and Dying    15 through the mid-​1980s, or even later—​the uncertain status of forgoing cardiopulmonary resuscitation loomed largest of all. Why this was so is not altogether clear, but it possibly resulted from the fact that, by that time, the administration of CPR to patients suffering a cardiac arrest had become standard operating procedure and that, in order to prevent its administration, efforts had to be made in advance by writing an order not to resuscitate. Such orders are referred to as DNR (do-​not-​resuscitate) orders, or orders not to resuscitate, or “no code,” but their legality was in doubt, at least in the minds of some physicians (Margolick 1982). Consequently, doctors sometimes undertook subterfuges to assure that CPR was withheld from some patients. Some of these procedures—​such as giving oral but not written orders—​ran the serious risk of being misinterpreted, thus resulting in some patients being resuscitated who were not supposed to be and vice versa. This sometimes resulted in, in the words of a grand jury report investigating the use of DNR orders in New York, “shocking procedural abuses” (Sullivan 1982). Although a few courts have had occasion to review and comment on DNR orders, the judicial authority is so thin that the vacuum has largely been filled by state legislative enactment of statutes authorizing either doctors to write DNR orders or patients to issue DNR directives or both. Concerns such as these led to the enactment of detailed statutes governing DNR orders in many states (for the citations to these statutes, see Meisel and Cerminara 2012, Section 6.02[C]‌[4]). Most of these statutes address formal procedural issues such as who may write a DNR order or issue a DNR directive, identification of patients with orders or directives, recordkeeping requirements, revocation of orders and directives, and similar matters. What they fail to address is the most controversial issue, namely the resolution of disputes that arise between family members demanding the administration of CPR and physicians who believe that its administration would be ineffective or even harmful to a patient. (This issue is addressed later under the heading of “Futility”.) What is most noteworthy is that the enactment of this legislation has taken what was once a covert clinical process and legitimated it. Whereas many doctors once feared the legal consequences of authorizing the withholding of CPR—​even though no doctor had ever been held civilly or criminally liable for doing so—​today a code status is routinely sought for virtually all patients admitted to a health care institution, and CPR is routinely withheld from those patients for whom a DNR order has been written.

Advance Directives For a long period of time—​as was the case with DNR orders—​the legal status and effect of advance directives was uncertain. Advance directives are instructions issued by patients while they possess decision-​making capacity providing guidance about how they wish to be treated if they lose decision-​making capacity. The oldest form of advance directive is the living will, first proposed in a law review article in 1969 (Kutner 1969). A living will is a written document stating that if the declarant (i.e., the future

16   Alan Meisel patient) loses decision-​making capacity and needs to be treated, he or she is to be treated in accordance with the instructions in the directive. A very simple (and not very useful) advance directive might state, “If I lose decision-​making capacity and am likely to die even with treatment, I do not want an extraordinary treatment.” A more specific advance directive might, for example, state, “If I lose decision-​making capacity and am suffering from kidney failure, I do not want dialysis administered.” Realizing that it is virtually impossible to anticipate in advance the entire gamut of treatments an individual might wish to decline (or have administered because an advance directive may request treatment as well as refuse it), it soon became clear that issuing such instructions would either be very vague (e.g., no “extraordinary” treatment) or very detailed yet still run the risk of not addressing the kind of treatment that was in fact under consideration. To deal with this problem, consideration began to be given to the use of an existing legal instrument, known as a durable power of attorney, to appoint another person to make health-​care decisions if the declarant no longer possessed decision-​making capacity. The difficulty with both of these forms of advance directives—​living wills and durable powers of attorney—​is that there was no explicit legal basis for them, and hence a physician who made treatment decisions for an incompetent patient on the basis of one of them had some reason to be concerned that such decisions were not legally valid and thus had some reason to fear the imposition of civil or criminal liability. Of course, it was not entirely clear—​indeed, until about 1990, it was fairly unclear—​how decisions for patients lacking decision-​making capacity should be made. There was uncertainty in many states about the authority of family members to make such decisions. There was uncertainty about whether a guardian needed to be judicially appointed to make such decisions, there was uncertainty about whether the surrogate’s or guardian’s decision to forgo life-​sustaining treatment needed judicial approval before it could be implemented, and there was the further uncertainty about whether decisions made on the basis of an advance directive were valid. This massive uncertainty was gradually resolved in the first instance—​in some states not until the 1990s or even later—​by the filing of lawsuits, usually by family members of incompetent patients, requesting a court to declare their authority to make medical decisions for an incompetent patient. Rarely were advance directives the central issue in the case, and thus what the courts had to say about advance directives were nonbinding “dicta.” Further, there were actually very few judicial decisions commenting on the validity of advance directives. What ultimately was responsible for legitimating advance directives was the enactment of legislation by state legislatures. What prompted the legislatures to act, however, is less clear, though a reasonable guess is that it was a combination of the headline-​ grabbing nature of end-​of-​life cases and the pleas of many family members, who never thought of filing a lawsuit, to their legislators. The adoption of uniform legislation on advance directives by the National Council of Commissioners on Uniform State Laws, first in 1985 (Uniform Rights of the Terminally Ill Act 1987), which was revised

Legal Issues in Death and Dying    17 in 1989 (Uniform Rights of the Terminally Ill Act 1999) and then completely reformulated in 1993 (Uniform Health-​Care Decisions Act 1999), may have also played an important role. Another important factor—​perhaps the most important—​was the passage of legislation by Congress in 1990 known as the Patient Self-​Determination Act (PSDA; Pub. L. No 101-​508 1990). The PSDA requires any health-​care institution accepting Medicare funds (which is almost all hospitals and a large proportion of long-​term care facilities) to inform patients of their rights under state law to make an advance directive. The legislation did not create a legal basis for advance directives. It merely required that, where the law of the state in which the health-​care institution was situated recognizes the validity of advance directives, patients be told that about this state law, so in essence the statute is merely a notification provision. Nonetheless, because it was a federal law, and because it was enacted in the wake of and partially as a result of the Supreme Court’s well-​ publicized Cruzan decision, it received a tremendous amount of publicity, so that not only did health-​care personnel become aware of it but so did many patients and families. Furthermore, each time patients were admitted to a health-​care institution, the process of informing them about advance directives resulted in a gradual process of educating the public about advance directives. As a result of the adoption of state advance directive legislation and the implementation of the PSDA, physicians and other health-​care professionals, as well as health-​care managers in hospitals, became aware of the validity of living wills and health care powers of attorney in general. This did not solve all of the issues surrounding their use, but it at least confirmed that it was not illegal, per se, to base end-​of-​life decisions on a living will or on the decisions of a health-​care proxy appointed by the patient using a health-​ care power of attorney.

Futility Until the 1990s, almost all litigation involving end-​of-​life issues followed a similar pattern in which a patient or surrogate requested a physician to withhold or withdraw life-​ sustaining medical treatment and the physician declined to do so on clinical, ethical, or legal grounds (or some mix of the three). The patient or surrogate then filed a lawsuit to have a court resolve the dispute. The lawsuit almost always took the form of a request for declaratory and/​or injunctive relief, that is, a court order declaring what the law is governing the matter in question and ordering the physician to implement the patient’s or surrogate’s request. Rarely was a lawsuit filed requesting monetary damages from the physician (or other health-​care provider) for treatment rendered without consent. After a decade or two of such cases, clinical practice markedly changed such that physicians came to understand the circumstances in which forgoing life-​sustaining treatment was legally required, and they much more readily acceded to such requests.

18   Alan Meisel In fact, so well-​accepted did the forgoing of life-​sustaining treatment become that eventually physicians began to initiate such efforts rather than waiting for surrogates to do so, and sometimes they did so well ahead of a family’s readiness to acknowledge that further treatment would not save the patient’s life or that further treatment was imposing far greater burdens than benefits on the patient. Cases of this sort have come to be known as “futility” cases because it was commonly said by physicians, in broaching this subject with families, that they were recommending forgoing life-​sustaining treatment because further treatment would be futile. The first of these futility cases is still probably the best known—​the Wanglie case in Minnesota (In re Wanglie 1991)—​even though it was never reviewed by an appeals court, and hence it does not constitute, strictly speaking, a legal precedent. Indeed, this is the one area of end-​of-​life law that has not yet reached a significant degree of consensus—​ largely because there has been so little law-​making, either judicial or legislative—​in this area. And because of the legal inactivity, the void has largely been filled by pronouncements from individual health-​care institutions and professional societies. About the best legal advice that can be given is that although patients and surrogates do not have a legal right to demand the provision of any and all treatment they might wish—​even those that are plausibly medically appropriate for the patient’s condition—​it is still highly imprudent for a physician to forgo—​especially to withdraw—​life-​sustaining treatment against the explicit wishes of the patient or family. This is most clearly the case when the patient is in possession of decision-​making capacity, and only a little less so when the patient lacks decision-​making capacity and a surrogate is making decisions on the patient’s behalf. Rather, in situations in which the attending physician believes that a particular treatment is not warranted and the patient—​or more likely the surrogate and other family members—​want it to be administered, the physician has several possible courses of action. One is for the physician to provide treatment and wait—​wait until the family comes to an understanding on its own timetable that further treatment is unlikely to benefit the patient, which might not occur before the patient dies. Another course of action is for the physician to attempt to transfer the patient’s care to another physician who does not object to providing the requested treatment (though the patient’s insurance company and/​or the hospital may have objections to continuing to pay for such treatment). The third course of action is to withhold or withdraw treatment against the surrogate’s wishes and risk whatever legal consequences might be forthcoming.

Actively Hastening Death and Palliative Care The development of end-​of-​life law has primarily been driven by physicians’ fears of legal liability for causing the death of a patient. Absent this fear, it is not clear that the

Legal Issues in Death and Dying    19 Quinlan case would have ever reached the courts (although the physicians in that case claimed that they were motivated by considerations of medical ethics as well). Indeed, the same can be said for virtually every one of the 150 or so end-​of-​life cases litigated to date. In reality, very little of the litigation has expressly sought to impose any kind of liability. The overwhelming proportion of litigated end-​of-​life cases have sought a declaratory judgment, an injunction, or both—​that is, a court order that some proposed course of forgoing life-​sustaining treatment was lawful. Although a few litigated cases have resulted from intrafamilial conflicts about whether life-​sustaining medical treatment should be forgone—​the most notable being the Schiavo case—​this constitutes a very small proportion of end-​of-​life cases. There has been only one criminal prosecution of physicians for forgoing life-​sustaining treatment (Barber v. Superior Court 1983), and it was resolved in their favor. In addition, there have been surprisingly few civil cases seeking monetary damages from physicians arising out of end-​of-​life decisions, and most have sought damages for continuing treatment rather than for discontinuing it. Beginning with the Quinlan case, the courts have fashioned a variety of overlapping explanations as to why forgoing life-​sustaining medical treatment does not result in legal culpability on the part of a patient’s physicians.

Causation One is a causal explanation. According to the courts, when life-​sustaining treatment is withheld or withdrawn, it is not the forgoing of treatment that causes the patient’s death. Rather it is the patient’s underlying medical condition—​which makes continued life impossible without the treatment in question—​that is the cause of the patient’s death. This obviously is an overly simplistic explanation because rarely, if ever, can causation be explained by a sole cause. A result usually has many causes, some more direct and some more attenuated than others, but they are causes in a factual sense nonetheless. In the instance of forgoing life-​sustaining treatment, it is clear that the patient would not have died under the circumstances that he or she did had treatment not been forgone, and therefore to deny that the physician’s actions did not play a causal role in the patient’s death is nonsense. The central issue is not a factual one but a moral and legal one: should the physician be morally and legally culpable for the role that his conduct played in bringing about the patient’s death? To answer this in the negative because the physician’s conduct is said not to be the cause of death—​if by the cause is meant sole cause—​is correct but misses the point.

Intent Another explanation as to why a physician should not be legally culpable for a patient’s death resulting from forgoing life-​sustaining treatment is that, in such situations, the physician lacks the intent necessary for criminal liability—​the intent to cause death.

20   Alan Meisel Courts have reasoned that when life-​sustaining treatment is forgone, the physician’s intent is to honor the patient’s wishes or to relieve the patient’s suffering or both, but not to cause death. Death is said to be merely the consequence of such conduct. This explanation is as flawed as that based on causation. In law, one is said to intend the natural and probable consequences of one’s acts, and the natural and probable consequence of withholding or withdrawing life-​sustaining treatment is that the patient will die; therefore, one’s intent, as far as the law is concerned, is for the patient to die. Also, in law, for there to be intent, one need not have the resulting consequence of an action be the purpose of one’s actions; it is enough that one knows with substantial certainty that a result will ensue for one to be said to “intend” it. In the end-​of-​life treatment context, this means that even if the physician’s purpose is not to cause the patient’s death, he still intends that result if he knows with substantial certainty that death will result from his actions—​that is, from forgoing life-​sustaining treatment.

Right to Refuse Treatment The last important explanation2 as to why forgoing life-​sustaining treatment should not be the basis for civil or criminal liability is that, historically, in American law (and its English antecedents), individuals have a legal right to be free from unwanted contact with their persons. Medical treatment to which there is no consent, or to which consent has been withdrawn, constitutes just such an interference with bodily integrity. In other words, forgoing life-​sustaining treatment constitutes an individual’s invocation of his or her right to refuse treatment, and as such, no penalty may be imposed on the physician who honors such a right by withholding or withdrawing treatment—​even if it results in the patient’s death—​as long as it is authorized by the patient or someone with the legal authority to act on the patient’s behalf. This is the most direct and most satisfying explanation of why an act—​forgoing life-​sustaining treatment—​that results in death, and that is intended to, still does not constitute the basis for civil or criminal liability.

Painting a Bright Line: Passively Versus Actively Hastening Death The upshot of all of this is that the courts have unanimously been able to declare that withholding or withdrawing of life-​sustaining treatment that results in a patient’s death—​as long as properly authorized by the patient or a surrogate—​cannot be the basis for civil or criminal liability. But what about patients (or surrogates) who are terminally ill but whose continued life is not dependent on any form of medical treatment? Some patients may be suffering greatly—​physically, emotionally, psychologically, spiritually, existentially, or some combination of these. They may wish for this suffering to end, and

Legal Issues in Death and Dying    21 they may wish to have that happen by ending their lives. But since there is no treatment to forgo, the only way to end their lives is through some active intervention, either at their own hands or those of someone else. In their decisions holding that withholding and withdrawing life-​sustaining medical treatment are legitimate, the courts—​although not directly asked to decide the question—​have nonetheless gone out of their way to draw a bright line between the practice they are authorizing (forgoing life-​sustaining treatment or passively hastening death) and the practice of actively hastening death and to emphasize that the latter is clearly illegal, indeed, criminal. More specifically, it is some form of criminal homicide, either murder or manslaughter. For the courts not to have made this distinction is unimaginable. First, and most clearly, both forms of actively hastening death, at least at first glance, clearly fit the classic definitions of the crimes of aiding suicide or homicide. Second, from a sociocultural perspective, it is hard to imagine in the 1970s or 1980s that any court in the United States would have held that the active ending of a life—​even the life of someone who is imminently dying and even by that person’s own actions—​could be considered socially, morally, or legally appropriate behavior. Third, merely legitimating passively hastening death—​that is, withholding or withdrawing life-​sustaining medical treatment—​ was at that time controversial enough. Had any court sought to extend that practice to the legitimization of actively hastening death, it would have been an even greater leap. Given the degree of opposition, at least in the early stages, to passively hastening death, to have attempted to legitimate actively hastening death as well would have been to create serious obstacles to legitimization of the former. Thus the keystone of the consensus about end-​of-​life decision making that evolved from the 1970s to the 1990s was that there is a clear, bright line between passively hastening death and actively hastening death. All the end-​of-​life cases to reach the courts have involved passively hastening death (or—​as it usually referred to in judicial opinions and often in clinical practice—​forgoing life-​sustaining treatment or withholding and withdrawing life-​sustaining treatment). It is what the courts and legislatures and clinicians and bioethicists have sanctioned—​as long as there is valid consent. Actively hastening death, by contrast, has traditionally been beyond the pale to all who subscribe to the consensus about end-​of-​life decisions. Actively hastening death—​whether by the patient’s own hand or through the efforts of a third person (a physician, a nurse, a family member or friend)—​is never to be countenanced. It is, simply, criminal homicide. Even if the patient has consented, the act is a crime because the law of murder and manslaughter vindicate society’s interests in every human life, not just the individual victim’s interest. One obvious consequence of drawing this bright line was to send a clear message to all clinicians caring for dying patients that any efforts to actively hasten a patient’s death were legally unacceptable. To the extent that actively hastening death was already being practiced—​albeit covertly because of its presumed illegality—​this had the effect either of reducing its practice or driving it further underground or both. Another, and less obvious, result may have had to do with the use of medications for pain relief. Physicians caring for the dying may have been made more uncomfortable than ever to prescribe—​and

22   Alan Meisel nurses to administer—​opiates and barbiturates for the relief of suffering for fear that it would hasten a patient’s death and be considered an “active,” rather than “passive,” hastening of death. In the longer run, however, drawing this bright line has had just the opposite results. As time went by, the disparate treatment of the two categories of the dying—​those being kept alive by medical treatment and those not—​became more apparent and more difficult to justify. Why the person dying of a painful cancer but not being kept alive by life support could obtain not hasten death but the patient with kidney failure could forgo dialysis and hasten his or her death was not readily apparent, at least not on compassionate grounds. This led to two diametrically opposed movements: one to legalize physician aid-​in-​dying and the other to improve palliative care. In truth, however, the line between passively and actively hastening death was never as clear or bright as its defenders would have it, and slowly, beginning in the late 1980s, they began to attempt to abolish it, with limited but increasing success. A number of factors have played an important role in this process. Although there have long been proponents of legalizing physician aid-​in-​dying in the United States and elsewhere, this movement had virtually no political traction until the 1990s. However, a series of events in the late 1980s and the early 1990s turned what was largely a pipedream into a serious political and legal movement. First, there has been an increasingly open and intense public debate about the subject. Articles in reputable medical journals—​such as “A Piece of My Mind:  It’s Over Debbie” (1988) in the Journal of the American Medical Association and “Death and Dignity: A Case of Individualized Decision Making” (Quill 1991) in the New England Journal of Medicine—​have exposed the fact that clinicians sometimes actively end the lives of seriously suffering patients. A number of surveys of health care professionals lead to the same conclusion (Asch 1996; Back et  al. 1996; “Poll Shows” 1992), which raises the important question of whether the practice of actively hastening death might be subject to abuse and whether legalization and regulation would help to assure that it is practiced only with the voluntary consent of competent patients. The practices of the crackpot Michigan pathologist, Jack Kevorkian, to create and make available a “suicide machine” to those who wished to use it played a key role in this movement. Despite the ghoulish nature of his undertaking, he did manage to bring the issue of physician aid-​in-​dying out of the closet and into the light of day so that millions of Americans who might otherwise be unfamiliar with the issue suddenly gained new insight into it. Second, public sentiment as expressed in public opinion polls for many decades has revealed that the public overwhelmingly supports the legalization of at least some forms of actively hastening death. At about the same time as Dr. Kevorkian’s activities were occurring and being widely publicized, more sober proponents of legalization began to coalesce and form movements in a number of states to legalize actively hastening death through legislation. This public sentiment has proved difficult to translate into law. As the matter is increasingly openly discussed, the arguments against legalization—​as well as an advertising blitz against legalization—​have more often than not carried the day.

Legal Issues in Death and Dying    23 Nonetheless, three states through legislation have legalized a form of actively hastening death—​often referred to as “physician aid-​in-​dying,” which involves a physician writing a lethal prescription for a competent, terminally ill patient who requests it. The first two were by referendum in Oregon in 1994 (which was reaffirmed in 1997) and Washington in 2008 and a third by the legislature in Vermont in 2013. Third, there have been a number of lawsuits brought to invalidate laws making assisting suicide a crime, at least insofar as those laws apply to competent, terminally ill people obtaining a prescription for a lethal dose of medication from their licensed physicians. The most important ones were filed in federal courts in Washington state and New York state challenging the constitutionality of those states’ statutes making it a crime to provide aid to another person in committing suicide. Although the federal courts of appeals held in both of these cases that the federal constitution prohibited states from criminalizing physician aid-​in-​dying for a terminally ill person, the U.S. Supreme Court reversed both decisions (Washington v. Glucksberg 1997; Vacco v. Quill 1997). It held that the U.S. Constitution does not guarantee such a right. However, the Court also ruled that states are not prohibited by the Constitution from legalizing if they wish to do so—​thus retrospectively validating the 1994 decision of Oregon voters and prospectively validating similar decisions in other states—​and that the practice of terminal sedation of terminally ill patients, with valid consent, did not offend the Constitution. Finally, the Court also signaled that if, in practice, significant barriers to palliative care for dying patients were found to exist, the Court might be willing to revisit its core holding that state statutes criminalizing aiding suicide do not offend the Constitutional rights of dying patients. Lawsuits challenging the constitutionality of similar laws in Alaska, California, and Florida have also been unsuccessful. However, in two other states (New Mexico [State v. Morris 2014] and Montana [Baxter v. State 2009]), judicial decisions have legalized it. Finally, changes in clinical practice have played an important role in breaking down the distinction between actively and passively hastening death. One important consequence of the public debate about the legalization of actively hastening death has been to shine light on the fact that many people near the end of life who experience serious pain, suffering, or both do not receive adequate treatment for it from their physicians. The increased awareness of this both by the public and those in the health professions has helped to bring about a revolution in the care of the dying, most notably by increasing the availability of palliative care and hospice services. Many of the pharmaceutical agents used in palliative medicine—​opiates, nonopiate analgesics, anxiolytics, and sedatives—​are capable of depressing respiration, and in large enough doses, they can cause a patient’s death. Determining just how much and what combination of medications to administer to a particular patient is as much an art as a science. Give too little, and the patient’s suffering will not be ameliorated; give too much and it may kill the patient. Moral philosophers have long reasoned that a patient’s death resulting from an unintended overdose of palliative medication should not result in moral culpability for the patient’s death. This conclusion is based on the principle of double

24   Alan Meisel effect—​borrowed from Catholic moral theology—​which holds that if one’s purpose in taking some action is a licit one, the fact that an unintended undesirable consequence also results is not the basis for culpability. Applied to palliative care, this means that the physician who prescribes or administers medication intended to relieve a patient’s suffering is not culpable for harm to the patient from that medication—​including actively hastening the patient’s death—​if that consequence was unintended, even if it was foreseeable. Although well accepted by many philosophers and theologians and understood by physicians, prior to the Supreme Court’s decisions in the physician aid-​in-​dying cases there was no sound legal basis for the doctrine of double effect, so that while it might excuse a physician from moral culpability, it was unclear what effect, if any, it had on legal culpability, which was probably the greater concern for most physicians. The cases before the Supreme Court did not expressly raise the question of whether there was a sound legal basis for the doctrine of double effect. Nonetheless, one of the reasons the Court gave for its unwillingness to find that state laws that made aiding suicide a crime did not violate the Constitution—​even when the person seeking to end his or her life was terminally ill and the person providing the aid was a licensed physician—​was that the doctrine of double effect protected physicians from liability for providing aggressive palliative care to such patients, and thus their needs for pain relief could be met without the legalization of physician aid-​in-​dying. Indeed, patients have a legal right to adequate medication for pain and suffering even if it does hasten death, and thus denying adequate medication, even if it is life-​threatening, constitutes a legal wrong. A small number of states have enacted statutes to provide some legal immunity to physicians from liability arising from death caused by palliative care in addition to that accorded by the Supreme Court’s opinion. To date, all the successful efforts in the United States to legalize actively hastening death have implicitly made a distinction between the two general ways in which death may be actively hastened in terminally ill patients. One way is by another person doing something to the terminally ill patient that ends the latter’s life, the classic example being a physician administering a lethal injection (often referred to as active euthanasia). The other way is by another person providing the terminally ill patient with something by which the patient can end his or her own life (referred to as physician-​ aid-​in-​dying or physician-​assisted suicide); here the classic example is a physician giving the patient a prescription for a lethal dose of medication that the patient self-​ administers. The distinction between the two, if there is indeed a moral one (Brock 1992), is that in the latter situation the patient is the final actor and thus retains, until the last moment, the right to change his or her mind. Only physician-​aid-​in-​dying has achieved any degree of legal recognition in the United States, active euthanasia still being considered a crime in all American jurisdictions. However, in Europe, quite the opposite has been true. Both the Netherlands and Belgium have given formal legal approval to active euthanasia with the patient’s consent, and Switzerland has given tacit legal approval to the practice.

Legal Issues in Death and Dying    25

Conclusion From the decisions of various federal (and mostly state) courts, beginning with the Quinlan case in 1975, a consensus emerged about the legality of withholding and withdrawing life-​sustaining medical treatment. By the time of the U.S. Supreme Court’s decision in the Cruzan case in 1990, this consensus had become fairly well established not just in law but in clinical practice. The Cruzan decision itself helped to cement various aspects of this consensus, and various law-​making efforts of courts and legislatures since then have only reinforced it. Although law has had a tremendous impact on clinical practice with regard to end-​of-​life decision making, the reverse is just as true. Clinical practice—​along with public opinion and the work of a presidential commission—​have greatly influenced the path that courts and legislatures have been willing to take.

Notes 1. The President’s Commission discussed a fourth category—​intended versus unintended but foreseeable consequences—​but the Conroy case did not discuss this one. This is also known as the principle of double effect, which is discussed later in this essay. 2. There are also two other reasons sometimes given by courts as to why forgoing life-​ sustaining treatment is not the basis for legal liability—​it is passive rather than active euthanasia, and there is a basis in advance directive statutes for it—​that are both unconvincing and have played minor role in judicial decisions.

References A piece of my mind: It’s over, Debbie. 1988. JAMA 259(2): 272. Asch, D. A. 1996. The role of critical care nurses in euthanasia and assisted suicide. New England Journal of Medicine 334(21):1374–​1402. Back, A. L., Wallace, J., Starks, H. E., and Pearlman, R. A. 1996. Physician-​assisted suicide and euthanasia in Washington State:  Patient requests and physician responses. JAMA 275(12):919–​925.

Barber v. Superior Court. 1983. 147 Cal. App. 3d 1006, 195 Cal. Rptr. 484. Baxter v. State. 2009. 224 P.3d 1211. Mont.

Brock, Dan W. 1992. Voluntary active euthanasia. The Hastings Center Report 22 (2):10–​22.

Cruzan v. Director. 1990. 497 U.S. 261. In re Conroy. 1985. 98 N.J. 321, 486 A.2d 1209. In re Quinlan. 1976. 70 N.J. 10, 355 A.2d 647. In re Schiavo. 2001. 789 So. 2d 248. Fla. In re Wanglie. 1991. No. P-​91-​283. Minn. 4th Dist. Ct. Hennepin County, July 1. Kutner, L. 1969. The living will: A proposal. Indiana Law Journal 44(1):539–​554.

26   Alan Meisel Margolick, David. 1982. Hospital is investigated on life-​support policy. New  York Times, June 20. http://​www.nytimes.com/​1982/​06/​20/​nyregion/​hospital-​is-​investigated-​on-​life-​ support-​policy.html. Meisel, Alan, and Cerminara, Kathy. 2012. The Right to Die:  The Law of End-​ of-​ Life Decisionmaking. New York: Wolters & Kluwer. Poll shows that 1 in 5 internists has helped a patient die. 1992. American Medical News. March 16: 9. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1983. Deciding to forego life-​ sustaining treatment. Washington, DC: Government Printing Office.

Pub. L. No 101-​508. 1990. §§ 4206, 4751.

Quill, T. E. 1991. Death and dignity: A case of individualized decision making. New England Journal of Medicine 324(10):691–​694.

State v. Morris. 2014. No. D-​202-​CV 2012-​02909. N.M. Dist. Ct. 2d Dist. Jan. 13.

Sullivan, Ronald. 1982. Queens hospital accused of denial of care. New York Times, November 17.

Superintendent of Belchertown State School and Another v. Joseph Saikewicz. 1977. 373 Mass. 728. 370 N.E.2d 417.

Uniform Health-​Care Decisions Act. 1999. 9 Uniform Laws Annotated Pt. 1B 143. St. Paul: West Publishing Company. Uniform Rights of the Terminally Ill Act. 1987. 9B Uniform Laws Annotated 609. St. Paul: West Publishing Company. Uniform Rights of the Terminally Ill Act. 1999. 9B Uniform Laws Annotated 127. St. Paul: West Publishing Company.

Vacco v. Quill. 1997. 521 U.S. 793. Washington v. Glucksberg. 1997. 521 U.S. 702.

Chapter 2

“ So What D o You Wa nt Us To D o ?”  Patients’ Rights, Unintended Consequences, and the Surrogate’s Role Mark P. Aulisio

Many stories have been written about my daughter. Others can write about the accident, about her life and about her death. But no one else can write about the profound pain and anguish we shared as a family. (Quinlan, 2005: xvi)

Introduction With the aging of the baby boomers—​what some have termed the “silver tsunami”—​ reliance on surrogate decision-​makers is sure to grow in the years ahead (Dunn and Alici, 2013). A recent study concluded that nearly half of all hospitalized adults in the United States aged sixty-​five or older need a surrogate decision-​maker within twenty-​ four hours of hospitalization (Torke et al., 2014). Almost 60% of these require decisions about life-​sustaining care. Thus approximately one in three hospitalized adults aged sixty-​five or older in the United States will have a family member or significant other placed in the position of having to make potentially life-​and-​death health-​care decisions for them within just two days of admission to the hospital. For those of us in and around health care, this prospect should be more than just a little unsettling. It certainly is for me. Indeed, in my experience with ethics consultation over the past twenty years, the most common triggers for a consult are problems related to surrogate decision-​making. Even a cursory look at media reports (e.g., Terry Schiavo in Florida, 2005; Barbara Wise in Ohio, 2013; Marlise Munoz in Texas, 2014)  and the increasingly voluminous academic literature suggests that my experience is far from idiosyncratic (Silveira, Kim, and Langa, 2010; Sulmasy and Snyder, 2010; Barrett and Scales, 2012; Hickman, Daly, and

28   Mark P. Aulisio Lee, 2012; Iverson et al., 2012; Majesko et al., 2012; Schenker et al., 2012; Sullivan et al., 2012; White et al., 2012; Iverson et al., 2013; Song and Ward, 2013). Why is surrogate decision-​making so fraught with problems, and what can be done to address them? In this article I attempt to answer these questions, offering at least a partial explanation for the first and some suggestions for the second. I consider the historical context in which surrogate decision-​making emerged by examining two key cases in the rise of the patients’ rights movement that involved surrogate decision-​makers, the landmark cases of Karen Quinlan and Nancy Cruzan. I argue that the success of the patients’ rights movement, spurred by Quinlan and Cruzan and championed in the rise of bioethics, has had some very negative unintended consequences; chief among these are problems associated with surrogate decision-​making. I then consider why surrogate decision-​making is so problematic and offer some suggestions as to how the surrogate’s role might be reconceived, lest we be washed away by the coming silver tsunami and the wave of distraught surrogate decision-​makers it will inevitably bring.

A Tectonic Shift: The Rise of Patients’ Rights and the Erosion of Physician Power On March 31, 1976, Joseph and Julia Quinlan won the right to have what was thought to be life-​sustaining respirator support withdrawn from their twenty-​two-​year-​old daughter, Karen (In re Quinlan, 1976). The New Jersey Supreme Court ultimately arrived at its decision by interpreting the US Constitution’s implied right to privacy—​the right of a person to decide personal issues without governmental interference—​to be sufficiently broad to allow the family of a dying patient to make even a life-​or-​death decision on the patient’s behalf (In re Quinlan, 1976; Annas, 1990; Pence, 2004). Although Quinlan, to the surprise of many, lived another ten years after the withdrawal of the respirator (she was weaned from the respirator and continued to receive nutrition and hydration through a feeding tube), the focus of controversy at the time was on whether withdrawal of the respirator was a form of “killing” rather than “letting die” (Pence, 2004). Indeed, a representative of hospital administration is reputed to have said to Julia Quinlan, “in this hospital we don’t kill people” in explaining the hospital’s opposition to removing respirator support (Quinlan and Quinlan, 1977). While this alleged comment sounds outrageous to us today, it is easy to forget almost forty years later that the official position of the American Medical Association and the practice standards of the day, such as they existed, supported the hospital’s position. At the time, “withholding” life-​sustaining treatment was considered to be sometimes justifiable, but withdrawing that same treatment, once initiated, was viewed as a form of “euthanasia” or “mercy killing,” the equivalent of “murder” (Pence, 2004). In light of this, Karen Quinlan’s physicians not only had ethical objections to withdrawal of

The Surrogate’s Role   29 respirator support but also feared a possible malpractice suit or even criminal prosecution for “killing” her (Pence, 2004). It is also easy to forget that the de facto authority to institute and continue life-​sustaining treatment at the time rested primarily with physicians (Aulisio, 2014). The Quinlans poignantly described their ordeal as a “nightmare” as they fought to be allowed to have what they considered to be “extraordinary” treatment withdrawn—​a treatment they said their daughter would not have wanted (Quinlan and Quinlan, 1977; Quinlan, 2005). Joseph and Julia Quinlan’s fight was, in a real sense, a fight for the right to become surrogate decision-​makers for their daughter. In support of this fight, they turned to their parish priest and local bishop and also to the bioethicist Robert Veatch, then staff director of the Death and Dying Group at The Hastings Center, with whom they met three times to discuss the tragic plight of their daughter (Jonsen, 2003). The Quinlans’ nightmare and the interpretation of the right to privacy that granted them the legal authority to make this “private” decision was, sadly and ironically, played out in the most public of ways. As Veatch put it just a few years later: It took the life-​support debate to the front pages and to the lead story of the evening news. It made clear that the physician who happened to inherit a patient in an emergency room did not automatically have custody of that patient and could not automatically insist that life-​support be continued against the wishes of the patient or surrogate. It brought surrogate decision making to the stage front and center. It identified privacy as a decisive legal and moral category. It introduced in a muddled way the concept of an ethics committee for the first time in the public discussion of life-​support. (Jonsen, 2003: 255)

As Veatch’s comments suggest, it would be hard to overstate the significance of the Quinlan case. The rise of surrogate decision-​making against physician power was yet another manifestation of an underlying a tectonic shift in society’s view of the physician–​patient relationship. A patient’s right to decide what should be done with his or her own body, the right to self-​determination (to paraphrase the language of Justice Cardoza in Schloendorff [1914]—​the often-​cited seminal case for Informed Consent Doctrine in Anglo/​American law [Schloendorff v.  Society of New  York Hospital]), the core of the increasingly robust notion of informed consent operative in Anglo/​American law and Western medicine was being extended to the refusal of even life-​sustaining treatment. Furthermore, patients would not cede back to physicians their right to “self-​determination” when unconscious and unable to make decisions even where life-​sustaining treatment was involved. The fact that Veatch headed up The Hasting Center’s Death and Dying Group and that he was consulted by the Quinlans shows the extent to which controversy over use of life support and who controls that use had been simmering in the early to mid-​1970s. The Quinlan court’s fear that a flood of life-​support cases was in the offing and its suggestion, albeit “muddled,” that ethics committees might be a judicial alternative only serves to underscore this.

30   Mark P. Aulisio Immediately following the Quinlan decision, states rushed to pass legislation affirming the rights of competent patients to refuse life support, as well as the rights of surrogates to make decisions on their behalf (Annas, 1990). In 1978 the US President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research was established with an expanded scope, going beyond its predecessor to include ethical problems in medicine (not just regulatory issues concerning human subjects research protections). From 1980 to 1983 that group worked on its now deservedly famous report, “Deciding to Forego Life-​Sustaining Treatment,” which, among other things, strongly affirmed (a) the rights of competent patients to refuse unwanted life-​sustaining treatments, (b) the rights of surrogates to make decisions on behalf of incompetent patients, and (c) the potential value of ethics committees as an alternative to the courts in assisting with these cases (US President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, 1983). Regarding the latter, ethics committees, which were found to be present in only around 1% of US hospitals at the time of the report (Youngner et al., 1983), rapidly expanded to over 60% just four years after its release (Fleetwood, Arnold, and Baron, 1989). Even while Joseph and Julia Quinlan’s ordeal endured, Lester and Joyce Cruzan found themselves trapped in a similar nightmare following a 1983 car accident that left their daughter Nancy in a persistent vegetative state at the age of twenty-​six. The Cruzans sought to have Nancy’s feeding tube removed, arguing that she would not have wanted to be kept alive in this state (Annas, 1990; Annas et al., 1990; Pence 2004). In the public discussion, the rhetoric of “killing” and “murder” characteristic of the Quinlan case gave way to that of “starving to death” on one side and “imprisoning and torturing” by medical technology on the other. Despite this rhetoric, Nancy Cruzan’s legal case focused primarily on the state’s right to set evidentiary standards regarding patient wishes and, only secondarily, on whether “artificial” nutrition and hydration should be viewed as a medical intervention (Annas, 1990; Annas et al., 1990; Pence, 2004). As the Cruzans sought to have their daughter’s feeding tube removed—​in accord, they argued, with her wishes—​ the State of Missouri argued that the Cruzans had failed to produce “clear and convincing evidence” of Nancy’s wishes, thus failing to meet the state’s evidentiary standard for withdrawal of the life-​sustaining treatment on an incapacitated patient (Annas, 1990; Annas et al., 1990; Pence, 2004). Despite the rhetorical flourishes on either side of this controverted case, the wishes of the patient were the focal point of debate, and it is clear that the tectonic shift signaled by Quinlan was still in motion—​patients, not physicians, should control decision-​making. Just as it is easy to forget that Quinlan occurred against a backdrop of American Medical Association prohibition of withdrawing life-​sustaining treatment and powerful physician hegemony in treatment decisions, it is easy to forget that the State of Missouri actually won the Cruzan case before the US Supreme Court. The court ruled that states did indeed have the right to set evidentiary standards with respect to the wishes of a formerly competent but now incompetent patient (Annas, 1990). Thus Missouri’s “clear and convincing evidence” standard, a standard that sat between “preponderance of

The Surrogate’s Role   31 evidence” and “beyond a reasonable doubt” in levels of stringency, was not taken to violate Nancy Cruzan’s (constitutional) right to be free of an unwanted medical intervention (Annas, 1990). Although reactions immediately after the decision were mixed, Cruzan remains remarkable for a number of reasons. First, it recognized that competent patients have a constitutional right (“liberty interest”) to refuse even life-​sustaining treatment. Second, it made clear that tube feeding (“artificial” nutrition and hydration) is not different from any other type of medical intervention for the purposes of this right. Third, it affirmed that patients do not lose their right, or “liberty interests,” to be free of unwanted medical interventions simply because they become incompetent (even if, as Justice Sandra Day O’Connor famously put it, the “challenging task of crafting appropriate procedures for safeguarding incompetents’ liberty interests is entrusted to the ‘laboratory’ of the States” (Cruzan v. Director, 1990). Most important for our purposes here, the Cruzan decision gave a huge push to surrogate decision-​making and to what eventually came to be called the advance directive movement (Aulisio, 2014). As Annas put it at the time, Since she [Justice O’Connor] agreed that few people will provide explicit instructions, she suggested that everyone appoint a proxy decision maker, and she noted that the Cruzan decision “does not preclude a future determination that the Constitution requires the States to implement the decisions of a duly appointed surrogate.” This will undoubtedly give a well-​deserved boost to durable powers of attorney, and most physicians will properly prefer a real person with whom they can discuss the patient’s options to a written living will that may require interpretation. (1990: 672)

Little did Annas know that the print would be barely dry on his article before Congress would pass the Patient Self Determination Act (November 5, 1990: Federal Patient Self Determination Act, 1990 [PSDA]) and the State of Missouri would (somewhat mysteriously) find new evidence of Nancy Cruzan’s wishes—​evidence that was, it said, sufficient to satisfy the “clear and convincing evidence” standard and allow for withdrawal of her feeding tube (December 14, 1990). Although public outcry arguably drove each turn of events, that it led directly to the passage of the federal PSDA is beyond dispute. The PSDA mandated that health-​care organizations receiving Medicare or Medicaid funds 1.  Give [patients] at the time of admission a written summary of: • [their] health care decision-​making rights. (Each state has developed such a summary for hospitals, nursing homes, and home health agencies to use.) • the facility’s policies with respect to recognizing advance directives. 2.  Ask [patients] if [they] have an advance directive, and document that fact [in the patient’s medical record] if they do. (It is up to [patients] to make sure they get a copy of it.) 3.  Educate their staff and community about advance directives.

32   Mark P. Aulisio 4.  Never discriminate against patients based on whether or not they have an advance directive. Thus, it is against the law for them to require either that [a patient] have or not have an advance directive. (http://​www.americanbar.org/​groups/​public_​ education/​resources/​law_​issues_​for_​consumers/​patient_​self_​determination_​act. html) As noted, the PSDA is commonly credited with having given rise to the advance directive movement (Aulisio, 2014). Although the relative success or failure of this movement has been a matter of some debate in the literature, it is clear that in the twenty-​five years since the passage of the PSDA we have seen the continued transformation of the physician–​ patient relationship in Western medicine, as well as the erosion of physician power and the ascendance of patients’ rights. Sandra Day O’Connor’s suggestion in the Cruzan decision of 1990 that everyone appoint a proxy for health-​care decision-​making both presaged and contributed to this continued transformation. It is astounding that in the span of forty to fifty years we have gone from near-​complete physician control of decision-​making at the end of life, including benevolent deception and outright lying to dying patients about their diagnoses (see Youngner, this volume), to routinely disclosing terminal diagnoses to patients and allowing them to make end-​of-​life decisions for themselves. It is beyond astounding to think that when patients are incapacitated we now look primarily not to physicians or other members of the health-​care team but to surrogates (typically family members) for even life-​and-​death decisions. Patients, at their most vulnerable moments, are no longer subject to the whim of physician power or at risk of being indefinitely imprisoned by medical technology but instead have family to ensure that their rights are respected. In this regard, the shift in power has been truly tectonic and the progress great. The success of the patients’ rights movement, however, spurred by Quinlan and Cruzan and unabashedly championed by bioethics, has experienced some unforeseen problems; chief among these is the highly problematic nature of surrogate decision-​making itself.

Unintended Consequence: The Burden of Surrogate Decision-​M aking When I first started doing ethics consultations almost twenty years ago, I encountered a case that has stayed with me, leaving an indelible mark on my work as a clinical ethicist. I have used the case elsewhere to raise questions about end-​of-​life decision-​making (Aulisio, 1999) and the role of ethics consultation (Aulisio, 2003). Here I use it toward a different end, that is, to highlight some of the reasons why surrogate decision-​making is so fraught with problems. I refer to the case as that of the “Beloved Brother”: A 72-​year-​old white male was admitted to the SICU for a ruptured aortic aneurysm. Although he initially did well and was discharged to a long term care facility, he was

The Surrogate’s Role   33 soon readmitted for a massive GI bleed. He underwent surgery but unfortunately re-​bled, resulting in cerebral ischemia with mental status changes. He became ventilator dependent, developed renal and liver failure, and sepsis. The patient was placed on dialysis. The patient had a sister and a brother, to whom he was very close. The patient’s sister was the recognized decision maker in the family. She insisted on aggressive therapy and wanted “everything done.” Aggressive therapy was instituted and continued for about three weeks without any improvement in status. After three weeks, the nursing staff felt that aggressive treatment was inappropriate and that the patient did not want it. They reported that the patient periodically grimaced when turned and suctioned. One nurse reported that prior to intubation the patient said “Why are you doing this to me? I’ve had enough. I’m dying, aren’t I?” Also, the nursing staff relayed that the patient’s brother said that the patient was angry that he was readmitted (and intubated). The nursing staff did not initially oppose aggressive therapy because they thought the sister “needed time” and would “come around.” The SICU attending now agreed with the nursing staff that continued aggressive therapy was inappropriate. They wanted to stop dialysis. The patient’s sister remained adamant that “everything be done.” The head nurse called for an ethics consult because her staff “cannot continue to do this to that poor man.” They were angry and upset with the patient’s sister. (Aulisio 2003: 3)

A colleague and I  responded to the request for consultation by first meeting with members of the nursing staff. They did not, as often happens, impugn the sister’s motives for insisting on continued aggressive treatment, nor did they question her devotion to her sick brother, the patient. Indeed, they relayed that the sister visited daily, always bringing something special to brighten the patient’s room and keeping vigil, hoping against hope that he might recover. They were, however, extremely upset with her apparent “inability” to accept the fact that her brother was dying and to just “let go.” Several nurses wept while recounting the patient’s winces, grimaces, and groans as they continued to “torture” (administer aggressive care to) him. Their anger and frustration was palpable. After meeting with other members of the care team, including the attending physician, and hearing a similar story, my colleague and I met with the patient’s sister and brother. We introduced ourselves but, to our surprise, the patient’s sister started the meeting. “I know why you’re here,” she said, as she started to cry. “They want you to convince me to give up on my brother.” Sobbing, she continued, “But you don’t understand … how could I live with myself if I gave up on him? I love him so much!” The sister’s words convey the unbearable burden that can come with being thrust into the role of surrogate decision-​maker, particularly when life-​or-​death decisions must be made. The anguish and distress of Quinlan and Cruzan is, in many cases, compressed not into years but months, weeks, or days, with the responsibility for the decision seemingly resting squarely on the shoulders of the surrogate. The now-​expansive literature on surrogate decision-​making underscores the traumatic nature of this burden (Wendler and Rid, 2011). For example, one study that

34   Mark P. Aulisio involved interviews with family members of 294 intensive care unit patients found that 33.1% of all family members exhibited symptoms consistent with moderate to major risk of developing posttraumatic stress disorder (PTSD; Azoulay et al., 2005). The same study showed PTSD symptoms in 47.8% of family members who “shared in decision making” and an especially troubling 81.8% of family members who “shared in end of life decisions” (Azoulay et al., 2005). Another study found that surrogates experience “significant emotional conflict between the desire to act in accordance with their loved one’s values and 1) not wanting to feel responsible for a loved one’s death, 2) a desire to pursue any chance of recovery, and 3) the need to preserve family well-​being” ( Schenker et al., 2012: 1657). Sadly, the problems discussed in the recent surrogate decision-​making literature look very much like those my colleague and I encountered in the “Beloved Brother” case, a case that occurred nearly twenty years ago. The patient’s sister couldn’t “live with herself ” if she “gave up” on her brother. She felt the weight of a life-​or-​death decision crushing down upon her and she was determined to give her sick brother “every chance” to recover. There were even hints of intrafamilial tension as the nursing staff indicated that the patient’s brother “said that the patient was angry that he was readmitted (and intubated).” Last, it is clear that the sister was under tremendous emotional strain, crying and even sobbing as she started the family meeting—​no doubt experiencing symptoms of PTSD. How might we lighten the burden of surrogate decision-​making for those who, like this sister, find themselves staggering beneath its weight?

Lightening the Load: Reconceiving the Surrogate’s Role Pitfalls of the Standard View By now nearly everyone in health care should be able to articulate, at least roughly, the standard view of the role of a surrogate decision-​maker. On the standard view, the surrogate decision-​maker’s role is to exercise what law somewhat unfortunately terms “substituted judgment” (Furrow, 2008). In other words, surrogate decision-​ makers are supposed to make the decisions that they think their loved ones would have made if able to make those decisions for themselves. The terminology “substituted judgment” is somewhat unfortunate because it lends itself to the interpretation that the surrogate substitutes his or her judgment for that of the patient and then makes decisions in light of that judgment. This interpretation plainly supplants the patient, who is unable to make decisions, in favor of the surrogate, who is able to make decisions—​sort of like a pinch hitter in baseball. “Substituted judgment” is, after all, a short step from “substituted decision-​maker,” and it is easy lose sight of

The Surrogate’s Role   35 the “make decisions that their loved ones would have made if able to make decisions for themselves” core of the concept. Even if health professionals themselves do not lose sight of this core idea of the surrogate’s role (on the standard view), they often unwittingly undermine it. Anecdotally, I can aver that if I had a dollar for every time I have heard an attending say to a family member, “So, what do you want us to do?” or, more directly “Well, you’re the decision-​ maker; what do you want us to do?”, I would have a small fortune. Distraught and grieving family members often buckle under the weight of this question. The situation is compounded when physicians merely inform surrogates of the variety of options before them or facilitate discussion of those options without providing a recommendation. Surrogates are then left to choose among the menu of options, each undoubtedly explained in great detail but with no directive guidance. This is an all-​too-​common occurrence. For example, in one multicentered study of audiotaped clinician–​family conferences concerning life-​support decisions in intensive care units at six hospitals, physicians refrained from making a recommendation in more than 47% of the cases (White et al., 2010). To make matters worse, the same study indicated that physicians refused to make a recommendation half the time even when explicitly asked by surrogates to do so! Such refusals to make a recommendation are, no doubt, well-​intentioned attempts to honor patient autonomy and the role of surrogate decision-​ makers. Unfortunately, this can amount to what I term the “information dump” view of autonomy, overwhelming surrogates with detailed information about a range of options but depriving them of at least one critical piece of information they need (and invariably want)—​that is, the recommendation of the care team. This approach would be problematic even when dealing with patients directly, but it is especially so where surrogates are concerned because it sends the clear message that “this is all on you—​you are the decision maker(s)!” Even when the core concept of the standard view of surrogate decision-​making (i.e., “make decisions that the patient would have made if able to make such decisions for him/​herself ”) is neither lost nor undermined, surrogates still struggle. Why? In a nutshell, surrogates often do not know their loved one’s wishes and have very little confidence about what decisions their loved ones would have made if faced with the choice at hand. This lack of confidence appears to be warranted, at least if the literature is to be believed. One comprehensive literature review with meta-​analysis revealed that patient-​designated and next-​of-​kin surrogates incorrectly predict patient preferences regarding end-​of-​life treatment about a third of the time (Shalowitz, Garrett-​Mayer, and Wendler, 2006). Worst still, patient designation of a surrogate and prior discussion does not seem to make much of a difference with respect to surrogate accuracy (Shalowitz et al., 2006). All of these problems taken together paint a rather bleak picture of the current state of affairs with respect to surrogate decision-​making. It is, as we know it, what clinicians would call a “train wreck”—​a patient on life support with multisystem organ failure and little hope for survival. If we are to move forward, the surrogate’s role must be reconceived, but how?

36   Mark P. Aulisio

Reconceiving the Surrogate’s Role In order to properly reconceive the surrogate’s role we need to look again at its moral and political basis. In the previous discussion of the Quinlan case, I suggested that the rise of surrogate decision-​making was a manifestation of an underlying tectonic shift in society’s view of the physician–​patient relationship. A patient’s right to decide what should be done to/​with his or her own body—​the right to self-​determination, which is the substantive core of the notion of informed consent—​was understood as extending even to the refusal of life-​sustaining treatment. Furthermore, this right was not understood as ceded back to physicians in the event that a patient lacked capacity to make a decision—​hence the need for surrogates. In the language of Cruzan, patients have constitutionally guaranteed “liberty interests” that must be safeguarded even when they are incompetent—​turning to surrogates, especially those who know and love the patient, is one way of safeguarding those “liberty interests.” Thus the moral and political basis for having surrogates stems from the substantive normative core of the notion of “informed consent” (i.e., a patient’s right to “self-​determination”) or the need to protect his or her constitutionally guaranteed “liberty interests” or, in the language I prefer, a patient’s right to live by his or her values. If this is the moral and political basis for having surrogates, then the moral and political authority for decision-​making comes from the patient. The patient is the source of decision-​making authority and, by extension, remains the “decision-​ maker” because it is respecting the patient’s right to live by his or her values that is the basis for having a surrogate in the first place. Normatively, we do not have an alternate (“surrogate”) locus of decision-​making authority. The normative locus is exactly the same as would be if the patient were alert and oriented times three; in other words, decisions to be made remain the patient’s decisions and not those of the family or members of the care team.

Implications for the Surrogate’s Role What then are the implications of this for our understanding of surrogate decision-​ making? One is that surrogates are not in fact decision-​makers. In practice, this means that we should ideally drop the “decision-​maker” language altogether when discussing the surrogate’s role because it is misleading. Conceptually, the substantive core of the surrogate’s role is obfuscated when we tag “decision-​maker” onto surrogate. This tag confounds surrogates and health professionals alike. As discussed previously, the latter often approach surrogates for “decisions” as if they (surrogates) are in fact “decision-​ makers” and, quite understandably, the former get the clear message that they (surrogates) are in fact “decision-​makers.” If we are to keep “decision-​making” language at all, it should be kept at arm’s length from “surrogate.” We should refer only to the “surrogate’s role in decision-​making” or something to that effect.

The Surrogate’s Role   37 This leads to a second implication of the previous discussion of the normative moral and political basis for the surrogate’s role. As we have seen, if the literature is correct, many surrogates are not very good at predicting patients’ wishes or making decisions that their loved ones would have made if able to choose for themselves. Surrogates do, however, know their loved ones and can help members of the care team learn their loved ones’ story, replete with their values, beliefs, and idiosyncrasies insofar as these might be relevant for decision-​making (Braun, Naik, and McCullough, 2009). Although their proposed terminology is rather unfortunate because it obscures rather than elucidates, Sulmasy and Snyder capture well what I take to be the core of the properly reconceived surrogate’s role: The key question under best interests is, “What do you think is best for your mother?” Under substituted judgment it is, “What would your mother choose if she could tell us?” The substituted interests model says, “Tell us about your mother.” A good surrogate can articulate the patient’s authentic values by describing the patient’s loves, beliefs, and fundamental moral commitments rather than just specific preferences. (2010: 1946)

This leads to a third implication of the discussion of the moral and political basis for the surrogate’s role. In the reconceived surrogate’s role, all involved are focused on patient values and seeing to it that those values are respected in decision-​making within the bounds of medically acceptable care. Why? Because, as discussed previously, the patient remains, in an extended but real sense, the decision-​maker. The surrogate’s role is about protecting a patient’s right to self-​determination, protecting those constitutionally guaranteed liberty interests, and making effective the patient’s right to live by his or her values, at a time when the patient cannot him or herself exercise that right. The fact that the decision should be the patient’s must be central to all discussion and deliberation going forward. Finally, if the this discussion of the moral and political basis for the surrogate’s role is on target, prospective surrogates should be informed about the surrogate’s role as early in the course of care as possible to ensure that they are willing and able to play it. Such an informed consent for possible surrogates could include, among other things, an explicit discussion of the “decision-​maker” misconception, an affirmation that all decision-​making going forward will as much as possible be toward the end of honoring and respecting the fact that decision-​making authority comes from the patient, an acknowledgement of the stresses and challenges for surrogates as reported in the literature, and so forth. Doing an informed consent for the surrogate’s role would also presumably have the added benefits of helping to identify people who may struggle in the surrogate’s role and enhancing clinician–​family communication, which has been shown in a number of studies to provide help surrogates, patients, and clinicians alike in dealing with these emotionally charged and stressful situations (Majesko et  al., 2012; Torke, Petronio, Purnell, et al., 2012; Torke, Petronio, Sachs, et al., 2012).

38   Mark P. Aulisio

Concluding Thoughts I return now to the case of the “Beloved Brother” because, although we probably could not have articulated it at the time, what my colleague and I actually did in the case is very close to what I am proposing here, albeit with one very significant difference. In response to the sister’s anguished assertion that the care team wanted us “to convince [her] to give up on [her] brother,” I emphasized that no one—​no doctor, no nurse or social worker, and certainly not two ethicists—​should ask her to give up on her brother. I told her also that we all should be focused on her brother so that we can respect and honor him as much as possible in what were surely his last days. This seemed to make her feel a bit better. My colleague then jumped in to affirm what had just been said and to ask her to explain in her own words what she took her brother’s condition to be and where she thought things were heading. She responded that she knew her brother was really sick and that he was probably going to die, but that she needed to give him every chance. She also said, emphatically, that there was “one doctor who said that there’s a chance he could get better!” My colleague gently probed a little to better understand when the comment was made and by whom. He also tried to elicit from her a sense of what she took “chance” and “get better” to mean. It turned out that the comment was made very early in the patient’s care, but it stuck with the sister as a buoy to her flagging hope. It also turned out that the sister took “chance” as something like a realistic chance and “better” to be, at a bare minimum, that he would be able to talk and interact with them in much the way he had before falling ill. At that point, my colleague clarified that her brother was dying and that there was no longer any real chance for him to return to anywhere near his baseline. The conversation then took a turn toward discussing what her brother was like. He was, she said, an avid gardener. He also, she added with a laugh, made a habit of playing with the ants that crawled about as he gardened—​earning him the endearing nickname “Ants” (her brother affirmed this with a chuckle). “Ants” was very active in a variety of other pursuits and had, it turned out, lived for over thirty years with his brother (who remained present but who had not said a word since the start of the meeting, sitting quietly at his sister’s side, save for the chuckle). My colleague asked her if she had ever spoken with her sick brother about what he would want if faced with this kind of decision. She was clear that she had not. I then asked her what she thought her brother, the patient, might want, and before she could reply her heretofore silent brother blurted out “Ain’t no way my brother’d want this! Ain’t no way!” I can still see the startled and relieved look on her face as she glanced at her him. Much more than merely coming to a clearer understanding of sick brother’s wishes, she, at that moment, I believe, came to a proper understanding of the surrogate’s role. She had come to recognize that the decision at hand was neither hers, nor ours, nor that of the care team but, rather, her sick brother’s. The burden of being the decision-​maker was lifted—​she qua decision-​maker had been supplanted by her sick brother. A day later, she had “come around” (to use the words of the nursing staff). Life supports were withdrawn with the sister, her brother, and pastoral care keeping vigil, and the patient was allowed to die.

The Surrogate’s Role   39 In thinking back on the case, I wonder how things might have been different had an informed consent for the surrogate’s role been done with Ants’s brother and sister early on. What if, from the outset, she had known that she was not the “decision-​maker”? What if, from the outset, all involved saw the decision as fundamentally not theirs but the patient’s? What if, from the outset, she and her nearly silent brother were focused on helping members of the care team learn the patient’s story, replete with his values, beliefs, and idiosyncrasies insofar as these might be relevant for decision-​making? What if, from the outset, all involved were asking how best to honor and respect the patient in his last days? Though Ants’s case is past, the “silver tsunami” is just offshore. It is high time to reconceive the surrogate’s role.

References Annas, G. J. (1990). “Nancy Cruzan and the right to die.” New England Journal of Medicine 323(10): 670–​673. Annas, G. J., B. Arnold, M. Aroskar, et al. (1990). “Bioethicists’ statement on the U.S. Supreme Court’s Cruzan decision.” New England Journal of Medicine 323(10): 686–​687. Aulisio, M. P. (1999). “Standards for decision making at the end of life.” In Advance directives and surrogate decision making in Illinois, edited by T. May and P. Tudico. Springfield: Southern Illinois University, 21–​33. Aulisio, M. P. (2003). “Meeting the need: Ethics consulation in health care today.” In Ethics consultation: from theory to practice, edited by M. P. Aulisio, R. M. Arnold, and S. J. Youngner. Baltimore: Johns Hopkins University Press, 1–​22. Aulisio, M. P. (2014). Advance directives and the physician-​patient relationship: A surprising metamorphosis. In Advance directives, edited by P. Lack, N. Biller-​Adorno, and S. Brauer. Dordrecht, The Netherlands: Springer, 127–​152. Azoulay, E., F. Pochard, N. Kentish-​Barnes, et al. (2005). “Risk of post-​traumatic stress symptoms in family members of intensive care unit patients.” American Journal of Respiratory and Critical Care Medicine 171(9): 987–​994. Barrett, K. A., and D. C. Scales (2012). “Considering the vulnerabilities of surrogate decision-​ makers when obtaining consent for critical care research.” Intensive Care Medicine 38(1): 4–​6. Braun, U. K., A. D. Naik, and L. B. McCullough. (2009). “Reconceptualizing the experience of surrogate decision making: reports vs genuine decisions.” The Annals of Family Medicine 7(3): 249–​253.

Cruzan v. Director, Missouri Department of Health. (1990). 497 U.S. 261.

Dunn, L. B., and Y. Alici. (2013). “Ethical waves of the silver tsunami: Consent, capacity, and surrogate decision-​making.” American Journal of Geriatric Psychiatry 21(4): 309–​313.

Federal Patient Self Determination Act. (1990). 42 U.S.C. 1395 cc (a).

Fleetwood, J. E., R. M. Arnold, and R. J. Baron. (1989). “Giving answers or raising questions? The problematic role of institutional ethics committees.” Journal of Medical Ethics 15(3): 137–​142. Furrow, B. R. (2008). Bioethics: Health care law and ethics. St. Paul, MN: Thomson/​West. Hickman, R. L., Jr., B. J. Daly, and E. Lee. (2012). “Decisional conflict and regret: Consequences of surrogate decision making for the chronically critically ill.” Applied Nursing Research 25(4): 271–​275.

40   Mark P. Aulisio In re Quinlan. (1976). 70 N.J. 10, A. d.

Iverson, E., A. Celious, C. R. Kennedy, et  al. (2012). “Real-​time perspectives of surrogate decision-​makers regarding critical illness research: Findings of focus group participants.” Chest 142(6): 1433–​1439. Iverson, E., A. Celious, C. R. Kennedy, E. Shehane, A. Eastman, V. Warren, and B. D. Freeman. (2013). “Perspectives of surrogate decision makers for critically ill patients regarding gene variation research.” Genetics in Medicine 15(5): 368–​373. Jonsen, A. R. (2003). The birth of bioethics. New York: Oxford University Press. Majesko, A., S. Y. Hong, L. Weissfeld, and D. B. White. (2012). “Identifying family members who may struggle in the role of surrogate decision maker.” Critical Care Medicine 40(8): 2281–​2286. Pence, G. E. (2004). Classic cases in medical ethics: Accounts of cases that have shaped medical ethics, with philosophical, legal, and historical backgrounds. Boston: McGraw-​Hill. Quinlan, J. (2005). My joy, my sorrow:  Karen Ann’s mother remembers. Cincinnati, OH:  St. Anthony Messenger Press. Quinlan, J., and J. Quinlan. (1977). Karen Ann:  The Quinlans tell their story. Garden City, NY: Doubleday. Schenker, Y., M. Crowley-​Matoka, D. Dohan, G. A. Tiver, R. M. Arnold, and D. B. White. (2012). “I don’t want to be the one saying ‘we should just let him die’: Intrapersonal tensions experienced by surrogate decision makers in the ICU.” Journal of General Internal Medicine 27(12): 1657–​1665.

Schloendorff v. Society of New York Hospital. (1914). N. Y., 105 N.E. 92.

Shalowitz, D. I., E. Garrett-​Mayer, and D. Wendler. (2006). “The accuracy of surrogate decision makers: A systematic review.” Archives of Internal Medicine 166(5): 493–​497. Silveira, M. J., S. Y. H. Kim, and K. M. Langa. (2010). “Advance directives and outcomes of surrogate decision making before death.” New England Journal of Medicine 362(13): 1211–​1218. Song, M. K., and S. E. Ward (2013). “Disconnect between emergency contacts and surrogate decision-​makers in the absence of advance directives.” Palliative Medicine 27(8): 789–​792. Sullivan, D. R., X. Liu, D. S. Corwin, et al. (2012). “Learned helplessness among families and surrogate decision-​makers of patients admitted to medical, surgical, and trauma ICUs.” Chest 142(6): 1440–​1446. Sulmasy, D. P., and L. Snyder (2010). “Substituted interests and best judgments: An integrated model of surrogate decision making.” JAMA 304(17): 1946–​1947. Torke, A. M., S. Petronio, C. E. Purnell, G. A. Sachs, P. R. Helft, and C. M. Callahan. (2012). “Communicating with clinicians: The experiences of surrogate decision-​makers for hospitalized older adults.” Journal of the American Geriatrics Society 60(8): 1401–​1407. Torke, A. M., S. Petronio, G. A. Sachs, P. R. Helft, and C. Purnelll. (2012). “A conceptual model of the role of communication in surrogate decision making for hospitalized adults.” Patient Education and Counseling 87(1): 54–​61. Torke, A. M., G. A. Sachs, P. R. Helft, K. Montz, S. L. Hui, J. E. Slaven, and C. M. Callahan. (2014). “Scope and outcomes of surrogate decision making among hospitalized older adults.” JAMA Internal Medicine 174(3): 370–​377. US President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. (1983). Deciding to forego life-​sustaining treatment:  A  report on the ethical, medical, and legal issues in treatment decisions. Washington, DC:  President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research.

The Surrogate’s Role   41 Wendler, D., and A. Rid. (2011). “Systematic review: The effect on surrogates of making treatment decisions for others.” Annals of Internal Medicine 154(5): 336–​346. White, D. B., S. M. Cua, R. Walk, et al. (2012). “Nurse-​led intervention to improve surrogate decision making for patients with advanced critical illness.” American Journal of Critical Care 21(6): 396–​409. White, D. B., G. Malvar, J. Karr, B. Lo, and J. R. Curtis. (2010). “Expanding the paradigm of the physician’s role in surrogate decision-​making: An empirically derived framework.” Critical Care Medicine 38(3): 743–​750. Youngner, S. J., D. L. Jackson, C. Coulton, B. W. Juknialis, and E. M. Smith. (1983). “A national survey of hospital ethics committees.” Critical Care Medicine 11(11): 902–​905.

Chapter 3

De ath at the Be g i nni ng  The Neonatal Intensive Care Unit Renee D. Boss

Introduction Providing intensive medical resources for very premature and sick infants is primarily a phenomenon of the developed world. The financial, technological, and labor-​intensive investments available in developed countries result in questions about whether those resources ought to be applied to all seriously ill infants and which stakeholders should participate in those decisions. Regardless of how far the envelope is pushed, there will always be a group of infants with life-​limiting conditions who will die in the first days, weeks, or months of life. How to best provide care for the infants and families in these circumstances continues to evolve. This article reviews how advances in obstetrics and neonatology have coincided with advances in palliative medicine and hospice care. Drawing from landmark ethical and legal cases, we examine how personal and clinical decisions about which babies should, and should not, receive life-​prolonging therapies have been debated in national and international conversations and in conversations at the bedside. These cases leave lingering uncertainties, including questions about personhood (should an infant with no forebrain and no capacity for consciousness receive a surgical feeding tube?), definitions of a life worth living (should parents have the right to request euthanasia of their infant with spina bifida?), and futility (can mechanical ventilation be withheld from an infant dying from an incurable metabolic disease?). We then evaluate how these ongoing challenges in the care of very sick infants continue to impact issues of informed consent, parent autonomy, resource allocation, and clinicians’ professional responsibilities.

Neonatal Intensive Care Unit    43

Prenatal Diagnosis and Decision Making Childbirth has always been a risky affair for mother and infant alike. Medical advances in perinatal care were initially focused solely on reducing maternal morbidity and mortality. Antibiotics for peripartum infections, interventions for preeclampsia, and greater delay between pregnancies because of access to birth control resulted in significant improvements in maternal outcomes. These advances made it possible for greater attention to be paid to improving fetal and infant outcomes. Prenatal fetal screening and testing has become widespread since the development of routine early pregnancy testing in the late 1970s. Safe access to therapeutic abortion in the case of serious fetal anomalies also fueled demand for prenatal testing. The evolution of prenatal testing led to new questions about maternal versus fetal best interests, as well as questions about what degree of severity of a fetal diagnosis would outweigh the risks of miscarriage associated with testing for that diagnosis. Parents who would consider pregnancy termination in the case of a serious fetal condition such as trisomy 13 have had to weigh the possibility that chorionic villous sampling or amniocentesis could lead to the death of a fetus whose testing may be normal. Options for noninvasive prenatal testing are continually expanding and are predicted to become standard of care. Cell-​free fetal DNA extracted from maternal serum in the first trimester, for example, may diagnose trisomies 21 and 18 with nearly 99% certainty (Chiu et al. 2011). The capacity for noninvasive fetal whole genome sequencing, as well as preimplantation embryonic genome sequencing, may be on the horizon. Such technologies have the potential to detect a much broader range of fetal diagnoses while minimizing the risk of fetal loss. This widened scope of prenatal testing raises a whole new set of questions about the impact on decisions regarding pregnancy termination and fetal treatments and interventions. Will more families be burdened by detection of a fetal genetic mutation of uncertain significance? Will clinicians receive prenatal requests for pregnancy termination for a fetal BRCA1 mutation? Will sex-​selection abortion become more common? Currently the availability of medical interventions to treat most serious fetal diagnoses lags significantly behind the ability to detect those diagnoses. A few fetal conditions can be ameliorated with maternal medications. For a handful of anatomic abnormalities, fetal surgery may be considered. Open fetal surgery entails significant risk of premature birth and its associated complications, including fetal loss; minimally invasive techniques are being developed. Though these options are largely considered experimental, they are actively advertised by some centers for fetal diagnoses associated with both significant potential neonatal mortality (e.g., congenital diaphragmatic hernia) and some with low potential mortality but high potential morbidity (e.g., spina bifida). Risky fetal interventions and pregnancy termination for nonlethal diagnoses can be

44   Renee D. Boss chosen by parents because the fetus lacks legal personhood. Parents are not compelled to justify their values about “acceptable” quality of life for their child when making prenatal decisions. A  parent could choose pregnancy termination for fetal spina bifida because they find lower extremity paralysis to confer an unacceptable quality of life. As discussed below, once an infant is born with that same diagnosis, parental values about quality of life are no longer absolute. Stakeholders in these decisions about prenatal testing and pregnancy termination are found, of course, not only among patients and clinicians. Even as technologies bring greater opportunities for prenatal diagnosis, there are increasing political and legal efforts to direct parent and clinician decision making. Several states have proposed legislation permitting clinicians to withhold information about a fetal diagnosis from parents if it is believed the parents will choose pregnancy termination based on those results. Other states require physicians to tell parents who are considering pregnancy termination about potential adverse maternal outcomes that have questionable if any scientific basis. These efforts raise several important concerns for clinicians. Should clinicians, for example, be allowed to hold fetal interests above maternal interests? And what is the responsibility of professional medical organizations to advocate professional ethical duties in the face of state legal limitations?

Counseling in the Window of Acute Neonatal Illness Though fetal treatments for serious diagnoses remain very limited, interventions to permit survival of sick and premature neonates have advanced significantly since 1975 when neonatal-​perinatal medicine became a board certified subspecialty. The late 1980s saw the FDA approval of commercial surfactants, which radically improved the survival for premature infants. While in the 1970s it was not uncommon for infants born only six weeks prematurely to succumb to respiratory insufficiency, today some infants born as many as eighteen weeks early will survive. Advances in medical and surgical management of congenital anomalies, metabolic syndromes, sepsis, and nutrition have resulted in dramatic increases in neonatal intensive care unit (NICU) admissions and lengths of stay. As many as 70% of NICU patients are in the NICU for more than twenty days, compared with less than 5% of adult ICU patients having an ICU stay of more than seven days (Gruenberg et al. 2006; Kornhauser and Schneiderman 2010). These gains in neonatal survival are accompanied by increased rates of serious, chronic infant disease that persist well beyond the neonatal period. Premature infants can have pulmonary insufficiency for months to years and require long-​term medications or home oxygen. Some require tracheotomy and mechanical ventilation in order to leave the hospital. Feeding and growth problems are common. Infants with intracranial bleeding from prematurity, or with brain ischemia related to birth asphyxia, are at

Neonatal Intensive Care Unit    45 high risk of cerebral palsy, mental retardation, blindness, deafness, and behavioral problems (Hack and Fanaroff 2000). Neonatal conditions resulting in subsequent chronic functional limitations account for a large proportion of pediatric health-​care resources and an increasing proportion of overall health-​care spending (Fiks et al. 2012; Moses et al. 2013). Because surfactant therapy dates only to the late 1980s, the earliest recipients have just reached adulthood. Children and adolescents who were born prematurely, especially before twenty-​seven weeks gestation, have been found to be at increased risk of anxiety disorders (Somhovd et al. 2012), significant motor impairment (de Kieviet et al. 2009), decreased cognitive scores and academic performance (Short et al. 2003), lower athletic and job competence (Grunau, Whitfield, and Fay 2004), and lower romantic confidence (Grunau, Whitfield, and Fay 2004). Nevertheless, data suggest parents may feel more positively than do clinicians about these challenges (Streiner et al. 2011), and adolescents themselves may feel no different from their peers (Dahl et al. 2006). For many newborns with life-​threatening illness, the initial hours and days after birth involve aggressive support of the neonate: artificial ventilation, pressors, central lines, surgery, blood products. Some infants die despite application of these therapies. Other infants quickly stabilize, inspiring optimism about their potential for recovery. But for some infants, aggressive resuscitation and intensive care barely achieve targeted cardiorespiratory and metabolic parameters. In such scenarios, there are concerns not only about the underlying diagnosis but about ongoing physiologic and especially neurologic damage, which increase the risk of serious long-​term disability should the infant survive. It is in this window of acute neonatal illness that clinicians and families face decisions about whether or not to withdraw life-​sustaining therapies. Acute infant suffering, the uncertain outcome of additional interventions, the burden of chronic pediatric illness, and values about a life worth living must all be weighed. In some cases, there is an element of time pressure to make a decision before spontaneous breathing may develop and result in survival with very poor neurodevelopmental prognosis. Ideal methods of counseling families in these situations are not clear, leaving substantial room for individual clinician judgment.

A Life Worth Living—​L andmark Cases with Residual Effects Several legal and ethical cases in the United States and elsewhere have challenged—​and in many cases redefined—​decision making for sick newborns. The Baby Doe cases (Antommaria 2006) involved disputes about whether parents whose infant had trisomy 21, or Down’s syndrome, could refuse surgical correction of gastrointestinal anomalies commonly associated with the syndrome.

46   Renee D. Boss The justification given by parents for refusing these life-​saving surgical interventions was the perceived unacceptable quality of life for persons with trisomy 21. The Baby Jane Doe case in 1983 involved a family who declined surgical repair of their infant’s meningomyelocele due to the potential of survival with cognitive impairment. These cases catalyzed public discourse about personhood, specifically whether it was acceptable to withhold medical and surgical interventions from infants solely on the basis of underlying cognitive impairment. Responding to these cases, the U.S. Surgeon General proposed the Baby Doe Amendment to preclude the withholding of medical treatment from children with disabilities. In its strictest interpretation, the amendment aimed to separate quality-​of-​life concerns from decisions about life-​ sustaining therapies. While these regulations were not upheld by the U.S. Supreme Court, federal funding to hospitals remains contingent on state systems for pursuing child abuse or neglect investigations for cases where life-​sustaining therapies are withheld from children with disabilities. One important outcome of the Baby Doe cases was the recommendation for the development of hospital ethics committees to participate in serious medical decisions (President’s Commission for the Study of Ethical Problems in Medical and Biomedical and Behavioral Research 1983). While mechanisms for ethical oversight of local medical decisions by hospitals are increasingly mandated, the quality of such mechanisms is variable and their standards not regularly assessed. Ongoing confusion about the reach of Baby Doe regulations and concerns about legal culpability persist among clinicians facing decisions about life-​sustaining therapies for infants with very poor neurologic prognosis. A strict interpretation of these rules suggests that neither parents nor clinicians can elect to withhold or withdraw medical interventions from an infant even if that infant is likely to never have meaningful interaction with his or her environment. Yet multiple major professional groups, including the American Academy of Pediatrics, have maintained that clinician, and especially parent, quality-​of-​life considerations should be a part of medical decisions for such infants (Bell 2007). Data suggest that neonatologists and families do sometimes decide to limit therapies solely based on quality of life concerns (Barton and Hodgman 2005; Singh, Lantos, and Meadow 2004). The tension between public legislation and debate, and private decisions made at the bedside, remains unresolved. A decade after Baby Doe, the Baby K case extended discourse about definitions of personhood and medical futility in cases of infant with severe neurologic impairments (Flannery 1995). Baby K was diagnosed prenatally with anencephaly, or congenital absence of the cerebrum. Infants with this condition are considered to never have the capacity for consciousness. The infant’s mother declined pregnancy termination and refused limitations of life-​sustaining therapies for the infant after birth. The treating physicians and the hospital argued that respiratory support for the infant was futile, but the state’s District Court argued that mechanical ventilation was not futile because it successfully treated respiratory failure. The court did not take into account the underlying diagnosis that resulted in respiratory failure or the fatal prognosis for anencephaly. The case raised many concerns about the scope of clinicians’ judgments about when

Neonatal Intensive Care Unit    47 a life-​sustaining therapy was indicated. It also drew sharp limits around quality-​of-​life considerations, suggesting that indefinite medical care should be provided to persons with minimal or no brain activity. In the past two decades, several cases have addressed medical advances in caring for extremely premature infants. These cases reflect a sensibility that not only should life-​ sustaining therapies be provided to all children regardless of neurodevelopmental status but these therapies may be considered obligatory even in the perinatal period for infants with minimal chance of survival. In Texas, a family chose compassionate care only for their daughter who was about to be born extremely prematurely at twenty-​two weeks gestation—​a gestational age where survival is less than 10%. The neonatologists were bound by the hospital to resuscitate the infant if she was born with any heart rate. The infant was resuscitated over the parents’ wishes and ultimately survived with profound disabilities. The family sued based on wrongful life; compensatory and punitive damages were awarded but were later reversed on appeal (Miller v. HCA 2003). The outcome is to extend restrictions on parent decision making even to the prenatal period. National efforts have also aimed at obligating neonatal resuscitation regardless of clinical circumstances. The 1986 Emergency Medical Treatment and Active Labor Act requires hospitals who receive Medicare funding to treat any patient presenting to the emergency department, such as a woman with imminent labor, without regard to that patient’s ability to pay for medical care. The Born Alive Infant Protection Act of 2002 extends the mandate to the medical stabilization of any infant born alive. How the act has actually impacted individual clinician care for extremely premature infants is not clear (Partridge et al. 2009; Sayeed 2005). What is clear is that, despite the fact that outcomes for the most extremely premature infants remain poor, the pendulum is swinging from medical interventions viewed as daring attempts to save a few infants to compulsory attempts to save all infants. In light of our rapidly advancing abilities to make prenatal diagnoses and political efforts to restrict abortion access, it is likely that more and more women who would chose to terminate or withhold resuscitation in the case of a severe fetal diagnosis may be prohibited from doing so. While sanctity of life concerns increasingly aim to limit medical decision making for newborns in the United States, other countries have focused not on eliminating quality-​of-​life considerations completely but on formalizing them. For example, the Groningen Protocol (Verhagen and Sauer 2005) was created in 2004 in the Netherlands. The Protocol went beyond authorizing the withholding of treatments and established a pathway for pediatricians to euthanize infants with no hope of meaningful survival. The Protocol formalized the medical and legal process of determining when it is acceptable to deliberately end the life of an infant perceived to have intractable suffering. Criticisms of the Protocol abound in the international literature, many focused on the permissibility of euthanizing infants not due to terminal conditions but purely because of conditions associated with diminished quality of life, such as spina bifida (Jotkowitz, Glick, and Gesundheit 2008; Kon 2008). Others argue that the degree of infant suffering is unknowable and often able to be addressed by pain and symptom management. The Protocol has rarely been used in clinical practice.

48   Renee D. Boss

Advances in End-​of-​L ife Care Across the Life Spectrum While survival for infants with life-​threatening conditions has greatly improved in recent decades, it will always be true that some infants will have diseases or malformations so severe that corrective interventions will not be possible. In addition, there will be some families and clinicians who determine that a therapeutic benefit, while possible, is outweighed by the associated burdens of those therapies. Helping these infants and families will remain a core challenge for pediatricians. End-​of-​life care in the perinatal and neonatal period continues to evolve. Academic scholarship in neonatology has traditionally focused on basic science research, or on clinical practices to improve survival and long-​term outcomes for neonates. A small but growing body of neonatology scholarship has examined how to maximize quality of life for newborns and families when the infant is unlikely to survive the hospitalization. For nearly three decades, researchers and clinicians have recognized that newborns, even those born at extremely premature gestational ages, are capable of sensing pain. A robust literature describes neonatal pain perception, including physiology, signs and symptoms, interventions, and long-​term outcomes. When exposed to painful stimuli, infants of varying gestational ages respond with autonomic and behavior changes, including tachycardia, changes in blood pressure, facial movements, and oxygen desaturations. Neonates exposed to many painful procedures, as occur with prolonged NICU hospitalizations, may develop hyperalgesia (Taddio 2002). Optimal strategies for preventing and treating neonatal pain have had to account for the significant differences in pharmacokinetics that characterize infants of varying postconceptional ages due to maturational differences in skin integrity, total body water, as well as hepatic and renal metabolism. Because of this, many fewer medications and medication-​delivery devices exist for infants than for older children or adults. Just as with pain management at the end of an adult’s life, clinicians may worry that aggressive neonatal pain management may hasten death. One study showed that median time to death for neonates after withdrawal of life-​sustaining therapies was approximately 20 minutes and was not different between infants who received no opiate and those who received suprapharmacologic opiate doses (Partridge and Wall 1997). Maximizing an infant’s quality of life in the NICU is, in part, complicated by the fact that there are no standard methods to assess a NICU infant’s quality of life. To date, there exists one quality-​of-​life measure designed specifically for infants, but it was validated on an outpatient population of healthy infants (Varni 2010). There exist no quality-​ of-​life measures for children of any age in an ICU. While a minority of adult quality-​ of-​life measures have been designed for patients in an ICU, as noted above, adults are typically in the ICU for days while infants are often in the NICU for weeks to months. Development of a standardized tool could both improve our ability to enhance quality of life and could serve as a resource for weighing benefits and burdens of care (Boss,

Neonatal Intensive Care Unit    49 Kinsman, and Donohue 2012). In the absence of this, maximizing an infant’s quality of life in the NICU may entail optimal pain management as above; minimizing thermal stress; avoiding exposure to intense light or noise, which are known to damage developing neonatal tissues; promoting non-​nutritive sucking and/​or feedings for comfort; and enabling families to spend as much time with the infant as possible. Skin-​ to-​skin between parents and neonates can have multiple benefits, including temperature control, decreased infant crying, and increased bonding (Moore, Anderson, and Bergman 2007). Whether or not to withhold nutrition and hydration for dying infants remains controversial. The American Academy of Pediatrics supports clinicians and families who make the decision to withhold artificial nutrition and hydration from dying infants and children in the case where those fluids are not felt to be of benefit (Diekema and Botkin 2009). Nevertheless, some argue that the situation is not analogous to an adult patient who chooses to forgo nutrition and hydration, as children are not able to make these decisions for themselves. Parents and clinicians alike may see a special significance of feeding for an infant, as this is a primary connection between parent and infant. Clinicians who care for an infant after the decision has been made to withhold fluid and nutrition may struggle to prepare families for the many days it will take their infant to die as well as the changes in their infant’s appearance (“How It Feels to Withdraw Feeding” 2012). Supporting families whose infant may not survive to ever leave the hospital after birth is challenging in many respects. As many as half of the families in these circumstances had no prenatal complications, no warnings that their infant might not be a healthy baby, and no chance to prepare for NICU hospitalization or infant loss. The family endures shock and disbelief not unlike that of a parent whose child dies a sudden death due to drowning or a motor vehicle accident. These families need intensive emotional, spiritual, informational, and logistical supports. Although families who knew prenatally that their infant would have a serious medical condition may have had more time to prepare, they are also in need of intensive supports. Most parents in these situations are young, have not experienced a life-​threatening illness in a loved one, and have limited exposure to serious medical decision making. Many mothers have the additional physical challenges related to postpartum healing and either maintaining or suppressing lactation. Siblings who are excited about a new baby yet who vary in their developmental capacity to understand the infant’s illness often need help understanding what is happening and why they may not be allowed to visit the NICU. Hospice and palliative medicine is becoming increasingly available to families in these situations. Palliative care can be incorporated into NICU care for all infants with life-​threatening conditions, including those born extremely premature. Palliative care may be routinely integrated by neonatology physicians and nurses into intensive care for all newborns or can be achieved via specialist consultation services for select infants. Many programs use a combination of these approaches. At the same time that NICUs were coming into existence, Dame Cicely Saunders catalyzed medical attention to end-​of-​life and palliative care. Hospital-​based palliative care programs for infants and children have developed more slowly than for adult patients.

50   Renee D. Boss The Institute of Medicine and the American Academy of Pediatrics both affirm that palliative care should occur alongside curative care for children with life-​threatening illnesses, regardless of whether the outcome is cure or death. Yet the resources to make that care possible for all of the children with life-​threatening illness are not yet in place. The majority of children who die do not receive hospice services, and very few infants receive hospice services in part because of limited training among adult and even pediatric hospice staff to provide the specialized services that infants may need.

Ongoing Challenges in End-​of-​L ife Care for Newborns Despite, and in some cases because of, the advances in caring for very sick newborns, there remain multiple complexities in determining the appropriateness of applying that technology to individual patients. Many of these complexities involve determining optimal care in the setting of extremely premature birth. The incidence of premature labor is increasing, in part due to increased use of assisted reproduction. As many as half of all pregnancies after assisted reproduction result in multiple gestation; these infants are at an increased risk of congenital anomalies, as well as of premature birth. Because of the potential for a poor outcome at the earliest gestational ages, the standard of care is for physicians to engage in collaborative decision making with families to determine whether neonatal resuscitation is consistent with their goals and values. Some have argued that policies that permit selective nonresuscitation at these gestational ages result in a self-​fulfilling prophecy whereby actual survival rates for these infants remain low. This may lead clinicians to underestimate potential survival when counseling families (Tyson et al. 2008). Yet when a pregnant woman presents with extremely premature labor, there are few objective criteria to reduce uncertainty about whether the application of intensive resuscitation and ICU care for this infant will result in a perfectly healthy infant, an infant who dies in the first hours of life, or an infant who survives but will require long-​term tracheotomy and a feeding tube. How then can optimal care for this individual infant be defined? Is optimal care that which minimizes pain and suffering, or that which extends life for as long as possible? What threshold of meaningful survival is worth what threshold of pain and suffering? What threshold of meaningful survival should exist before parents are offered an option of resuscitation? Counseling of these families has traditionally emphasized short-​term outcomes—​ those complications that could occur in the first hours, days, and weeks in the NICU. Less emphasis has been placed on the potential long-​term problems such as cerebral palsy, mental retardation, learning disability, and behavioral problems. Despite professional expectations that neonatologists will help families make decisions for their infant, neonatologists have generally not been trained to have difficult conversations with

Neonatal Intensive Care Unit    51 families about their goals and values related to disability or quality of life (Boss et al. 2009). Extensive data about what parents need from this counseling are not available. The subjective variability that can characterize these conversations should not be underestimated, particularly with regard to the value placed on the benefits and burdens of care. How should clinicians help parents discuss what they understand to be a life worth living? Communication education must be prioritized in order to provide this standard of care and to obtain genuinely informed consent from families. Decision support tools, which have helped adult patients make serious medical decisions in other contexts, may also be of use in these settings. It is unclear how, or if, resource allocation considerations should be incorporated into decisions about whether to engage in prolonged, expensive NICU care, which is predicted to have >90% chance of death before hospital discharge or survival with a lifetime of medical, educational, and financial needs. During prenatal counseling, discussions of a family’s economic resources and potential to care for a child with neurodevelopmental disability and complex medical problems are not uniformly included. Nevertheless, systematic barriers to health-​care access can make it more likely that women without economic resources will deliver infants at risk of needing NICU care. For example, a pregnant woman in the United States with private health insurance can, on the day of delivery, request and obtain a tubal ligation to prevent further pregnancies. But Medicare requires that pregnant women wait thirty days between the request for tubal ligation and the actual procedure. This systematic barrier increases a poor woman’s risk of both undesired and inadequately spaced pregnancies, which also places those infants at risk of complications. Medical interventions for the most premature infants vary across hospitals and geographic regions. NICUs are concentrated in certain geographic areas; women who deliver premature infants in these cities are more likely to have their extremely premature infant resuscitated and treated. Mothers who deliver premature infants at hospitals with religious affiliations often are not given the option of nonresuscitation; the clinicians at these centers are under no obligation to tell families that they could have the option of compassionate care only at another facility. These restrictions are not limited to religious health systems; some clinicians and hospitals have gone so far as to seek court-​ordered hospitalization for women with extremely premature labor who do not want medical interventions. In some cases, judicial approval has allowed a hospital to deliver medical care against a woman’s wishes, in order to protect the life of the fetus (Cantor 2012). Variability in medical care for sick infants in the United States may also vary by patient race and ethnicity. Although the overall infant mortality rate has steadily declined over the past four decades, most of the improvement has been among white infants. Infant mortality rates are higher among non-​Hispanic black, American Indian/​ Alaskan Native, and Puerto Rican infants than for non-​Hispanic white infants (Hessol and Fuentes-​Afflick 2005). The greatest racial disparities in infant mortality seem to be related to preterm birth, even though data suggest that extremely premature non-​ Hispanic black infants are intubated and resuscitated more commonly than are Hispanic and non-​Hispanic white infants (Hauck, Tanabe, and Moon 2011; Moseley et al. 2004).

52   Renee D. Boss

Conclusions In parts of the world with the resources to support neonatal intensive care, families of very sick infants have many more options, and are asked to make many more decisions, than did families just a few decades ago. Beginning even before birth, families and clinicians may struggle to weigh predicted benefit and burden in the setting of uncertainty; some of these choices are irrevocable. Hospital policies and legal milieus may compel different decisions despite identical circumstances. Authentic informed consent from families will be enhanced by educational efforts targeted at helping clinicians talk with families about deeply held values.

References Antommaria, A. M. 2006. “Who should survive? One of the choices on our conscience”: Mental retardation and the history of contemporary bioethics. Kennedy Institute of Ethics Journal 16(3):205–​224. Barton, L., and J. E. Hodgman. 2005. The contribution of withholding or withdrawing care to newborn mortality. Pediatrics 116(6):1487–​1491. Bell, E. F. 2007. Noninitiation or withdrawal of intensive care for high-​risk newborns. Pediatrics 119(2):401–​403. Boss, R. D., N. Hutton, P. K. Donohue, and R. M. Arnold. 2009. Neonatologist training to guide family decision making for critically ill infants. Archives of Pediatrics and Adolescent Medicine 163(9):783–​788. Boss, R. D., H. I. Kinsman, and P. K. Donohue. 2012. Health-​related quality of life for infants in the neonatal intensive care unit. Journal of Perinatology 32(12):901–​906. Cantor, J. D. 2012. Court-​ordered care—​A complication of pregnancy to avoid. The New England Journal of Medicine 366(24):2237–​2240. Chiu, R. W., R. Akolekar, Y. W. Zheng, T. Y. Leung, H. Sun, K. C. Chan, et al. 2011. Non-​invasive prenatal assessment of trisomy 21 by multiplexed maternal plasma DNA sequencing: large scale validity study. BMJ 342:c7401. Dahl, L. B., P. I. Kaaresen, J. Tunby, B. H. Handegard, S. Kvernmo, and J. A. Ronning. 2006. Emotional, behavioral, social, and academic outcomes in adolescents born with very low birth weight. Pediatrics 118(2):e449–​e459. de Kieviet, J. F., J. P. Piek, C. S. Aarnoudse-​Moens, and J. Oosterlaan. 2009. Motor development in very preterm and very low-​birth-​weight children from birth to adolescence:  A  meta-​ analysis. JAMA 302(20):2235–​2242. Diekema, D. S., and J. R. Botkin. 2009. Clinical report—​Forgoing medically provided nutrition and hydration in children. Pediatrics 124(2):813–​822. Fiks, A. G., S. Mayne, A. R. Localio, E. A. Alessandrini, and J. P. Guevara. 2012. Shared decision-​ making and health care expenditures among children with special health care needs. Pediatrics 129(1):99–​107. Flannery, E. J. 1995. One advocate’s viewpoint: Conflicts and tensions in the Baby K case. The Journal of Law, Medicine and Ethics 23(1):7–​12; discussion 13–​14.

Neonatal Intensive Care Unit    53 Gruenberg, D. A., W. Shelton, S. L. Rose, A. E. Rutter, S. Socaris, G. McGee. 2006. Factors influencing length of stay in the intensive care unit. American Journal of Critical Care 15(5):502–​509. Grunau, R. E., M. F. Whitfield, and T. B. Fay. 2004. Psychosocial and academic characteristics of extremely low birth weight (< or =800 g) adolescents who are free of major impairment compared with term-​born control subjects. Pediatrics 114(6):e725–​e732. Hack, M., and A. A. Fanaroff. 2000. Outcomes of children of extremely low birthweight and gestational age in the 1990s. Seminars in Neonatology 5(2):89–​106. Hauck, F. R., K. O. Tanabe, and R. Y. Moon. 2011. Racial and ethnic disparities in infant mortality. Seminars in Perinatology 35(4):209–​220. Hessol, N. A., and E. Fuentes-​Afflick. 2005. Ethnic differences in neonatal and postneonatal mortality. Pediatrics 115(1):e44–​e51. How it feels to withdraw feeding from newborn babies. 2012. British Medical Journal 345:e7319. Jotkowitz, A., S. Glick, and B. Gesundheit. 2008. A case against justified non-​voluntary active euthanasia (the Groningen Protocol). The American Journal of Bioethics 8(11):23–​26. Kon, A. A. 2008. We cannot accurately predict the extent of an infant’s future suffering: The Groningen Protocol is too dangerous to support. The American Journal of Bioethics 8(11):27–​29. Kornhauser, M., and R. Schneiderman. 2010. How plans can improve outcomes and cut costs for preterm infant care. Managed Care 19(1):28–​30.

Miller v. HCA. 2003. 47 Tex Sup J 12, 118 S.W.3d 758.

Moore, E. R., G. C. Anderson, and N. Bergman N. 2007. Early skin-​to-​skin contact for mothers and their healthy newborn infants. Cochrane Database of Systematic Reviews 3:CD003519. Moseley, K. L., A. Church, B. Hempel, H. Yuan, S. D. Goold, and G. L. Freed. 2004. End-​of-​life choices for African-​American and white infants in a neonatal intensive-​care unit: A pilot study. Journal of the National Medical Association 96(7):933–​937. Moses, H. III, D. H. Matheson, E. R. Dorsey, B. P. George, D. Sadoff, and S. Yoshimura. 2013. The anatomy of health care in the United States. JAMA 310(18):1947–​1963. Partridge, J. C., M. D. Sendowski, E. A. Drey, and A. M. Martinez. 2009. Resuscitation of likely nonviable newborns: Would neonatology practices in California change if the Born-​Alive Infants Protection Act were enforced? Pediatrics 123(4):1088–​1094. Partridge, J. C., and S. N. Wall. 1997. Analgesia for dying infants whose life support is withdrawn or withheld. Pediatrics 99(1):76–​79. President’s Commission for the Study of Ethical Problems in Medical and Biomedical and Behavioral Research. 1983. Deciding to Forego Life-​Sustaining Treatment. Washington, DC: Author. Sayeed, S. A. 2005. Baby doe redux? The Department of Health and Human Services and the Born-​Alive Infants Protection Act of 2002: A cautionary note on normative neonatal practice. Pediatrics 116(4):e576–​e585. Short, E. J., N. K. Klein, B. A. Lewis, S. Fulton, S. Eisengart, C. Kercsmar, et al. 2003. Cognitive and academic consequences of bronchopulmonary dysplasia and very low birth weight: 8-​ year-​old outcomes. Pediatrics 112(5):e359. Singh, J., J. Lantos, and W. Meadow. 2004. End-​of-​life after birth: Death and dying in a neonatal intensive care unit. Pediatrics 114(6):1620–​1626.

54   Renee D. Boss Somhovd, M. J., B. M. Hansen, J. Brok, B. H. Esbjorn, and G. Greisen. 2012. Anxiety in adolescents born preterm or with very low birthweight: A meta-​analysis of case-​control studies. Developmental Medicine and Child Neurology 54(11):988–​994. Streiner, D. L., S. Saigal, E. Burrows, B. Stoskopf, and P. Rosenbaum. 2001. Attitudes of parents and health care professionals toward active treatment of extremely premature infants. Pediatrics 108(1):152–​157. Taddio, A., V. Shah, C. Gilbert-​MacLeod, and J. Katz. 2002. Conditioning and hyperalgesia in newborns exposed to repeated heel lances. JAMA 288(7):857–​861. Tyson, J. E., N. A. Parikh, J. Langer, C. Green, and R. D. Higgins. 2008. Intensive care for extreme prematurity—​Moving beyond gestational age. The New England Journal of Medicine 358(16):1672–​1681. Varni, J., C. Limbers, K. Neighbors, K. Schulz, J. Lieu, R. Heffer, et al. 2010. The PedsQL™ Infant Scales:  Feasibility, internal consistency reliability, and validity in healthy and ill infants. Quality of Life Research 20(1):45–​55. Verhagen, E., and P. J. Sauer. 2005. The Groningen protocol—​Euthanasia in severely ill newborns. The New England Journal of Medicine 352(10):959–​962.

Chapter 4

Dying Children a nd t h e Kindness of St ra ng e rs John D. Lantos

The central experience for families of dying children in the United States today is the experience of isolation. Childhood death has become so unusual that our communities have forgotten their rituals. The younger the child, the more true this is. Everybody knows how to respond to the death of an adult or even a teenager. But childhood death is different. When a child has a life-​threatening illness, they are moved out of their own community and into the strange world of the tertiary-​care medical center. A child’s own pediatrician or family doctor is usually not involved in the care. Instead, critically ill children are usually cared for by doctors and nurses who never knew the child or family when they were well. Thus, children and their families rely almost entirely on the kindness of strangers. In this chapter, I discuss the implications of this modern way of death by drawing on fiction, poetry, film, and memoir to try to understand what the experience is like for the parents and families. I then discuss the ways that medical centers have tried to respond to the unique challenges of childhood illness and death and review some studies of pediatric palliative care. These two sources of insight—​narrative and quantitative studies—​ are complementary. Together, they can help us understand what is happening and how, perhaps, to make it better.

The Kindness of Strangers Pedro Almodovar’s film, All About My Mother, a profound meditation on the way we die now, opens with a mysterious close-​up of something blurry and unrecognizable. As the camera pulls back and the picture comes into focus, we recognize a plastic bag of intravenous fluid. The camera pans along the IV tubing. Clear fluid drips into the Buretrol. We are at a patient’s bedside. According to cinematic conventions, we now expect to see the patient.

56   John D. Lantos But Almodovar plays with our expectations. The camera pans along the wall, showing the various outlets and nozzles for oxygen and suction and then, instead of showing us the patient, the frame shifts and the screen fills with the image of a complex machine. It has many little white and black and red and blue dials and buttons. The camera pans across the panel of buttons and comes to rest on a sheet of paper scrolling out underneath ten delicate markers. The markers are eerily still. They trace flat lines across the scrolling paper. The image of the flat line is one that we all recognize. It has become a trope in popular culture. Flat lines mean death. Brain death. We realize, disturbingly, that this hospital scene is not taking us where we thought it was going. There is no living patient. The first living character we see is neither patient nor doctor. She is the transplant coordinator. She is making a phone call. She reports on the patient’s blood type, weight, and other information that is necessary in order to allocate the organs. The transplant coordinator, Manuela, is the main character in the movie. In addition to helping doctors to procure organs, she also helps to train the doctors who must approach the loved ones of a patient who has died. She used to be an actress. Now, she role-​plays the part of a grieving relative in order to help the doctors learn how to break the bad news and make the “ask.” She helps them learn how to try to transform the story of a loved one’s death into a story of rebirth and hope. The loved one dies but his heart lives on. Through these enigmatic scenes, we glimpse the complex social organization that makes death an occasion for communal altruism. Later in All About My Mother, Manuela takes her son to see the play A Streetcar Named Desire. We see them watching the famous scene where Blanche says her famous line, a line that, in this context, seems to be as much about transplantation as it is about Blanche, “I have always depended upon the kindness of strangers.” After the play, her son wants to get an autograph from the actress who plays Blanche. He waits in the rain by the stage door until she comes out to her waiting limo. She ignores him. He knocks on her limo window. She disdainfully motions to the driver. The car drives away. The boy chases after her down the rain-​swept street, still intent on getting an autograph. As her car careens around a corner and he races after it, he is struck by an oncoming car. He falls to the street, unconscious and bleeding in the pouring rain. Then Manuela is in the ICU waiting room. The doctors, the very doctors whom she has trained to ask grieving family members about organ donation, come out of the ICU, just as they did in her role-​plays. She has become the character whom she impersonated. She collapses in tears.

The Tension Between Despair and Hope That tension between despair and hope is a central problem for doctors and parents who care for dying children. When an elderly person dies, it is sad, but it also seems somehow

Dying Children and the Kindness of Strangers    57 natural. The elderly have lived their lives. Their death doesn’t represent dashed hopes in the way that the death of a child does. When a child dies, the customary words of comfort don’t apply. The fabric of normal life has been torn. One central and eternal problem for both pediatricians and parents in such situations is to find a reason for hope in a child’s illness. One way is to hope for a miraculous cure. This hope can focus on the individual child who is critically ill. Parents hope for a new operation, a new drug, or a new device that will transform a tragic situation into a joyous one. Sometimes, this works. Most doctors have had cases where they did not think survival was possible but, against all the odds, the child pulled through. Such cases make it hard to ever give up. Another way to preserve hope is to participate in clinical research. That offers hope for future children. Overall, in the past 100 years, medical research has fulfilled its promise. We can now prevent polio, treat pneumonia, save premature babies, and cure leukemia. When treatments are not yet available, we give money for more research. Hope comes from our determination to make progress. Many parents get involved in fundraising for groups that advocate for more research, groups such as the March of Dimes. But no matter how many new discoveries we make, some children still get incurably sick. When that happens, parents find themselves in the world of the sick. Often, neither friends nor family members know how to deal with the stress, the grief, and the chronicity of life-​threatening childhood illnesses. Thus, parents find themselves bonding with other parents of sick children, people with whom they have little in common except the experience of caring for a dying child. Novelist Lorrie Moore captured this in the title of her story about her own experiences when her child was critically ill. She called the story, “People Like That Are the Only People Here” (Moore 1988). Parents and families form new communities in the “new normal” of tertiary-​care pediatrics. Almodovar is one of many writers and filmmakers to explore the experiences of parental grief and loss. There are common themes. Most parents’ first response, when told that their child is critically ill, is denial. This is quickly following by a visceral and paralyzing grief. A third common experience is the radical uncertainty that accompanies many illnesses today and that makes decision-​making particularly fraught. Fourth, the experience of caring for a critically ill child often results in a deep and profound sense of isolation.

Denial Novelist Alexander Hemon wrote about what it was like for him when his daughter was diagnosed with a brain tumor. His first response is denial: [The doctor] showed us the MRI images on his computer:  right at the center of Isabel’s brain, lodged between the cerebellum, the brain stem, and the hypothalamus, was a round thing. It was the size of a golf ball, Dr. Tomita suggested, but I’d never been interested in golf and couldn’t envision what he meant.

58   John D. Lantos It is clear from this passage that the reality of his daughter’s brain tumor has not even begun to register. Instead, his mind drifts off on an irrelevant tangent. Lorrie Moore describes a similar experience when she was told that her baby had a Wilms’ tumor (Moore 1988, 215). She thinks, Wait a minute. Hold on here. The Baby is only a baby, fed on organic applesauce and soy milk—​a little prince!—​and he was standing so close to her during the ultrasound. How could he have this terrible thing? It must have been her kidney. A fifties kidney. A DDT kidney. The Mother clears her throat. “Is it possible it was my kidney on the scan? I mean, I’ve never heard of a baby with a tumor, and, frankly, I was standing very close.” She would make the blood hers, the tumor hers; it would all be some treacherous, farcical mistake. “No, that’s not possible,” says the Surgeon.

The parents’ denial suggests the failure of imagination that accompanies an experience that none of us dare to imagine could happen to us. Hemon writes, “How can you possibly ease yourself into the death of your child? For one thing, it is supposed to happen well after your own dissolution into nothingness. Your children are supposed to outlive you by several decades, during the course of which they live their lives, happily devoid of the burden of your presence, and eventually complete the same mortal trajectory as their parents: oblivion, denial, fear, the end. They’re supposed to handle their own mortality, and no help in that regard (other than forcing them to confront death by dying) can come from you—​death ain’t a science project. And, even if you could imagine your child’s death, why would you?”

Profound Grief Vicki Foreman, writing about the death of her premature baby, describes what grief was like for her: I learned about grief during this time. I learned that no matter the true temperature, grief made the air crisp and cold; that it caused me to drive slowly, carefully; there was very little I could eat. The world receded. Everything took place in slow motion and was viewed as if down a very long telescope. So much was unfamiliar that if asked my name, I had to think for long moment. Grief is a visceral process. (Foreman 2009, 1).

Amy Kubelbeck (2011) wrote about what it was like to receive a prenatal diagnosis of hypoplastic left heart syndrome: “People often use physical terms to try to describe what it feels like to hear devastating news—​that it’s like being punched in the stomach, like being hit by a truck, or like the world is crashing in on them. To me it felt like falling backward, as

Dying Children and the Kindness of Strangers    59 though the tiled concrete floor, the clay underground, all the subterranean layers of rock were simply and soundlessly parting to let me through to some other dimension.” Eighth-​century Chinese poet Bai Juyi wrote of similar feelings in a way that illustrates how such feelings are timeless and universal. The sickness came, took only ten days, even though we’d raised you for three years. Miserable tears, crying voices, everything hurt painfully. Your old clothes lonely on the hanger, the medicine at your bedside. I sent you through the deep village lanes, I saw the tiny grave in the field. Don’t tell me it’s three li away—​ this separation is till the end of days. (Bai Juyi c. 8th century)

Uncertainty Manuela’s son is healthy and dies a sudden traumatic death. For Hemon and Moore, the diagnosis leads to choices about treatment. They have time. They are in a state that psychologist Barbara Sourkes calls “the living-​dying interval” (Sourkes 1982). She notes, “In the past, the illness trajectory moved directly from diagnosis to death, with little intervening time or space. This ‘new’ middle phase can unfold in many guises: a cycle of remissions and relapses, a gradual downhill course, or prolonged remission implying a cure … The challenge which faces the patient and family is to maintain a semblance of normal life in the ‘abnormal’ presence of life-​threatening illness.” This “living-​dying interval” is a time when prognosis is usually uncertain and various decisions must be made about how relentlessly to pursue treatment or whether to accept that death is inevitable and perhaps preferable. Doctors are often unable to help very much since they, too, are often uncertain about whether a child will live or die. They talk in probabilities. They talk of populations. But the probabilistic population statistics always allow room for both hope and despair. Clinical uncertainty amplifies moral uncertainty. It creates situations in which nobody can be sure that they are doing the right thing but in which they must do something. They must make a choice to either continue or discontinue treatment. There is no neutral default position. To err on the side of continuing treatment sometimes leads to a painful, futile, and costly prolongation of the dying process or, what may be worse, to a child’s survival with such severe neurological damage that life may not seem to be worth living. To err on the side of discontinuing treatment may lead to the death of a deeply beloved child who might have survived. It is a decision that no parent should have to make but one that we now force on parents every day.

60   John D. Lantos Adults, of course, face the same choices. But they can usually participate in the decisions. Or they will have expressed preferences or values in advance. Older children can do the same. For young children, we must try to imagine what they would want. In his book A Personal Matter, Nobel laureate Kenzaburo Oe tries to imagine what a critically ill baby might experience, “What did death mean to a baby like that? Or, for that matter, life? Only a few hours on this earth, spent in crying, tongue fluttering in his stretched pearly-​red mouth” (Oe 1972, 116). Oe concluded that the experience of such a baby was essentially unimaginable. Lorrie Moore tries to imagine what her child is feeling. It is almost unbearable for her. How can it be described? How can any of it be described? It is a horror and a miracle to see him. He is lying in his crib in his room, tubed up, splayed like a boy on a cross, his arms stiffened into cardboard “no-​no’s” so that he cannot yank out the tubes. There is the bladder catheter, the nasal-​gastric tube, and the Hickman, which, beneath the skin, is plugged into his jugular, then popped out his chest wall and capped with a long plastic cap. There is a large bandage taped over his abdomen. Groggy, on a morphine drip, still he is able to look at her when, maneuvering through all the vinyl wiring, she leans to hold him, and when she does, he begins to cry, but cry silently, without motion or noise. She has never seen a baby cry without motion or noise… . She would crawl up and lie beside him in the crib if she could. But instead, because of all his intricate wiring, she must lean and cuddle, sing to him, songs of peril and flight: “We gotta get out of this place, if it’s the last thing we ever do.” (Moore 1998, 237–​238)

In Moore’s account, the experience of the caregiver is one of pity, anger, sorrow beyond words, and an overwhelming helplessness. If she doesn’t give her permission for treatment, he will be dead. If she does, she is authorizing his suffering. Hemon also finds the experience of watching his child in the ICU to be almost unbearable. “In the I.C.U., we found her entangled in a web of I.V.s, tubes, and wires, paralyzed by Rocuronium (called ‘the rock’ by everyone there), which had been administered to prevent her from ripping out her breathing tubes. We spent the night watching her, kissing the fingers on her limp hand, reading or singing to her.” As these parents go through the experience of caring for a crucially ill child, they have no idea whether the decisions they are making are the right ones. Parents have so many more options than ever before. They can choose hospice and palliative care. They can choose mechanical ventilation, dialysis, and ECMO in the ICU. They can go to free-​standing, for-​profit chemotherapy centers in Houston or Phoenix for an unapproved cocktail of cancer chemotherapy or have their child’s heart augmented by a shiny “Berlin Heart.” More and more often, death seems almost optional, a choice rather than an inevitability. Ultimately, the rightness or wrongness of those decisions will be determined by the outcome. There may be redemption at the end of the long, dark tunnel. Or their may just be darkness. While they wait, their experiences are of uncertainty, helplessness, denial,

Dying Children and the Kindness of Strangers    61 fear of the future, and anger at the doctors and the brutal nature of medical care. And, ultimately, that profound sense of isolation.

Isolation Moore, Hemon, and Oe all talk about feeling utterly alone. Hemon calls his memoir “The Aquarium” because he felt, during the time when his daughter was sick, as if he was in an aquarium, looking out at all the people with “normal” lives. Moore writes, “In the end, you suffer alone. When your child has cancer, you are instantly whisked away to another planet” (Moore 1988, 60). Oe calls his novel A Personal Matter (Oe 1972, 68) and describes his experience as like being in a cave, deep underground. He says, “It’s entirely a personal matter. What I’m experiencing personally now is like digging a vertical mine shaft in isolation; it goes straight down to a hopeless depth and never opens on anybody else’s world. So I can sweat and suffer in that same dark cave and my personal experience won’t result in so much as a fragment of significance for anybody else.” Hemon writes about the difficulty of connection with friends, “We instinctively protected our friends from the knowledge we possessed; we let them think that words had failed, because we knew that they didn’t want to learn the vocabulary we used daily. We were sure that they didn’t want to know what we knew; we didn’t want to know it, either.”

Pediatric Palliative Care Most children’s hospitals now offer pediatric palliative care. The services that are offered differ from center to center. All, it seems, are designed to help parents deal with the denial, grief, uncertainty, and isolation. They also try to deal with the physical symptoms that accompany life-​threatening illnesses. The need for attention to palliation is clear. Wolfe and colleagues interviewed 103 parents of children who died of cancer between 1990 and 1997 at two Boston hospitals (Wolfe et al. 2000). According to the parents, 89% of these children experienced a lot or a great deal of suffering during the last month of life. But palliative care goes far beyond the treatment of pain. It focuses more on helping parents to understand their new role, helping them understand what it means to be the parent of a critically ill or dying child. Hinds and colleagues interviewed 62 parents of children with cancer and tried to understand how the parents coped with their changed role (Hinds et al. 2009). They found that parents defined their roles as protective—​they said that they needed to be there for their child, to always convey love, and to not allow the child to suffer. Many

62   John D. Lantos parents turned to their religious faith to help them through the last days and talked specifically about prayer and acceptance of what God sent their way. Renjilian and colleagues found similar responses (Renjilian et al. 2013). Parents said things like, “We will always choose the option that most improves her quality of life” and “We always said we would do anything to fight this disease.” Many said, “I don’t want my child to be in pain.” These empirical studies suggest ways that that well-​organized pediatric palliative care programs can help parents make sense of a situation that at first seems senseless. One problem with most of the studies, however, is that they are generally (and perhaps inevitably) retrospective studies. They are done after a child has died. This leads to two problems. First, they don’t capture the prognostic uncertainty that is an inevitable feature of the real world of decision-​making in which some children die and others do not. Palliative care is not just for children who die. It is for children who have complex and potentially lethal conditions. Parents whose children were critically ill but did not die might have different views of the situations that they faced than do the parents whose children died. For example, a common theme in studies of parents whose children died is that doctors gave them false hope or withheld information about how sick the child really was. By contrast, parents whose children were close to death but who survived are more likely to be angry at doctors for giving a bleak prognosis and taking away hope (French 2012; Miller 2011). In addition to this selection bias in the studies, there is another problem with the existing literature. The studies are inevitably retrospective ones that rely on parents’ recall of the events at the time of their child’s death. But, as Meert and colleagues note, “Parents’ perceptions and desires expressed after their child’s death may or may not relate to their actual needs at the time of the death. Parents are often angry after their child’s death; anger and blame directed at others may be a form of unresolved grief ” (Meert et al. 2008).

Small Good Things Are we doing a good job caring for critically ill and dying children? It is hard to know. The work is difficult for professionals in ways that are different from the ways in which it is difficult for parents. They, too, want to do what is best for the children. They, too, struggle with the prognostic uncertainties. They, too, try to understand whether continued treatment is going to be worth all the pain and suffering and to guide parents through the moral and spiritual and emotional mazes. Bedside nurses bear the brunt of the moral distress that sometimes accompanies these situations. More than doctors, they suffer physical symptoms that include headaches, abdominal pain, palpitations, and nausea. The moral distress of the professional caregivers is an indication and perhaps a pale reflection of the anxiety and depression

Dying Children and the Kindness of Strangers    63 that parents experience. There is no way to sugar-​coat the fundamental truth that caring for a critically child is emotionally devastating for everyone involved. But there are also moments of triumph, celebration, and exultation. There is good news mixed with the bad. Sometimes the good news comes in the form of a cure. Sometimes, it is accepting death and allowing the death to be a good one, a painless one, a death surrounded by loved ones. Children’s hospitals and pediatric palliative care services are changing the way we think about what is possible. Like organ procurement agencies, they sometimes find silver linings in the most tragic situations, the situations where bad things happen to good people. Parents talk about these silver linings. They talk about how they never would wish their experiences on anyone but how they also, sometimes, feel a sort of gratitude for parts of the experience. Perhaps the best story ever written about that sort of redemption is Raymond Carver’s short story, “A Small Good Thing.” The story is about the tragic death of an eight-​year-​old child, Scotty, who is hit by a car on his birthday as he is walking to school. For the next twenty-​four hours, Scotty is in the “living-​ dying interval” but, as is so often the case, nobody is quite sure whether he will recover or die. His parents try to get answers from the doctors, try to understand what is happening. Carver writes this about the father, “Until now, his life had gone smoothly. So far, he had kept away from any real harm, from those forces he knew existed and that could cripple or bring down a man if the luck went bad, if things suddenly turned. He pulled into the driveway and parked. His left leg began to tremble” (Carver 1986, 379). The mother starts to pray for the first time in many years: “I almost thought I’d forgotten how, but it came back to me. All I had to do was close my eyes and say, ‘Please God, help us—​help Scotty,’ and then the rest was easy. The words were right there” (384). The end of the story is the surprise. Before Scotty’s accident, they had ordered a birthday cake from a local baker. The baker has been calling them at all hours, demanding that they come and pay for the cake. After Scotty’s death, they go to see him. When he realizes what they have been going through and how his strident requests for payment made it all worse, he feels terrible. He apologizes profusely and offers them some fresh baked cinnamon rolls. Carver writes, He served them warm cinnamon rolls just out of the oven, the icing still runny. He put butter on the table and knives to spread the butter. Then the baker sat down at the table with them. He waited. He waited until they each took a roll from the platter and began to eat. “It’s good to eat something,” he said, watching them. “There’s more. Eat up. Eat all you want. There’s all the rolls in the world in here.” They ate rolls and drank coffee. Ann was suddenly hungry, and the rolls were warm and sweet. She ate three of them, which pleased the baker. “Smell this,” the baker said, breaking open a dark loaf. “It’s a heavy bread, but rich.” They smelled it, then he had them taste it. It had the taste of molasses and coarse grains. They listened to him. They ate what they could. They swallowed the dark bread. It was like daylight under the fluorescent trays of light. They talked on into the early morning, the high, pale cast of light in the windows, and they did not think of leaving.

64   John D. Lantos Carver’s story has all the elements of parental experience around the death of a child. The profound grief. The terrible isolation. It also, beautifully, portrays the opportunities for human connection that arise in these dark and dire situations. The goal for professionals who care for children and families facing life-​threatening illness must always be to offer kindness to strangers.

References “All About My Mother.” http://​www.sonyclassics.com/​allaboutmymother/​. Accessed December 6, 2015. Bai Juyi. c. 8th century. This parting is for all time. http://​www.consolatio.com/​2007/​09/​bai-​ juyi-​this-​p.html. Accessed November 19, 2013. Benham French, K. B. December 6, 2012. “Never let go.” Tampa Bay Times. http://​www.tampabay.com/​specials/​2012/​reports/​juniper. Accessed March 20, 2013. Carver, R. 1986 “A Small Good Thing.” http://​www.classicshorts.com/​stories/​sgthing.html Foreman, V. 2009. This lovely life. A memoir of premature motherhood. New York: Houghton Mifflin Harcourt. Hemon, A. “The Acquarium.” New Yorker Magazine, Jue 13, 2011. http://​www.newyorker.com/​ magazine/​2011/​06/​13/​the-​aquarium. Accessed December 6, 2015. Hinds, P. S., L. L. Oakes, J. Hicks, et al. 2009. “Trying to be a good parent as defined by interviews with parents who made phase I, terminal care, and resuscitation decisions for their children.” Journal of Clinical Oncology 27: 5979–​5985. Kubelbeck, A. 2011. “Waiting with Gabriel.” Current Problems in Pediatrics and Adolescent Health Care 41: 113–​114. Meert, K. L., S. Eggly, Pollack, M., et al. 2008. “Parents’ perspectives on physician-​parent communication near the time of a child’s death in the pediatric intensive care unit.” Pediatric Critical Care Medicine 9: 2–​7. Miller, T. 2011. “Shalebug.” Current Problems in Pediatrics and Adolescent Health Care 41: 121–​123. Moore, L. 1988. People like that are the only people here. In Birds of America. New Yorker Magazine, Jan 27, 1997. New York: Picador: New York. http://​www.newyorker.com/​magazine/​1997/​01/​27/​people-​like-​that-​are-​the-​only-​people-​here. Accessed December 6, 2015. Oe, K. 1972. A Personal Matter. New York: Grove. Carver, R. 1986 “A Small Good Thing.” http://​www.classicshorts.com/​stories/​sgthing.html Moore, L. 1998. People like that are the only people here. New Yorker Magazine, Jan 27, 1997. http://​www.newyorker.com/​magazine/​1997/​01/​27/​people-​like-​that-​are-​the-​only-​people-​ here. Accessed December 6, 2015. Renjilian, C. B., J. W. Womer, K. W. Carroll, T. I. Kang, and C. Feudtner. 2013. “Parental explicit heuristics in decision-​ making for children with life-​ threatening illnesses.” Pediatrics 131: e566–​e572. Sourkes, B. (1982). Armfuls of time. Pittsburgh, PA: University of Pittsburgh Press Wolfe, J., H. E. Grier, N. Klar, et al. 2000. “Symptoms and suffering at the end of life in children with cancer.” New England Journal of Medicine 342: 326–​333.

Chapter 5

Medical Fu t i l i t y and P otent ia l ly I nappropriate T re atme nt Douglas B. White and Thaddeus M. Pope

Introduction In this article we provide a historical, ethical, and conceptual review of medical futility disputes in the intensive care unit (ICU). We place particular emphasis on the role that physician power plays in these disputes. Specifically, we analyze the circumstances and arguments proposed to justify when physicians may stop life-​sustaining treatment without the consent of either the patient or surrogate. We begin the article by reviewing the history of the medical futility movement and the causes of medical futility disputes. Second, we summarize the major positions and policy statements addressing how such disputes should be resolved. Third, we turn from an objective, descriptive approach to a more normative approach. We highlight the value-​laden nature of most “futility” judgments. Finally, we outline how clinicians should respond to requests for ICU interventions that they deem medically or ethically inappropriate.

History of the Futility Debate Both the word “futility” and the concept of medical overtreatment have their origins in Ancient Greece (Amundsen, 1978). But the contemporary medical futility debate over the appropriateness of life-​sustaining treatment in the ICU did not begin until the late 1980s (Blackhall, 1987; Brody & Halevy, 1995; Lantos et al., 1988; Lantos et al., 1989; Murphy, 1988; Tomlinson & Brody, 1988; Truog et al., 1992; Youngner, 1988). A quarter century later, medical and broader societal consensus remains elusive.

66    Douglas B. White and Thaddeus M. Pope

Origins in Ancient Greece The word “futility” is derived from the Latin futilis, which means leaky (http://​www. oed.com/​view/​Entry/​75839?). The connection between “leaky” and the contemporary meaning “useless” and “failing utterly of the desired end” is based in the Greek myth of the fifty daughters of Danaus (Bernat, 2008: 226). On instructions from their father, all except one of the daughters killed their husbands. As punishment, they were condemned to spend eternity carrying water in a sieve or perforated device. This represented a “futile” act that could not produce the required result, a repetitive task that could never be completed. Not only the term “futility” but also the more precise concept “medical futility” or “clinical futility” dates to Ancient Greece. For example, Hippocrates warned, “Whenever the illness is too strong for the available remedies, the physician surely must not expect that it can be overcome by medicine … To attempt futile treatment is to display an ignorance that is allied to madness” (Schneiderman and Jecker, 2011: 6; Schneiderman and de Ridder, 2013). Accordingly, Hippocrates admonished, the physician must “refuse to treat those who are overmastered by their disease, realizing that in such cases medicine is powerless” (Hippocrates of Cos, 1959: 193). At around the same time, Plato similarly recognized, “For those whose lives are always in a state of inner sickness Asclepius [who was a legendary, indeed divine, physician] did not attempt to prescribe a regime to make their life a prolonged misery … A life with preoccupation with illness and neglect of work is not worth living” (Schneiderman and Jecker, 2011: 17; Schneiderman and de Ridder, 2013). And the playwright Sophocles observed, “For death is not the most odious thing; it is rather craving death, but lacking the means to die” (Sophocles, 1894: 1007). Over the next twenty-​four centuries, commentators continued to discuss medical futility (Bernat, 2008; Helft et al., 2000; Luce, 2010; Wilkinson & Savulescu, 2011). For example, in the 1930s Thomas Horder, physician to several British kings and prime ministers, argued, “It is the duty of a doctor to prolong life and it is not his duty to prolong the act of dying” (Voluntary Euthanasia [Legalisation] Bill, 1936). But Horder’s comments, like those of most commentators before 1988, were framed as an argument for the appropriate goals of medicine, not as an argument about the proper balance of power between physicians and patients/​surrogates.

Introduction of Life-​Sustaining Treatment The contemporary debate over medical futility began in the late 1980s, but its foundations were laid almost three decades earlier. The antecedent building blocks of the medical futility movement are two revolutionary developments in the 1960s:  (1)  the introduction of life-​ sustaining technology and (2)  the rise of patient autonomy. The 1960s were a time of remarkable medical innovation (Luce and White, 2009). Cardiopulmonary resuscitation (CPR) was invented in 1960. That same year, the

Medical Futility and Potentially Inappropriate Treatment    67 Scribner shunt was invented, permitting long-​term treatment of chronic renal failure. During the remainder of the decade, other life-​sustaining treatments were introduced, including both positive pressure mechanical ventilation and extracorporeal life support. As a result of this new technology, critical care grew into a specialty in its own right. By the end of the 1960s, most US hospitals had ICUs. But these developments did not lead to uniformly positive results. While life-​sustaining treatment permits the prolongation of the lives of many seriously ill individuals, it does not always offer realistic prospects for benefit. These “halfway” technologies can sustain biological life for practically indefinite periods of time but may not themselves lead to improvement or cure. Moreover, consistent with the technological imperative, the trend for all these life-​sustaining measures has been toward supporting sicker and more complex patients who would not have been considered appropriate candidates in the past. For example, CPR was not originally intended for patients with a terminal stage incurable disease. But such patients are now presumed to have consented to CPR unless they or their surrogates authorize a do not resuscitate (DNR) order. More recently, the indications for extracorporeal membrane oxygenation have also been expanding (Paden, 2014).

Rise of Patient Autonomy At the same time that life-​sustaining treatment was being introduced, major social changes were occurring. Most notably, the field of bioethics emerged in the 1960s’ context of civil rights and women’s rights. In contrast to the paternalism of the previous era, bioethics adopted a patient-​centered framework that defined medical ethics in terms of individual rights and patient autonomy. One important manifestation of this emphasis was the doctrine of informed consent (Canterbury v. Spence, 1972; Natanson v. Kline, 1960; Salgo v. Stanford, 1957). Soon it was recognized that a corollary of informed consent was the right to refuse treatment. During the 1970s and 1980s, appellate courts across the country decided numerous “right to die” cases in which patients and patients’ families wanted to withdraw or withhold life-​sustaining treatment but health-​care providers were reluctant to cede to such requests. These cases firmly established the right of patients to refuse life-​ sustaining treatment. These cases also established the right of surrogates to exercise this right for patients who were incapacitated and unable to exercise it for themselves. Today, patients and surrogates can decide whether proposed treatment is beneficial given their own values and particular circumstances (Pope, 2007).

From Negative Rights to Positive Rights By the early 1990s, most questions about the scope and extent of the patient’s negative right to refuse treatment had been answered. But new questions were just emerging (Helft et al., 2000). The new questions focused not on negative rights but on positive

68    Douglas B. White and Thaddeus M. Pope rights. Do patients or their surrogates have an affirmative right to demand and receive life-​sustaining treatment, even when their clinicians judge that treatment medically or ethically inappropriate? There are two distinctive features of a medical futility dispute. First, ICU patients dependent on life-​sustaining medical treatment almost never have decision-​making capacity. They lack the ability to understand the significant benefits, risks, and alternatives to proposed health care and to make and communicate a health-​care decision. They cannot direct their own medical treatment. Consequently, medical treatment decisions for ICU patients must be made by a surrogate. Second, the surrogate may request aggressive treatment interventions for an imminently dying or catastrophically chronically ill patient, but that patient’s clinicians may consider such treatment to be medically or ethically inappropriate. The clinician thinks that life-​sustaining measures are non-​beneficial and that the appropriate treatment plan is for comfort measures only. The surrogate, on the other hand, rejects this proposed treatment plan and directs the clinician to continue life-​sustaining measures.

Prevalence of Futility Disputes Today, medical futility disputes occur frequently in health-​care facilities across the United States and are one of the primary reasons for hospital ethics consultations. One major academic medical center estimates that 20% of the patients in its ICUs are receiving futile or probably futile treatment (Huynh et al., 2013; Huynh et al., 2014).

Causes of Futility Disputes Medical futility disputes typically result from surrogates requesting interventions that clinicians do not want to provide. To understand the causes of these conflicts, it is useful to examine both sides of the conflict: why surrogates insist and why clinicians resist. First, this section summarizes the main reasons that surrogates demand nonrecommended treatment. Second, it lists the main reasons that clinicians are unwilling to provide the desired treatment.

Reasons Surrogates Request Nonrecommended Treatment The reasons that surrogates request nonrecommended treatment can be grouped into four categories: (1) distrust, (2) cognitive issues, (3) psychological and emotional issues, and (4) religion and miracles.

Medical Futility and Potentially Inappropriate Treatment    69

Distrust Surrogates are aware of the limits of prognostication, sometimes because of a prior error in the patient’s treatment. Accordingly, they may doubt that things are actually as bad as physicians represent. Moreover, with greater access to medical information, surrogates are more confident in opposing health-​care providers. Distrust is especially prevalent among African Americans and Hispanics. There is substantial evidence that these patients and surrogates are more likely to request unconditional prolongation of life-​ sustaining treatment and are less likely to agree with a recommendation to withdraw or withhold. Furthermore, even nonminorities are increasingly distrustful, due to growing attention on providers’ economic incentives.

Cognitive Issues Unfortunately, surrogates frequently do not understand the clinical status of the patients they represent. Some studies have shown that less than half of surrogates, regardless of educational level, had adequate knowledge of what was going on and what would happen to the patient. Sometimes the surrogate may lack the capacity to make the relevant decisions. But even when the surrogate has the capacity, three key iatrogenic causes can lead to surrogate misunderstanding. First, providers often fail to clearly explain the patient’s condition and prognosis with clear, jargon-​free language. Second, providers may place undue pressure on the surrogate and fail to allow sufficient time to process information. Third, different specialists often supply the surrogate with uncoordinated, even conflicting information.

Psychological and Emotional Issues In addition to cognitive issues, many surrogates have clinically diagnosable conditions such as stress, depression, and anxiety. These psychological problems may impair the surrogate’s decision-​making capacity. But even if the surrogate has capacity, he or she may find it difficult to carry out his or her fiduciary duties because of loyalty, guilt, or uncertainty about the patient’s wishes or because of other family dynamics (Paris et al., 2010). Furthermore, some surrogates may have “dubious motives” in that they are looking out for their own interests rather than the patient’s.

Religion and Miracles While some requests for inappropriate treatment can be explained by the surrogate’s distrust or by cognitive or psychological impairment, the most intractable futility disputes tend to be those in which the surrogate’s decision is based on a value difference over “odds and ends”—​or the questions regarding what is a worthwhile chance or what is a worthwhile outcome. Religion is at the bottom of the most intractable futility disputes (Zier et al., 2009). For example, in the widely discussed Samuel Golubchuk case (Golubchuk v. Salvation Army Grace Hospital 2008), physicians determined that their patient had minimal brain function and that his chances for recovery were slim. Golubchuk’s insurmountable problem with wound infections required providers to

70    Douglas B. White and Thaddeus M. Pope repeatedly surgically hack away at his skin ulcers. Several physicians resigned because they thought this was tantamount to “torture.” Still, Golubchuk’s adult children argued that taking their father off life support would be a sin under their Orthodox Jewish faith (Jotkowitz, Glick & Zivotofsky, 2010). Even apart from religion, many surrogates request treatment with very low odds of success because they think that any chance is worth taking when the stakes are life and death. Indeed, a majority of the American public believes in miracles. They believe that even if doctors say futility had been reached, divine intervention by God could save their family member. Sometimes this belief in miracles is fueled by health-​care facilities’ own advertising (Vater et al., 2014).

Reasons Clinicians Resist Surrogate Requests Since the surrogate speaks for the patient, respect for autonomy means that providers should generally follow surrogate decisions. But medical futility concerns the limits of autonomy. The five main reasons that providers resist surrogate treatment requests are because they want to (1) avoid patient suffering, (2) respect patient autonomy, (3) protect the integrity of the medical profession, (4) avoid moral distress, and (5) promote good stewardship.

Avoid Patient Suffering Perhaps the most significant reason that providers resist surrogate requests is the desire to avoid being engaged in causing suffering. In the Golubchuk case, one physician described the surrogate-​requested treatment as “abomination,” “immoral,” and “tantamount to torture.” In another case from Boston, a staff member said, “This is the Massachusetts General Hospital, not Auschwitz.” Of course health-​care providers are willing to conduct procedures that cause pain and discomfort (e.g., chemotherapy), but they are willing to do this only because side effects are outweighed by benefits. In other words, providers are not really opposed to being complicit in causing patient suffering per se. They are opposed to participating in “unwarranted suffering.”

Respect Patient Autonomy Providers want to do what they think the patient would have wanted. They often doubt that the surrogate’s decision (to continue life-​sustaining treatment) accurately reflects the patient’s wishes, preferences, or best interest. Indeed, significant empirical evidence shows that surrogates frequently do not make the same treatment decision that the patient would have made for him or herself. Significant evidence shows that most patients prefer less aggressive medicine at the end of life.

Protect the Integrity of the Medical Profession Physicians want to defend the integrity of the medical profession. They do not want to be beholden to provide whatever treatment patients or surrogates want. Physicians are not “indentured servants,” “reflexive automatons,” or “vending machines.” The medical profession is largely a self-​governing one with its own standards of professional practice.

Medical Futility and Potentially Inappropriate Treatment    71

Avoid Moral Distress Medical futility is the leading cause of moral distress among nurses. Nurses “know the right thing to do,” but institutional constraints make it nearly impossible to pursue that course of action. Moral distress can drive people from the profession and thus reduce access. It can also reduce staffing levels and make people operate poorly, thus adversely impacting other patients’ quality of care.

Promote Good Stewardship Finally, providers want to be good stewards of health-​care resources. In most futility disputes they balance autonomy against beneficence or nonmaleficence. But sometimes, autonomy must be balanced against justice. While end-​of-​life costs have been a major subject of health policy debates, they have not played, and generally should not play, a role at the bedside. On the other hand, physicians do want to be careful with the allocation of “hard” resources like ICU beds, particularly in true triage situations.

Summary Identifying and understanding the causes of medical futility disputes is important for their successful collaborative resolution. In particular, with a better understanding of surrogates’ motivations and rationales for resisting provider recommendations to discontinue life support, the clinician can better develop empirically derived interventions. The clinician can incorporate targeted interventions into dispute resolution mechanisms and deploy them according to the precise basis for surrogate dissent. For example, if the basis for disagreement is prognostic distrust, it might be effective to offer an independent second opinion or a time-​limited trial of the disputed therapy.

Major Positions Writers and policymakers have articulated eight major positions on medical futility. The first four are definitional. Two are narrowly circumscribed by objective clinical criteria: (1) physiological futility and (2) medical ineffectiveness. Two other positions also purport to be neutral and scientific—​(3) quantitative futility and (4)  qualitative futility—​but they actually include value-​laden criteria. The goal of defining these terms was to justify the unilateral withholding or withdrawing of treatment even over surrogate objections, but finding consensus on these definitions proved problematic and elusive. Thus many institutions, professional associations, and commentators have turned from a definitional to (5)  a procedural, process-​based approach. Another position focuses on the appropriateness of the surrogate decision-​maker instead of the appropriateness of the treatment itself: (6) surrogate selection. Some approach medical futility

72    Douglas B. White and Thaddeus M. Pope as one type of (7) conscience-​based objection. Finally, some have moved the focus from the individual patient to the broader population: instead of framing the debate in terms of autonomy and beneficence, they frame it in terms of (8) distributive justice.

Definitional Approach: Physiological Futility The narrowest and most clearly defined type of “medically futility” is referred to as “physiological futility.” Physiologically futile interventions are considered inappropriate because they have a zero percent chance of being effective at achieving their intended goal. Commentators have offered a multitude of colorful examples, including prescribing antibiotics for a viral illness, performing CPR in the presence of rigor mortis, and offering chemotherapy for an ulcer. With physiological futility, the clinician does not make any evaluative assessment that the treatment’s effect is too unlikely, too small, or not worthwhile. There is no normative disagreement. Instead, clinicians can readily ascertain physiological futility based solely on their clinical knowledge. The basis for refusing treatment is a scientific and empirical one: the treatment simply will not work. This use of “futile” is true to the word’s core meaning. However, this objectivity comes at a steep price. Physiological futility has a very limited applicability. Decisions about withholding and withdrawing treatment are usually based on probabilities as opposed to certainties. Clinicians can rarely be certain that there is a 100% probability that a given intervention will have zero effect.

Definitional Approach: Medical Ineffectiveness Slightly broader in scope than physiological futility is “medical ineffectiveness.” This term is used in some state laws. In Maryland, for example, the Health Care Decisions Act contains a provision recognizing the authority of a physician to decline to provide “medically ineffective treatment” (Maryland Code Health–​General, 2015:  §5-​ 611[b]‌[1]). “Medically ineffective treatment” means that “to a reasonable degree of medical certainty, a medical procedure will not (1) prevent or reduce the deterioration of the health of an individual; or (2) prevent the impending death of an individual.” Take for example a patient dying of metastatic cancer who is diagnosed with an aortic stenosis. Replacing the valve is not physiologically futile, because the procedure can work from a cardiac perspective. But it does nothing to prevent the patient’s death hours or days later. Notably, an earlier version of the legislation defined “medically ineffective” treatment as that which would not “prevent or postpone for more than an insignificant length of time the death of an individual.” As enacted, the definition omits the verb “postpone.” The legislature decided to allow physicians to certify as “medically ineffective” interventions like CPR that might interrupt and delay death but would not materially alter the dying process. Some commentators defend a similar concept of “imminent demise futility.” The closeness and impending nature of

Medical Futility and Potentially Inappropriate Treatment    73 death is crucial. In contrast, the broader notion of “lethal condition futility” focuses on the impossibility of the patient’s long-​term survival with her underlying disease.

Definitional Approach: Quantitative Futility While few disputed interventions are physiologically futile, many more are probably ineffective. Accordingly, some commentators have proposed a quantitative standard for determining when life-​sustaining treatment is inappropriate. Proponents of quantitative futility note that clinical studies and scoring systems like APACHE (Acute Physiology and Chronic Health Evaluation) can provide enough information to provide an empirical basis for establishing percentage thresholds. Perhaps the most common quantitative futility standard holds that a treatment should be regarded as medically futile if it has not worked in the past 100 cases (Schneiderman et al., 1990; Jecker & Schneiderman, 1995; Schneiderman and Jecker, 2011). At first blush, quantitative futility seems ethically defensible. The clinician determines only whether the requested intervention can reasonably achieve the patient’s goals. The clinician does not necessarily challenge the patient’s goals themselves. However, quantitative futility suffers from three serious problems. First, it is unclear where the threshold percentage should be set. Should it be 2%, 1%, or even 0.1%? Setting the threshold of probability is a value judgment about which there is considerable variability (Rhodes, 1998; Wilkinson & Truog, 2013; Veatch, 1973; Veatch, 1977; Veatch & Spicer, 1992). Any threshold is arbitrary and determined by social policies, not by a clinician’s medical judgment or epidemiological skills (Prendergast, 1997; Veatch and Spicer, 1992). Second, even if health-​care providers and society could settle on a threshold percentage, available measures from population-​level studies are very imprecise (Gabbay et  al., 2010; Rubenfeld & Crawford, 1996). It is difficult for physicians to ascertain whether a given threshold standard applies to any particular patient. Third, morbidity and mortality rates are often based on self-​fulfilling prophecies.

Definitional Approach: Qualitative Futility With a quantitative standard, the clinician starts with the patient’s own goals and determines whether those goals are sufficiently achievable. In contrast, when applying a qualitative futility standard, the clinician questions whether the patient’s goals themselves are beneficial, meaningful, or worthwhile. Qualitative futility has two primary forms. The first form of qualitative futility focuses on balancing benefits and harms. It asserts that an intervention is medically inappropriate when it is “disproportionate treatment,” when its prospective benefits are outweighed by its associated burdens to the patient. For example, providing long-​term dialysis to a patient with severe dementia arguably prolongs agony with no likelihood of a good outcome. Similarly, performing CPR on a patient with metastatic cancer is arguably disproportionately burdensome. It cannot

74    Douglas B. White and Thaddeus M. Pope restore the patient but, by causing multiple rib fractures, would serve to increase the patient’s suffering. The second form of qualitative futility asserts that treatment is inappropriate when it simply cannot provide the patient a quality of life worth living (e.g. with permanent unconsciousness). The expected outcome is of little or no value in light of the patient’s extremely poor condition or prognosis. Providing the disputed treatment might not cause the patient to suffer, but neither does it offer the patient any reasonable benefit. Qualitative futility is heavily criticized. While an individual may make a personal choice to forgo treatment, it is a very different matter for a clinician to make that decision on the patient’s behalf (Burt, 2002; Rubin, 1998). Most problematic is the risk that a clinician might judge the patient’s quality of life to be far less than the patient would. For example, people with physical and cognitive disabilities obtain many satisfactions and rewards in their lives, but others may not adequately recognize those possible benefits of treatment. Schneiderman has proposed a clear definition of qualitatively futility in two situations: when a patient is (1) permanently unconscious and/​or (2) permanently totally dependent on intensive medical care (Schneiderman and Jecker, 2011; College of Physicians and Surgeons of Manitoba, 2007, Rural Physician Action Plan). This proposal is in itself controversial. Some Canadian authorities have referred to the former as the “minimum goal” of life-​sustaining treatment (College of Physicians and Surgeons of Manitoba, 2007.). When permanently unconscious, as in a PVS, patients have no thoughts, sensation, purposeful action, social interaction, awareness of self, or awareness of their environment. Therefore, the minimum goal does not comment on the quality of the patient’s experience; rather it states that the patient must simply experience his or her own existence. Several professional medical associations have similarly identified permanent unconsciousness as a contraindication for life-​sustaining measures (Renal Physicians Association, 2010). Schneiderman and Jecker (2011) argue that patients who are totally and permanently dependent on intensive care are in much the same position as those who are permanently unconscious. They are “preoccupied” with their illness to the point that they cannot meaningfully participate in the human community. However, many of these patients are interacting with their surroundings and often with their families. It is very hard to predict that the patient will be permanently dependenty on ICU care. Among other factors, many of these patients can actually be cared for in vent facilities when a bed is available.

Procedural Process-​Based Approach Lawyers, bioethicists, health-​care providers, and policymakers have had enormous difficulty defining treatment that is “futile” or “medically inappropriate.” Years of debate have failed to produce any consensus. By the late 1990s, many had abandoned delineating clinical indications that would “define” medical futility (Burns & Truog, 2007; Caplan, 2012). Paraphrasing Justice Potter Stewart’s comment on pornography, many concluded

Medical Futility and Potentially Inappropriate Treatment    75 that we can only “know it” when we “see it.” Instead of taking a definitional approach, many have turned to a procedural or process-​based approach (Moratti, 2009; Stewart, 2011). Important components of a fair process include a second opinion from a qualified physician, a case review by the hospital ethics committee, attempts to transfer the patient to another institution when such a transfer might be appropriate, and informing surrogates of their right to seek judicial intervention. Existing evidence suggests that most clinician–​surrogate disputes can be resolved through these steps, because most futility conflicts arise not from intractable value conflicts but from breakdowns in communication that are amenable to communication interventions (Bosslet et al., 2015). Several considerations justify a procedural approach to conflict resolution rather than giving all decision-​making authority either to surrogates or to individual clinicians. Giving sole authority to surrogates is problematic because, although it is generally accepted that patients/​surrogates should be allowed to choose from accepted, available treatment options, there is no positive right to interventions that are outside the boundaries of accepted practice. In addition, surrogates sometimes experience strong emotional and psychological barriers to authorizing decisions to forego life support, even when those decisions are clearly consistent with the patient’s values and preferences. Giving unilateral authority to surrogates may create a disincentive for them to genuinely consider clinicians’ perspectives and to move through the emotional challenges of foregoing treatment when doing so is consistent with a patient’s values and preferences. Giving sole authority to individual clinicians is problematic because there is well-​ documented variability between clinicians in their judgments about what is appropriate care in such cases, raising the concern of undue variability in treatment decisions (Wilkinson & Truog, 2013). Giving all authority to clinicians may also create a disincentive for clinicians to fully engage in the time-​consuming, challenging conversations often required to support surrogates and achieve mutually agreeable decisions. A process-​based approach to conflict resolution is recommended in part because the cases in question are ethically controversial, have important interests at stake, and exist in a gray zone of medical practice in which there are no explicit rules that define the boundaries of acceptable practice. It is ethically important to incorporate multiple perspectives to minimize the risk that the values of any one individual will carry undue weight. In addition, process-​based approaches better fulfill democratic ideals for resolving conflicts involving fundamental interests, such as transparency, legitimacy, accountability, and opportunity for appeal. Procedural fairness or “due process” is especially important because these cases generally involve patients who are vulnerable by virtue of incapacity, have little choice regarding their treating clinicians, and, because of their overwhelming illness, have severely limited ability to seek out other caregivers (White & Wicclair, 2012). Practically speaking, process-​based approaches may allow mutually agreeable solutions to emerge as the conflict-​resolution process unfolds over time. Within institutions, a process-​ based approach to conflict resolution may lessen arbitrariness by ensuring broader input, consistent evaluation of such requests, and the possibility for continuous institutional learning about how to manage future cases (Bosslet et al., 2015).

76    Douglas B. White and Thaddeus M. Pope

Surrogate Selection Approach A surrogate is an “extension of the patient” and stands in the shoes of the patient. Accordingly, the surrogate must make the choice that the patient, if competent, would have made for him or herself. The standards for surrogate decision-​making are substantially uniform across the country and generally involve a two-​step hierarchy. First, the surrogate should try to infer the patient’s wishes from his or her prior statements and conduct and make decisions that conform as closely as possible to what the patient would have done under the circumstances. Second, in the absence of reliable evidence of the patient’s expressed wishes, values, or preferences, the surrogate must rely on more objective grounds and shift the focus from the wishes of the patient to the welfare of the patient. Surrogates who deviate from these decision-​making standards, which they are supposed to employ, can be replaced (Pope, 2010a; Pope 2011; Pope, 2014). Clinicians sometimes take a surrogate selection approach to resolving medical futility disputes. Indeed, some jurisdictions have special tribunals, such as the Ontario Consent and Capacity Board Approach, which have adjudicated dozens of cases in which clinicians challenged surrogates whose requests for life-​sustaining treatment seemed inconsistent with a patient’s wishes or best interest (Chidwick 2013; Cuthbertson, 2013).

Conscience-​Based Objection Approach While most positions focus on the clinician’s professional objections to the requested treatment, one focuses on the clinician’s personal, conscience-​based objections. Almost every state has a health-​care conscience clause. These laws permit clinicians to decline the provision of services that violate their religious or moral beliefs. Because some clinicians equate the administration of “futile” treatment with torture and inhumanity, they may make conscience-​based refusals pursuant to such laws. But conscience clauses are usually materially limited to balance clinician rights against patient needs. Accordingly, most conscience clauses require “treat ’til transfer.” That is, they condition the provider’s right to refuse on transferring the patient to another provider who is willing to comply with the patient’s or surrogate’s treatment request. In most intractable futility disputes, transfer is not possible. Therefore clinicians cannot effectively implement their right to conscientiously object and refuse treatment.

Distributive Justice Approach Most major positions on medical futility focus on the principles of autonomy and beneficence applied to the individual patient, but increasingly, commentators are defending a justice-​based approach (Niederman & Burger, 2010; Huynh et al., 2013; Huynh et al., 2014; Savulescu, 2013). ICU resources are limited and require careful allocation. When they are provided to patients with little capacity to benefit, this creates an opportunity

Medical Futility and Potentially Inappropriate Treatment    77 cost for other patients who have a comparatively greater capacity to benefit. For example, critically ill patients boarded in the emergency department or outside referring hospitals have higher ICU and hospital mortality. But calculating negative externalities in the ICU is complicated (Truog & White, 2013).

Major Policy Statements Dozens of medical associations and other bodies have issued policy statements on medical futility. Seven of the most significant, in chronological order, are those from (1) the American Thoracic Society (ATS), (2) Society of Critical Care Medicine (SCCM), (3) American Medical Association (AMA), (4) Texas Advance Directives Act, (5) General Medical Council, (6) American College of Physicians, and (7) a multisociety policy statement from five major critical care associations.

American Thoracic Society (1991) In 1991 the ATS took a definitional approach. It declared that the prolongation of biological life is not itself a legitimate goal of ICU care. There must be a reasonable likelihood of “meaningful survival.” The ATS specifically clarified that surviving with permanent unconsciousness is not a meaningful survival.

Society of Critical Care Medicine (1997) In contrast to the ATS policy, the SCCM (1997) criticizes definitional approaches as “insensitive” to the values of individual patients. Apart from physiological futility, the SCCM encourages clinicians to explore and address conflicts. It concludes that a process-​based approach better assures “respect and consideration” of diverse perspectives.

American Medical Association (1999) Like the SCCM, the AMA (1999) also concluded that it was time to abandon definitional approaches. The AMA specified a seven-​stage process. Four steps are aimed at deliberation and negotiation, including the engagement of consultants such as social workers, chaplains, palliative care, and the ethics committee. Two steps are aimed at securing alternatives if conflict remains intractable: transfer to another physician or to another facility. Only after completing the previous six steps may a clinician proceed to unilaterally withhold or withdraw treatment.

78    Douglas B. White and Thaddeus M. Pope

Texas Advance Directives Act (1999) In 1999 Texas enacted a statute that follows the AMA approach. If a clinician wants legal immunity for refusing to honor a surrogate’s request for continued life-​sustaining medical treatment, the clinician must commence a multistage review process (Texas Directives Act, 1999). Life-​sustaining treatment must be provided during this review process. The first stage entails an ethics committee review of the attending physician’s determination. The surrogate must be notified of the ethics committee review process at least forty-​eight hours before the committee meets. The surrogate is also entitled to attend the meeting and to receive a written explanation of the committee’s decision. If the ethics committee agrees with the treating physician that treatment is inappropriate, the provider must attempt to transfer the patient to another provider who is willing to comply with the surrogate’s treatment request. The current provider is obligated to continue providing treatment for ten days after the surrogate is given the ethics committee’s written decision. If the patient has not been transferred or granted an extension, then the provider may unilaterally stop treatment on the eleventh day (Fine, 2000; Fine, 2009a; Fine, 2009b; Fine & Mayo, 2003; Halevy & Brody, 1996; O’Callaghan, 2008; Truog, 2007; Truog, 2009a; Truog, 2009b; Zientek, 2005).

General Medical Council (2010) The General Medical Council is the physician licensure and registration authority in the United Kingdom (General Medical Council, 2010). Its guidance provides that physicians need not provide treatment if they believe it would not be “clinically appropriate.” The UK courts have confirmed that neither patients/​surrogates nor the courts can require clinicians to provide treatment contrary to their clinical judgment, but a clinician’s decision to withhold or withdraw treatment must be transparent. He or she must tell surrogates the reasons for this judgment and their options, including seeking a second opinion and statutory review (Tracey v. Cambridge University Hospitals NHS Foundation Trust 2014.). Other European authorities have similarly afforded greater discretion to clinicians in medical futility disputes, not only permitting but mandating them to avoid “therapeutic obstinacy” (Lambert, 2015).

American College of Physicians (2012) The American College of Physicians is one of the largest physician groups in the United States. Its 2012 code of ethics confirms that physicians need not provide physiologically futile treatment, and it approves unilateral DNR orders where CPR would be medically ineffective (Snyder, 2012). But for the more common situation in which there is “some small prospect of benefit,” there is “no easy, automatic solution.” The American College of Physicians encourages a process-​based approach.

Medical Futility and Potentially Inappropriate Treatment    79

ATS/​AACN/​ACCP/​ESICM/​SCCM (2015) In June 2015, the American Thoracic Society issued a new policy statement supported by the American Association of Critical Care Nurses, the American College of Chest Physicians, the European Society of Intensive Care, and the Society of Critical Care Medicine. The multisociety statement on responding to requests for potentially inappropriate treatments in intensive care units provides guidance for clinicians to prevent and manage disputes in patients with advanced critical illness. The recommendations immediately below closely track this policy statement.

How Clinicians Should Respond We provide justifications for taking a procedural approach above, in the section on Major Positions. Here, we recommend and outline a specific type of procedural approach. Hospitals should develop and adopt conflict resolution processes that contain the nine characteristics detailed below, which are broadly informed by the conflict resolution literature, philosophical conceptions of procedural justice, and professional society guidelines (Bosslet et al., 2015). At bottom, the process is intended to help determine whether the surrogate-​requested intervention falls outside the boundaries of good medical practice according to the best judgment of those involved.

Implement Preventative Strategies Institutions should implement strategies to prevent intractable treatment conflicts, including proactive communication and early involvement of expert consultation. When responding to requests for futile interventions, clinicians should seek to understand the reasons for such requests, empathically correct misperceptions, provide emotional support, and explain why the requested interventions will not be provided. If disagreement persists, clinicians should generally obtain expert consultation to assist with conflict resolution. Clinicians should consider seeking expert consultation to provide intensive psychosocial support to the surrogate.

Refuse Requests for Physiologically Futile, Proscribed, and Discretionary Treatments “Proscribed treatments” are those that are prohibited by applicable laws, judicial precedent, or widely accepted public policies (e.g., organ allocation strategies) (Bosslet et al., 2015). “Discretionary treatments” are those for which there are laws, judicial precedent,

80    Douglas B. White and Thaddeus M. Pope or policies that give physicians permission to refuse to administer them (e.g., laws pertaining to “medically ineffective treatment”) (Bosslet et al., 2015). In responding to requests for either proscribed or discretionary treatments, clinicians should carefully explain the rationale for treatment refusal and, if there is uncertainty regarding the interpretation and application of the relevant rule, should generally seek expert consultation to confirm the rule’s correct interpretation. Clinicians should not provide physiologically futile interventions because they clearly fall outside the scope of good medical practice. Nor should clinicians provide proscribed or discretionary treatments because when society has legitimately developed rules to govern controversial aspects of medical practice, clinicians are justified in acting according to those rules as part of their professional role (Bosslet et al., 2015). Other than in these narrow circumstances, in general terms, clinicians should conceptualize their judgments that requested treatments are inappropriate as preliminary claims in need of confirmation, rather than conclusions to be immediately acted upon. Similarly, clinicians should conceptualize the requested interventions that are the subject of such claims as “potentially inappropriate treatment” to emphasize the tentative nature of the inappropriateness judgments.

Attempt Negotiated Agreement If conflict persists, clinicians should redouble efforts to reach a negotiated agreement with surrogates. Clinicians should generally seek the assistance of consultants skilled in mediation and conflict resolution. These consultants should be separate from the hospital review committee that evaluates whether the requested treatments are inappropriate. The consultant should ensure frequent, skillful communication between parties, foster negotiation, and provide psychosocial support to the clinicians and surrogates. Both clinicians and surrogates should be encouraged to reconsider their positions as new information becomes available (Pope & Waldman, 2007; Spielman, 1994).

Give Notice of the Process to Surrogates Surrogates should be informed in writing and verbally about the procedural conflict-​ resolution mechanism and invited to participate in the process.

Obtain a Second Medical Opinion Clinicians should obtain a second opinion from another independent clinician with expertise in the patient’s condition, addressing both the patient’s prognosis and the judgment that requested treatment is inappropriate.

Medical Futility and Potentially Inappropriate Treatment    81

Provide Review by an Interdisciplinary Hospital Committee If disagreement persists, the case should be evaluated by an interdisciplinary institutional committee whose members are not directly involved with the patient’s care and who do not otherwise have a conflict of interest regarding the case. The committee should be interdisciplinary in nature with community representation if possible, in accord with existing recommendations regarding the composition and competencies of hospital ethics committees. The committee should be able to convene and proceed with case review in a timely fashion.

Offer Surrogates the Opportunity for Transfer If the hospital committee agrees with the clinicians’ judgment that the requested treatments are inappropriate but the surrogate remains unpersuaded, surrogates should be informed of their right to seek transfer to another institution. Clinicians and the institution should offer to assist surrogates in seeking an alternate provider because doing so requires expertise that patients and families generally do not have. This includes identifying and contacting alternative providers, explaining the clinical situation to other clinicians, and assisting with the logistics of such a transfer process.

Inform Surrogate of Their Opportunity to Pursue Extramural Appeal Whenever possible, disputes should be resolved by the aforementioned measures. But if they prove ineffective, surrogates should be informed of their right to an independent appeals mechanism, generally through judicial review. If the internal dispute resolution process has affirmed the clinician’s judgment, then there is a presumption that the requested treatment is not appropriate. Consequently, it is the surrogate’s burden to rebut this presumption rather than the clinician’s burden to affirm it.

Implement the Decision of the Resolution Process If the hospital committee or appellate body (if consulted) agrees with the surrogate’s request for life-​prolonging treatment, then clinicians should either provide such treatments or transfer the patient to a willing provider. On the other hand, if the hospital committee affirms the treating clinicians’ judgment, no alternate providers can be found, and the independent appeal mechanism is either not undertaken by the surrogate or affirms the clinicians’ position, then clinicians may refuse to provide or withdraw

82    Douglas B. White and Thaddeus M. Pope the disputed treatments. A care plan should be developed that includes provision of other indicated treatments, including those focused on achieving patient comfort.

Role of Judicial or Other Independent Review The legitimacy of decisions that arise from a purely procedural conflict-​resolution process, like that outlined above, hinges on adherence to principles of fair process, such as diverse representation of views, transparency, legitimacy, freedom from conflict of interest, and independence. But intramural resolution processes have been criticized as not adequately achieving these standards. There are concerns about variable expertise, financial or relational conflicts of interest, and inadequate community representation (Pope, 2009; Pope, 2010b; Pope, 2015). Others have questioned whether seeking an alternate provider at other institutions will reliably shed light on community norms. In practice, a clinician’s refusal to accept a patient in transfer may result from financial considerations or a desire to not become embroiled in another hospital’s controversial case, rather than the belief that the requested treatments are inappropriate. In the United States, the courts generally fulfill the appellate function for conflict resolution. Although there are important strengths of the court as the appeals mechanism, there are also serious limitations, such as the time-​consuming and adversarial nature of court proceedings (Curtis & Burt, 2010; White & Pope, 2012). The medical profession must partner with policymakers to develop and evaluate novel extrajudicial appeals mechanisms, such as regional ethics committees and quasi-​judicial bodies to resolve conflicts when surrogates wish to challenge the decision rendered by the intramural resolution process about boundaries of accepted practice.

Conclusion Significant controversy about how to manage requests by patients or surrogates for treatments that clinicians believe should not be administered remains. A fair process of conflict resolution can settle most such disputes, but this is not a complete and permanent solution. The medical profession must partner with policymakers and social scientists to conduct public engagement efforts to develop substantive laws/​policies about when the use of life-​prolonging technologies should be not be used. Developing clear societal rules about the appropriate boundaries of medical practice near the end of life would foster transparency in limit setting and may allow more efficient resolution of individual cases. To be clinically useful, such rules will require a high level of detail and specificity about which treatments are proscribed under which circumstances. To

Medical Futility and Potentially Inappropriate Treatment    83 be ethically acceptable in pluralistic societies, the development of such rules will require the input diverse of clinician, patient, and stakeholder groups.

References American Medical Association. 1999. Medical futility in end-​of-​life care: Report of the Council on Ethical and Judicial Affairs. JAMA 281(10): 937–​941. American Thoracic Society. 1991. Withholding and withdrawing life-​sustaining therapy. American Review of Respiratory Diseases 144(3): 726–​731. Amundsen, Darrel W. 1978. The physician’s obligation to prolong life: A medical duty without classical roots. Hastings Center Report 8: 23–​30. Bernat, James L. 2008. Ethical issues in neurology. Philadelphia: Lippincott Williams. Blackhall, Leslie D. 1987. Must we always use CPR? New England Journal of Medicine 317(20): 1281–​1285. Bosslet, Gabriel T. et  al. 2015. An official ATS/​AACN/​ACCP/​ESICM/​SCCM policy statement:  responding to requests for potentially inappropriate treatments in intensive care units. American Journal of Respiratory and Critical Care Medicine 191(11): 1318–​1330. Brody, Baruch A., and Amir Halevy. 1995. Is futility a futile concept? Journal of Medicine and Philosophy 20(2): 123–​144. Burns, J. P., and Robert D. Truog. 2007. Futility: A concept in evolution. Chest 132(6): 1987–​1993. Burt, Robert A. 2002. The medical futility debate: Patient choice, physician obligation, and end-​of-​life care. Journal of Palliative Medicine 5(2): 249–​254.

Canterbury v. Spence. 1972. 464 F.2d 772 (D.C. Circuit).

Caplan, Arthur L. 2012. Little hope for medical futility. Mayo Clinic Proceedings 87(11): 1040–​1041. Chidwick Paula. 2013. Best interests at end of life: an updated review of decisions made by the Consent and Capacity Board of Ontario. Journal of Critical Care 28(1): 22–​27. College of Physicians and Surgeons of Manitoba. 2007. Statement No. 1602: Withholding and Withdrawing Life-​Sustaining Treatment. Curtis J. R., and Robert A. Burt. 2010. Futility in the intensive care unit: Hard cases make bad law. Critical Care Medicine 38(8): 1742–​1743.

Cuthbertson v. Rasouli. 2013. 2013 SCC 53; 3 S.C.R. 341.

Ethics Committee of the Society of Critical Care Medicine. 1997. Consensus statement of the Society of Critical Care Medicine’s Ethics Committee regarding futile and other possibly inadvisable treatments. Critical Care Medicine 25(5): 887–​891. Fine, Robert L. 2000. Medical futility and the Texas Advance Directives Act of 1999. Proceedings of the Baylor University Medical Center 13(2): 144–​147. Fine, Robert L. 2009a. Point:  The Texas Advance Directives Act effectively and ethically resolves disputes about medical futility. Chest 136(4): 963–​967. Fine, Robert L. 2009b. Rebuttal from Dr. Fine. Chest 136(4): 971–​972. Fine, Robert L., and Tom W. Mayo. 2003. Resolution of futility by due process: Early experience with the Texas Advance Directives Act. Annals of Internal Medicine 138(9): 743–​746. Gabbay, E., J. Calvo-​Broce, K. B. Meyer, T. A. Trikalinos, J. Cohen, and D. M. Kent. 2010. The empirical basis for determinations of medical futility. Journal of General Internal Medicine 25(1): 1083–​1089.

84    Douglas B. White and Thaddeus M. Pope General Medical Council. 2010. Treatment and care towards the end of life: Good practice in decision making. London: General Medical Council.

Golubchuk v. Salvation Army Grace Hospital. 2008. MBQB 49.

Halevy, Amir, and Baruch A. Brody. 1996. A multi-​institution collaborative policy on medical futility. Journal of the American Medical Association 276(7): 571–​574. Helft, Paul R., Mark Siegler, and John Lantos. 2000. The rise and fall of the futility movement. New England Journal of Medicine 343(4): 293–​296. Hippocrates of Cos. 1959. The art. In W. H. S. Jones, trans. Hippocrates. Vol. 2. Cambridge, MA: Harvard University Press. Huynh T.N. et al. 2013. The frequency and cost of treatment perceived to be futile in critical care. JAMA Internal Medicine 173(20): 1887–​1894. Huynh T.N. et  al. 2014. The opportunity cost of futile treatment in the ICU. Critical Care Medicine 42(9): 1977–​1982. Jecker, Nancy S., and Lawrence J. Schneiderman. 1995. Judging medical futility:  An ethical analysis of medical power and responsibility. Cambridge Quarterly of Healthcare Ethics 4(1): 23–​35. Jotkowitz, A., S. Glick, and A. Z. Zivotofsky. 2010. The case of Samuel Golubchuk and the right to live. American Journal of Bioethics 10(3): 50–​53.

Lambert v. France. 2015. European Court of Human Rights Application no. 46043/​14.

Lantos, John D., Steven H. Miles, M. D. Silverstein, et al. 1988. Survival after cardiopulmonary resuscitation in babies of very low birthweight: Is CPR futile? New England Journal of Medicine 318(2): 91–​95. Lantos, John D., P. A. Singer, R. M. Walker, G. P. Gramelspacher, G. R. Shapiro, M. A. Sanchez-​ Gonzalez, C. B. Stocking, S. H. Miles, and M. Siegler. 1989. The illusion of futility in clinical practice. American Journal of Medicine 87(1): 81–​84. Luce, John M. 2010. A history of resolving conflicts over end-​of-​life care in intensive care units in the United States. Critical Care Medicine 38(8): 1623–​1629. Luce, John M., and Douglas B. White. 2009. A history of ethics and law in the intensive care unit. Critical Care Clinics 25(1): 221–​237. Maryland Code. 2015. “Health–​General” § 5-​611. Moratti, Sofia. 2009. The development of “medical futility”: Towards a procedural approach based on the role of the medical profession. Journal of Medical Ethics 35(6): 369–​372. Murphy, D.J. 1988. Do-​not-​resuscitate orders. Time for reappraisal in long-​term-​care institutions. Journal of the American Medical Association 260(14): 2098–​2101.

Natanson v. Kline. 1960. 186 Kan. 393, 350 P.2d 1093.

Niederman, Michael S., and Jeffrey T. Berger. 2010. The delivery of futile care is harmful to other patients. Critical Care Medicine 38(Suppl.): S518–​S522. O’Callaghan, Nora. 2008. When Atlas shrugs: May the state wash its hands of those in need of life-​sustaining medical treatment? Health Matrix 18(2): 291–​372. Paden, M.L. 2014. Update and outcomes in extracorporeal life support. Seminars in Perinatology 38(2): 65–​70. Paris, John J., P. Angelos, and M. D. Schreiber. 2010. Does compassion for a family justify providing futile CPR? Journal of Perinatology 30(12): 770–​772. Pope, Thaddeus M. 2007. Medical futility statutes: No Safe harbor to unilaterally refuse life-​ sustaining treatment. Tennessee Law Review 71(1): 1–​81. Pope, Thaddeus M. 2009. Multi-​institutional healthcare ethics committees: the procedurally fair internal dispute resolution mechanism, Campbell Law Review 31(2): 257–​331.

Medical Futility and Potentially Inappropriate Treatment    85 Pope, Thaddeus M. 2010a. Surrogate selection: An increasingly viable, but limited, solution to intractable futility disputes. Saint Louis University Journal of Health Law and Policy 3(2): 183–​252. Pope, Thaddeus M. 2010b. The case of Samuel Golubchuk: The dangers of judicial deference and medical self-​regulation. America Journal of Bioethics 10(3): 59–​61. Pope, Thaddeus M. 2011. Legal briefing: Futile or non-​beneficial treatment. Journal of Clinical Ethics 22(3): 277–​296. Pope, Thaddeus M. 2014. Dispute resolution mechanisms for intractable medical futility disputes. New York Law School Law Review 58(2): 347–​368. Pope, Thaddeus M. 2015. The Texas Advance Directives Act: must a death panel be a star chamber? American Journal of Bioethics 15(8): 41–​43. Pope, Thaddeus M., and Ellen A. Waldman. 2007. Mediation at the end of life: Getting beyond the limits of the talking cure. Ohio State Journal on Dispute Resolution 23(1): 143–​194. Prendergast, Thomas J. 1997. Resolving conflicts surrounding end-​of-​life care. New Horizons 5(1): 62–​7 1. Renal Physicians Association. 2010. Shared decision making in the appropriate initiation of and withdrawal from dialysis. 2nd ed. Rockville, MD: Renal Physicians Association. Rhodes, Rosamond. 1998. Futility and the goals of medicine. Journal of Clinical Ethics 9(2): 194–​205. Rubenfeld, Gordon D., and S. W. Crawford. 1996. Withdrawing life support from mechanically ventilated recipients of bone marrow transplants: A case for evidence-​based guidelines. Annals of Internal Medicine 125(8): 625–​633. Rubin, Susan B. 1998. When doctors say no: The battleground of medical futility. Bloomington:  Indiana University Press.

Salgo v. Leland Stanford Board of Trustees. 1957. 317 P.2d 170, 154 Cal. App. 2d 560.

Savulescu, Julian. 2013. Just dying: The futility of futility. Journal of Medical Ethics 39(9): 583–​584. Schneiderman, Lawrence J., and Michael de Ridder. 2013. Medical futility. In James L. Bernat and H. Richard Beresford, eds. Ethical and legal issues in neurology. Handbook of Clinical Neurology 118. Edinburgh, England: Elsevier, 167–​179. Schneiderman, Lawrence J., and Nancy S. Jecker. 2011. Wrong medicine: Doctors, patients, and futile treatment. 2nd ed. Baltimore: Johns Hopkins University Press. Schneiderman, Lawrence J., Nancy S. Jecker, and A. R. Jonsen. 1990. Medical futility: Its meaning and ethical implications. Annals of Internal Medicine 112(12): 949–​954. Snyder, Lois. 2012. American College of Physicians Ethics Manual Sixth Edition. Annals of Internal Medicine 156: 73–​104. Sophocles. 1894. Electra. Richard C. Jebb, ed. Cambridge: Cambridge University Press. Spielman, Bethany. 1994. Collective decisions about medical futility. Journal of Law, Medicine and Ethics 22: 152–​160. Stewart, Cameron. 2011. Futility determination as a process:  Problems with medical sovereignty, legal issues and the strengths and weakness of the procedural approach. Journal of Bioethical Inquiry 8(2): 155–​163. Texas Advance Directives Act. 1999. Texas Health and Safety Code §166.046. Tomlinson, Tom and H. Brody 1988 Ethics and communication in do-​not-​resuscitate orders. New England Journal of Medicine 318(1): 43–​46.

Tracey v. Cambridge University Hospitals NHS Foundation Trust. 2014. EWCA Civ 822.

Truog, Robert D. 2007. Tackling medical futility in Texas. New England Journal of Medicine 357(1): 1–​3.

86    Douglas B. White and Thaddeus M. Pope Truog, Robert D. 2009a. Counterpoint:  The Texas Advance Directives Act is ethically flawed: Medical futility disputes must be resolved by a fair process. Chest 136(4): 968–​971. Truog, Robert D. 2009b. Rebuttal from Dr. Truog. Chest 136(4): 972–​973. Truog, Robert D., A. S. Brett, and J. Frader. 1992. The problem with futility. New England Journal of Medicine 326(23): 1560–​1564. Truog, Robert D., and Douglas B. White. 2013. Futile treatments in intensive care units. JAMA Internal Medicine 173(2): 1894–​1895. Vater, L. B., J. M. Donohue, R. Arnold, D. B. White, E. Chu, and Y. Schenker. 2014. What are cancer centers advertising to the public? A content analysis. Annals of Internal Medicine 160(12): 813–​820. Veatch, Robert M. 1973. Generalization of expertise. Hastings Center Report 1: 29–​40. Veatch, Robert M. 1977. The technical criteria fallacy. Hastings Center Report 7: 15–​16. Veatch, Robert M., and Carol M. Spicer. 1992. Medically futile care: The role of the physician in setting limits. American Journal of Law and Medicine 18(1): 15–​36. White, Douglas B., and Thaddeus M. Pope. 2012. The courts, futility, and the ends of medicine. JAMA 307(2): 151–​152. White, Douglas B., and Mark Wicclair. 2012. Limits on clinicians’ discretion to unilaterally refuse treatment. American Journal of Critical Care 21(5): 361–​364. Wilkinson, Dominic J. C., and Julian Savulescu. 2011. Knowing when to stop: Futility in the ICU. Current Opinion in Anaesthesiology 24(2): 160–​165. Wilkinson, Dominic J. C., and Robert D. Truog. 2013. The luck of the draw: Physician-​related variability in end-​ of-​ life decision-​ making in intensive care. Intensive Care Medicine 39(6): 1128–​1132. Voluntary Euthanasia (Legalisation) Bill [H.L.]. 1936. HL Deb 01 December, Vol. 103 cc465–​505. http://​hansard.millbanksystems.com/​lords/​1936/​dec/​01/​voluntary-​euthanasia-​legalisation​bill-​hl#S5LV0103P0_​19361201_​HOL_​21 Youngner, Stuart J. 1988. Who defines futility? JAMA 260(14): 2094–​2095. Zientek, D. M. 2005. The Texas Advance Directives Act of 1999: An exercise in futility? HEC Forum 17(4): 245–​259. Zier, L. S., J. H. Burack, G. Micco, et al. 2009. Surrogate decision makers’ responses to physicians’ predictions of medical futility. Chest 136(1): 110–​117.

Chapter 6

C onscientious Obj e c t i on Mark R. Wicclair

Introduction Until early in the last half of the 20th century, conscientious objection was largely a phenomenon associated with military service. Two U.S. court decisions, one in 1972 and the other in 1973, contributed significantly to expanding its scope into health care. In the earlier case, a U.S. District Court in Billings, Montana issued a preliminary injunction enjoining a Catholic hospital from denying a physician permission to perform a sterilization procedure within the hospital (369 F.Supp. 948). The court’s opinion was based in part on the hospital’s acceptance of federal funds. The second case was the landmark U.S. Supreme Court Roe v. Wade decision in 1973 holding that women have a constitutional right to abortion (410 U.S. 113 (1973)). Despite the court’s ruling and the resulting change in the legal status of abortion, many physicians continued to object to it on moral or religious grounds. Later in 1973, the U.S. Congress enacted the Church Amendment (42 U.S.C. § 300a-​7(a–​b)), which provided the first federal legal protection specifically for health professionals and institutions who object to abortion on moral or religious grounds. It stated that the receipt of government funding “does not authorize any court or any public official or other public authority” to require (a) clinicians to perform or assist in the performance of abortions or sterilizations or (b) institutions to make their facilities available or provide personnel for abortions or sterilizations. The language of the Church Amendment, as well as its inclusion of sterilization, is attributable in part to the earlier Montana District Court decision. Another factor that contributed significantly to expanding the scope of conscientious objection into health care is more closely related to the subject of this handbook. During the last half of the 20th century, there was a substantial increase in the ability to prolong the lives of patients. Advances in life-​sustaining measures included dialysis, CPR, mechanical ventilation, and medically provided nutrition and hydration. Some physicians believed that if it is medically possible to prolong a patient’s life, it is imperative to do so. As a result, patients and family members

88   Mark R. Wicclair sometimes were not given an opportunity to decide whether or not they wanted life-​sustaining measures. Over time, however, a consensus developed that patients with decision-​making capacity have an ethical and legal right to accept or refuse life-​sustaining measures (Meisel 1992; Meisel and Cerminara 2009). Nevertheless, some physicians could not in good conscience honor patient or surrogate requests to forgo life-​sustaining measures, and several leading professional codes and policy guidelines recommended against requiring such physicians to act against their consciences (American Thoracic Society 1991; Guidelines on the Termination of Life-​ Sustaining Treatment and the Care of the Dying 1987; President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1983). Typically, state advance directive laws also provided some accommodation for health professionals who could not in good conscience forgo life-​sustaining treatment (Meisel and Cerminara 2009). The scope of conscientious objection in health care generally as well as in the specific area of end-​of-​life care has broadened substantially. Conscientious objection in reproductive health has expanded beyond abortion and sterilization to include contraception and fertility services. Further expanding the scope of conscientious objection, some health professionals engaged in research and genetic screening have refused to participate in activities that are against their consciences. In the area of end-​of-​life care, beyond objections to forgoing life-​sustaining measures, some health professionals have conscience-​based objections to participation in donation after circulatory determination of death (DCDD), declaring death using the whole brain (neurological) criterion, and offering palliative sedation to unconsciousness.

What Is Conscientious Objection? A health professional’s objection to providing a medical service is an instance of conscientious objection only if the objection is conscience based. There are several different conceptions of conscience (Benjamin 2004; Skorupski 2010; Wicclair 2013). However, for the purposes of analyzing conscientious objection in relation to end-​of-​life care, it suffices to stipulate that an objection is conscience based only if it is based on the health professional’s moral beliefs.1 For example, suppose an ICU physician refuses to participate in DCDD. If her refusal is based on her moral belief that DCDD is ethically unacceptable, then her refusal is conscience based and is an instance of conscientious objection. However, ICU physicians might object to participating in DCDD for other reasons. For example, a physician might have no ethical objection to DCDD and yet refuse to participate in the process to avoid uncomfortable interactions with surrogates or additional time on call. As another instance of a refusal to provide a service that is not conscience-​based, physicians and nurses who have no ethical objection to palliative sedation to unconsciousness might refuse to offer it or participate in its provision because it is not within the scope of their clinical competence.

Conscientious Objection   89 Clinicians might refuse to provide a medical service out of a concern for their own interests. For example, a clinician might refuse to provide end-​of-​life care for a Medicaid patient due to low reimbursement rates, or a clinician might refuse to treat patients with a deadly infectious disease such as SARS or Ebola out of a concern for his own health. Insofar as the first clinician’s refusal is based on financial considerations rather than a moral objection to providing the type of care requested (e.g., palliative sedation to unconsciousness), the objection is not conscience based, and insofar as the second clinician’s refusal to treat is based on a concern for his own health rather than his moral beliefs, the objection is not conscience based. Another possible reason for refusing to provide a requested medical service is a health professional’s belief that it is outside the boundaries of accepted clinical practice. For example, say a physician who is not ethically opposed in principle to providing palliative sedation to unconsciousness refuses family members’ request for it because the patient does not satisfy clinical guidelines. The physician cites a report on “Sedation to Unconsciousness” by the American Medical Association (AMA) Council on Ethical and Judicial Affairs. The report states that it is an option for patients in the “final stages of terminal illness” and is to be considered “an intervention of last resort to reduce severe, refractory pain or other distressing clinical symptoms that do not respond to aggressive symptom specific palliation.”2 In the physician’s judgment, the patient is not in the final stages of terminal illness and there are other means to offer him adequate relief from pain and discomfort. The physician’s refusal to offer palliative sedation is not conscience based. By contrast, a physician might believe that offering palliative sedation to unconsciousness never is morally permissible because doing so is inconsistent with her conception of acceptable medical practice. Insofar as offering palliative sedation to unconsciousness is within the currently recognized scope of accepted medical practice and the physician nevertheless refuses to offer it because of her unshakable belief that it would be morally wrong to do so, her refusal is an instance of a conscience-​ based objection. In the four U.S.  states in which physician assisted suicide is legally permitted (Oregon, Washington, Vermont, and Montana), clinicians might refuse to honor requests for suicide assistance based on their moral beliefs; and in the three states in which the practice is authorized by statute (Oregon, Washington, and Vermont), clinicians have a legal right to refuse (ORS 127.800–​995; RCW 70.245.010–​904; and 18 V.S.A.  ­chapter  113). However, in the states in which suicide assistance is not legally permitted, clinicians who have no ethical objection to suicide assistance can refuse to honor requests for it because of their unwillingness to risk legal penalties, professional censure, or loss of license to practice. Moreover, even if a practitioner in those states has an ethical objection to physician-​assisted suicide, there is no need to appeal to her moral beliefs to justify a refusal to provide it. Insofar as a practice such as suicide assistance or active euthanasia is legally prohibited, there is no expectation that physicians will provide it and no need to assert a conscience-​based objection to justify not providing it.

90   Mark R. Wicclair Generally, then, conscience-​based refusals to provide a medical service occur within a distinctive context: The service is legal, professionally accepted, clinically appropriate, and within the scope of the clinician’s competence, and providing it is contrary to the clinician’s moral beliefs.

Why Is the Exercise of Conscience Valuable and Worth Protecting? The exercise of conscience (i.e., an ability to act in accordance with one’s ethical beliefs) is valuable and worth protecting for several reasons (Wicclair 2011). The primary reason is the connection between the exercise of conscience and moral integrity. To maintain moral integrity, a person must refrain from performing actions that are against her conscience. As Blustein maintains, when one acts against one’s conscience, “one violates one’s own fundamental moral or religious convictions, personal standards that one sees as an important part of oneself and by which one is prepared to judge oneself ” (Blustein 1993, 295). Moral integrity in turn is valuable and worth protecting for several reasons (Wicclair 2011). First, it can be an essential component of a person’s conception of a good or meaningful life and can have intrinsic worth or value to a person. Second, a loss of moral integrity can be devastating. It can result in strong feelings of guilt, remorse, and shame, as well as loss of self-​respect. It also can be experienced as an assault on one’s self-​image or identity. Third, although the available evidence is equivocal, it has been claimed that a loss of moral integrity can result in a general decline in a person’s moral character. For example, Hepler advances a claim along these lines in relation to members of his profession (pharmacy): “We would be naive to expect a pharmacist to forsake his or her ethics in one area (e.g., abortion) while applying them for the patient’s welfare in every other area” (2005, 434). Finally, it can be claimed that moral integrity generally has intrinsic worth or value. To be sure, insofar as moral integrity can involve a commitment to any ethical or religious beliefs and principles, it does not guarantee ethically acceptable behavior. Nevertheless, similar to courage and honesty, which also can serve immoral ends and produce undesirable consequences, it might be claimed that moral integrity is a virtue and its value is not exclusively a function of ends and consequences. The exercise of conscience is valuable and worth protecting for several additional reasons. The first derives from the value of autonomy and the associated principle of respect for autonomous agents: “to respect an autonomous agent is, at a minimum, to acknowledge that person’s right to hold views, to make choices, and to take actions based on their personal values and beliefs” (Beauchamp and Childress 2009, 63). Second, the value of the exercise of conscience can be said to derive in part from the value of moral/​cultural diversity, and protecting its exercise can be defended as a requirement of toleration of moral/​cultural diversity (Wear, LaGaipa, and Logue 1994). Third, the notion of ethical epistemic modesty or humility can be cited as a basis for respecting and protecting

Conscientious Objection   91 the exercise of conscience. Ethical epistemic modesty is the view that, although ethical beliefs can be correct or incorrect and justified or unjustified, we might be mistaken when we think that a particular ethical belief is correct or justified. This recognition suggests “modesty” or “humility” and a rejection of dogmatism in relation to beliefs that we do not accept (Sulmasy 2008). A fourth reason applies specifically to health professionals. It can be argued that a failure to accommodate conscience-​based refusals to provide a medical service may discourage people who value moral integrity from entering health professions. An unintended consequence might be to preselect individuals who are ethically insensitive. Finally, protecting the exercise of conscience by health professionals can promote diversity within the health professions. For all these reasons, then, it can be maintained that the exercise of conscience is valuable and worth protecting.

Exemptions for Conscience-​based Objections Even if it is granted that the exercise of conscience is valuable and worth protecting, it still is necessary to consider when health professionals with a conscience-​based objection should be exempted from providing a medical service within the scope of their clinical competence. If an intensivist has a conscience-​based objection to offering palliative sedation to unconsciousness, should she be exempt from offering it? If an internist on an in-​patient general medicine service has a conscience-​based objection to withholding medically provided nutrition and hydration, should he be exempted from offering it as an option and writing orders to forgo it? If a nurse in a long-​term care facility has an objection to caring for patients after medically provided nutrition and hydration (MPNH) are discontinued, should she be exempted from caring for such patients? If a surgeon or a nurse has a conscience-​based objection to DCDD, should either be exempted from participating in it?

Two Extreme Approaches Answers to these and similar questions range over a continuum. At one extreme, the consistent answer is a simple and unnuanced “no:” Clinicians with conscience-​based objections to legal, professionally accepted, and clinically appropriate medical services within the scope of their clinical competence should not be exempted from providing them. We refer to this as the “no-​exemption” approach to conscience-​based refusals by health professionals. At the other extreme, the consistent answer is a simple and unnuanced “yes:” Clinicians should be exempted from providing any medical service that is against their consciences. This approach is referred to as “conscience absolutism.” Both extreme approaches can be challenged.

92   Mark R. Wicclair The following statement by Julian Savulescu, although he subsequently qualifies it, typifies the no-​exemption approach: “If people are not prepared to offer legally permitted, efficient, and beneficial care to a patient because it conflicts with their values, they should not be doctors” (2006, 94).3 A New York Times editorial expressed a similar sentiment in relation to pharmacists: “Any pharmacist who cannot dispense medicines lawfully prescribed by a doctor should find another line of work” (“Moralists at the Pharmacy” 2005). This view is seconded by the author of a syndicated op-​ed piece who proclaimed: [N]‌o one has the right to refuse to perform some foreseeable aspect of their job … [A] candidate for a pharmacy job [should] understand that she might have to hand out contraceptive pills and devices. She should either resolve to mind her own business or keep searching the want ads … You don’t like what the job requires? Fine. Get another job. (Pitts 2005)

The no-​exemption approach is similar to one of the five “conflict management models” for managing conscience-​based objections that Rebecca Dresser identifies. It “rules out the possibility of conscientious objection. According to this model, people who enter a profession agree to conform to its basic standards” (2005, 9). In defense of the no-​exemption approach, it can be argued that whereas health professionals have a professional obligation to put the interests of patients above their own (Pellegrino and Thomasma 1993), conscience-​based refusals place the interests of practitioners above the interests of patients. Accordingly, it is concluded that practitioners who refuse to provide a medical service for reasons of conscience fail to fulfill their professional obligations (Rhodes 2006). A model-​conscience protection statute and a model-​conscience protection policy for pharmacists exemplify conscience absolutism. The Protection of Conscience Project Model Protection of Conscience Act prohibits compelled direct and indirect participation in specified activities if a health-​care professional has a conscience-​based objection to such participation.4 The Pharmacists for Life International Model Conscience Clause states: The rights of conscience of any person being a duly licensed pharmacist, who shall object on personal, ethical, moral or religious grounds to the performance of any act in the normal course of professional performance or dispensing, shall be respected. Further, such a refusal to perform any act or the omission of any act based on such a claim of conscience, shall not form the basis for any claim for damages or any recriminatory or discriminatory action against such a person. (emphasis added)5

According to conscience absolutism, in addition to being exempted from providing a medical service that violates a health professional’s conscience, the practitioner should be exempted from directly or indirectly participating in its provision, facilitating patient access to it, or informing patients about it.

Conscientious Objection   93 The primary basis for conscience absolutism is the value of moral integrity. It is argued that exemptions are needed to enable health professionals to maintain their moral integrity. In addition, it is claimed, a health professional can maintain his moral integrity only if he avoids direct as well as indirect participation in activities that violate his conscience. Karen Brauer, president of Pharmacists for Life, appeals to this line of reasoning to defend the view that pharmacists with a conscience-​based objection to filling prescriptions should not be expected to facilitate a transfer or provide referrals: That’s like saying, “I don’t kill people myself but let me tell you about the guy down the street who does.” What’s that saying? “I will not off your husband, but I know a buddy who will?” It’s the same thing. (Stein 2005)

There is at least a grain of truth in both extreme views. On the one hand, the no-​ exemption approach serves as a reminder that health professionals have responsibilities to patients, surrogates, other clinicians, administrators, and organizations. Hence, at the very least, in considering whether health professionals with conscience-​based objections to providing a medical service should be exempted from providing it, a relevant consideration is whether an exemption would interfere with their ability to satisfactorily discharge those responsibilities. On the other hand, conscience absolutism serves as a reminder that health professionals also are moral agents. Hence, at the very least, in considering whether health professionals with conscience-​based objections to providing a medical service should be exempted from providing it, a relevant consideration is the impact on their moral integrity. If there is a grain of truth in both of two extreme views, a natural strategy is to try to find a middle ground that captures the valuable insights of each. Dan Brock proposes an approach along these lines, which he refers to as the “conventional compromise.” Applied to physicians and pharmacists, it exempts them from providing a service or product that is against their conscience only if the following three conditions are satisfied (2008, 194): 1.  The physician/​pharmacist informs the patient/​customer about the service/​product if it is medically relevant to their medical condition. 2. The physician/​pharmacist refers the patient/​customer to another professional willing and able to provide the service/​product. 3.  The referral does not impose an unreasonable burden on the patient/​customer. An important objective of a “middle ground” approach is to strike an appropriate balance between protecting the moral integrity of health professionals and protecting the legitimate interests of patients, surrogates, other clinicians, administrators, and organizations. However, depending on the circumstances, the three conditions of the conventional compromise may limit moral integrity more than necessary to adequately protect patients and their families. For example, suppose an ICU physician has a moral

94   Mark R. Wicclair objection to palliative sedation to unconsciousness, and informing family members of this option would compromise her moral integrity. Depending on ICU staff resources, it may be possible to assign the responsibility of informing family members to other willing physicians. In addition, none of the three conditions addresses the interests of other clinicians, administrators, and organizations. The third condition requires that referrals do not impose an unreasonable burden on the patient/​customer. However, depending on institutional resources, referrals can impose unreasonable burdens on other clinicians, which is a relevant consideration when determining whether to accommodate.

Reasonable Accommodation There is a more satisfactory approach to managing health-​care providers’ conscience-​ based objections. We refer to this as “reasonable accommodation.” The general aim is to accommodate clinicians’ claims of conscience without unjustifiably compromising other values and interests. Reasonable accommodation guidelines include the following: 1. Whenever feasible and apt, clinicians should provide advance notification to patients/​surrogates, administrators, employers, etc. 2. Accommodation should not impede a patient’s/​ surrogate’s timely access to information, counseling, and referral. 3. Accommodation should not impede a patient’s timely access to health care services. 4.  Accommodation should not impose excessive burdens on other clinicians, administrators, or organizations. There are three major differences between these reasonable accommodation guidelines and the conventional compromise requirements. First, the reasonable accommodation guidelines are more context dependent. Second, the guidelines include a provision to protect patients and surrogates from the burdens of preventable eleventh-​ hour conscience-​based refusals. Third, protections are not limited to patients. The reasonable accommodation guidelines are more context dependent insofar as they require a consideration of outcomes rather than performing specified actions. For example, whereas the conventional compromise requires clinicians to inform and refer, the reasonable accommodation guidelines require only that accommodation does not impede a patient’s/​surrogate’s timely access to information and referral. Accordingly, if it would compromise a clinician’s moral integrity to provide information about an option, such as palliative sedation to unconsciousness, she would not be required to do so unless there were no alternative means for patients/​surrogates to receive information about it in a timely manner. The physician may need only to direct the patient or surrogate to another clinician who will provide the information. The timing of disclosing a conscience-​based refusal to patients/​surrogates can significantly affect the impact on them. Patients are prone to be especially vulnerable and

Conscientious Objection   95 dependent at times when they are in need of medical care. A corresponding point applies to surrogates. To minimize burdens on patients/​surrogates, whenever feasible and apt, practitioners who have a conscience-​based objection to a medical service within the scope of their clinical competence should provide advance notice of their objection to patients/​surrogates and not wait until it becomes a clinically suitable option. Clinicians may be able to minimize harms and burdens by giving (prospective) patients an opportunity to learn about the provider’s conscience-​based objections during their initial meeting. It is then that finding another provider is the least upsetting and distressing. Advance notification also can maximize a patient’s opportunity to establish a relationship with another practitioner—​one who has no conscience-​based objection to providing a medical service that the patient someday may want. However, advance notification has to be exercised with care so as to minimize the risk of unnecessary distress to patients. For example, if a physician informs a patient who presents with a suspicious lump in her breast that he is conscientiously opposed to forgoing MPNH, the patient may mistakenly infer that she has terminal cancer and is dying. The same observation would apply if a physician were to disclose his conscience-​ based objection to forgoing MPNH when first informing a patient that she has acute myelogenous leukemia. It would be no more apt to disclose a conscience-​based objection to forgoing MPNH in such circumstances than it would be to initiate a discussion of the patient’s end-​of-​life preferences. For patients who are in hospitals or nursing homes, advance notification may be infeasible or may accomplish little because they are being cared for by multiple, sometimes rotating providers, and their ability to select clinicians may be very limited. Accordingly, for hospital and nursing home staff, the primary recipient of advance notification may be administrators and supervisors. In addition to reducing burdens on administrators and supervisors, advance notification can increase the likelihood that staffing assignments and scheduling can be made to facilitate accommodation. For example, if a surgical nurse informs a supervisor when she is offered a position that she objects to participation in DCDD, an accommodation is more likely to be feasible than if she waits to inform the supervisor until she is asked to participate in the procedure. Finally, since advance notification can enable other clinicians who might be asked to substitute to make any necessary professional and personal adjustments, such notification can also minimize inconvenience and disruption to them. When clinicians do not object in principle to a procedure or practice, advance notification can be more challenging but still not infeasible. For example, Dr. A, an NICU physician, and Nurse B, an NICU nurse, are not ethically opposed in principle to providing aggressive treatment to preterm neonates. However, they both have conscience-​ based objections to continuing aggressive life support for Baby C, a preterm infant with an extremely poor prognosis. Although continued aggressive treatment is contrary to Dr. A’s conception of “good medicine” and Nurse B’s conception of “good nursing practice,” it does not violate established professional norms and is not clearly outside the boundaries of “appropriate medical/​nursing care.” To facilitate advance notification of NICU administrators, Dr. A and Nurse B should attempt to identify their respective

96   Mark R. Wicclair general criteria for deciding when providing aggressive treatment to premature newborns is contrary to their conceptions of “good medicine” and “good nursing practice,” respectively. Generally, to facilitate advance notification, clinicians should attempt to anticipate the types of situations in which they are likely to request exemptions. In addition to the potential negative impact on patients and surrogates, conscience-​ based refusals can impose significant burdens on other clinicians, administrators, or organizations. For example, to accommodate a critical-​care nurse who has a conscience-​ based objection to DCDD, the supervisor will have to make the necessary work assignments, which may require significant modifications and re-​scheduling. Although doing this once may not be a substantial burden, doing it repeatedly may meet the “excessive” threshold. In addition, other nurses must be assigned to assist in DCDD, which might increase their workloads or necessitate unwanted changes in their work schedules. Similarly, to enable a hospital or nursing home nurse to be exempted from caring for patients who are unable to take nutrition and hydration orally and who themselves or whose family members have decided to forgo MPNH, it might be necessary to switch assignments with another nurse, which may be an unwelcome change for the latter. Accommodating physicians can be no less burdensome on department heads and other clinicians. For example, if an on-​call intensivist refuses to participate in withdrawing life support for patients who are to undergo DCDD, other intensivists will have to be available to supervise and perform established DCDD procedures. Hence, accommodating her can impose burdens on the person who is responsible for scheduling arrangements and on the other intensivists. Again, although this may not be unduly burdensome on one occasion, repeated accommodation can reach the “excessive” threshold. Unfortunately, there is no simple rule for determining when burdens are “excessive,” in part because excessiveness is largely context dependent. Whether an accommodation will impose excessive burdens depends on a variety of contextual factors, including the number of staff members whose clinical competencies overlap with those of the clinician who seeks accommodation, the willingness of those clinicians to provide the medical service from which the clinician requests an exemption, the existing responsibilities and workload of administrators and staff, and the availability of funds to pay overtime or hire additional staff. Moreover, the assessment of burdens is in part subjective insofar as burdensomeness depends on a person’s life and perceptions. For example, whereas a schedule shift or additional time on service may not be perceived as an excessive burden by a clinician who has no partner or children, it might be perceived as an excessive burden by a clinician who is a single parent of a young child or a clinician who has other professional responsibilities (e.g., research, consulting, or teaching). The second and third guidelines (accommodation should not impede a patient’s/​surrogate’s timely access to information, counseling, and referral; accommodation should not impede a patient’s timely access to health care services) can be grounded in health professionals’ obligations to promote patients’ autonomy, health, and well-​being. These obligations are cited in major professional codes, such as the AMA Code of Medical Ethics (AMA Council on Ethical and Judicial Affairs 2010) and the American Nurses Association Code of Ethics for Nurses (Fowler 2008).

Conscientious Objection   97 In a section on informed consent, the AMA Code of Medical Ethics includes the following implicit recognition of the value of patient autonomy:  “The patient should make his or her own determination about treatment” (Informed Consent, Opinion 8.08). An opinion on “therapeutic privilege” (the practice of withholding medical information from patients “in the belief that disclosure is medically contraindicated”) recognizes an obligation to respect patient autonomy:  “It [therapeutic privilege] creates a conflict between the physician’s obligations to promote patients’ welfare and respect for their autonomy by communicating truthfully … Withholding medical information from patients without their knowledge or consent is ethically unacceptable” (Withholding Information from Patients, Opinion 8.082). In addition, an AMA House of Delegates Health Policy statement calls on physicians to respect patient autonomy in the context of decisions to forgo life-​sustaining measures: “The principle of patient autonomy requires that physicians must respect the decision to forgo life-​sustaining treatment of a patient who possesses decision-​ making capacity” (Decisions Near the End of Life, House of Delegates Health Policy 140.966).6 The following excerpts from the AMA Code of Medical Ethics express a strong commitment to patient well-​being: The medical profession has long subscribed to a body of ethical statements developed primarily for the benefit of the patient. As a member of this profession, a physician must recognize responsibility to patients first and foremost. (Preamble, Principles of Medical Ethics)

A physician shall, while caring for a patient, regard responsibility to the patient as paramount (Principle VIII, Principles of Medical Ethics). The relationship between patient and physician is based on trust and gives rise to physicians’ ethical obligations to place patients’ welfare above their own self-​interest and above obligations to other groups, and to advocate for their patients’ welfare. (The Patient–​Physician Relationship, Opinion 10.015)

Three provisions of the American Nurses Association Code of Ethics for Nurses have been characterized as “the most fundamental values and commitments of the nurse” (Fowler 2008, 146). The first provision requires “respect for the inherent dignity, worth, and uniqueness of every individual” (147). An interpretive statement explicitly links this provision to patient self-​determination: “Respect for human dignity requires the recognition of specific patient rights, particularly, the right of self-​determination” (148). Each of the other two provisions expresses a commitment to patient health and well-​being. Provision 2 states: “The nurse’s primary commitment is to the patient” (150). The third provision states: “the nurse promotes, advocates for, and strives to protect the health, safety, and rights of the patient” (152). Lack of timely access to health services, counseling, and referral can substantially limit patient autonomy and can have a serious negative impact on a patient’s health and

98   Mark R. Wicclair well-​being. Accordingly, the third and fourth reasonable accommodation guidelines are based in part on the principle that accommodations for physicians and nurses should be compatible with their ability to fulfill their fundamental obligations to patients. The timeliness of access is context dependent. For example, whether a delay of an hour or a day fails to satisfy the timeliness requirement depends on a number of factors, including the urgency of the service, the patient’s medical condition, the impact on the patient and family members, and the willingness of the patient and family members to accept a delay. A report of the AMA Council on Ethical and Judicial Affairs published in 2007 considered and rejected a resolution proposed by the Medical Student Association to adopt a specific policy on conscientious objection that would limit it to nonemergent situations and require physicians with conscience-​based objections to “provide alternative(s) which include a prompt and appropriate referral” (CEJA Report 6-​A-​07).7 The report recommended no new policy statements. It did acknowledge that physicians have a “prerogative” to “refuse to provide a treatment to which they object on the basis of religious or moral beliefs.” However, it observed that “other Principles balance this prerogative with obligations to respect patients and their ability to access available medical care.” The reasonable accommodation guidelines provide a framework for striking an appropriate balance. The American Nurses Association Code of Ethics for Nurses includes the following general statement that addresses nurses’ conscience-​based objections:  “The nurse is obliged to provide for the patient’s safety, to avoid patient abandonment, and to withdraw only when assured that alternative sources of nursing care are available to the patient” (160–​161). This statement provides only limited guidance for determining whether to accommodate nurses with conscience-​based objections. The reasonable accommodation guidelines provide a more comprehensive framework—​arguably one that is consistent with the “the most fundamental values and commitments of the nurse” cited above. Typically, when ethical guidelines of other health professional organizations address conscientious objection, they are limited in scope (e.g., limited to forgoing life-​ sustaining treatment) and/​or specificity. For example a 1991 American Thoracic Society position paper states: [I]‌f carrying out … a request [by an informed and competent patient to forgo life-​ sustaining therapy] would violate the personal moral code of a physician or other health care provider, that individual generally has the right not to participate in the process. If this occurs, others should be made available to carry out the patient’s request. (American Thoracic Society 1991, 478)

This statement is limited to forgoing life-​sustaining therapy, and the qualifier “generally” suggests that there are exceptions, but no criteria for identifying any are provided. In addition, although the statement advises that “others should be made available to carry out the patient’s request,” it fails to provide a time-​frame or specify whether

Conscientious Objection   99 the physician has an obligation to carry out the patient’s request if “others” cannot be “made available.”8 In contrast to such limited guidelines, the reasonable accommodation guidelines provide a more comprehensive and detailed framework for assessing requests for accommodation by health professionals with conscience-​based objections to proving a medical service. However, since the reasonable accommodation guidelines are context dependent, determining whether they are satisfied in a particular case will depend on contextual factors such as those previously identified.

Constraints on Conscience-​based refusals There is an important distinction between two types of conscience-​based refusals: (1) a refusal to provide a medical service because a health professional believes that the service (e.g. DCDD or palliative sedation to unconsciousness) is unethical and (2) a refusal to provide a service only to specified classes of patients (e.g. African Americans, Muslims, lesbian women, or gay men). Depending on the nature of the class, conscience-​ based refusals of the second kind can be based on invidious discrimination. It is a settled view—​one based on defensible and widely shared conceptions of justice, equality, dignity, and respect—​that racial, ethnic, religious and gender-​based prejudice or bias are ethically wrong. Various professional codes of ethics, such as the AMA Code of Medical Ethics (AMA Council on Ethical and Judicial Affairs 2010) and the American Nurses Association Code of Ethics for Nurses (Fowler 2008), prohibit invidious discrimination. Hence, general ethical considerations as well as professional codes of ethics support a constraint against conscience-​based refusals based on invidious discrimination. It is, of course, possible to question whether a particular specification of the scope of invidious discrimination is justified. For example, although it is a settled view that race-​based prejudice is ethically unacceptable, it might be questioned whether moral disapproval of gay, lesbian, bisexual, or transgendered patients reflects prejudice or unjustified bias. The AMA added sexual orientation to the specified types of prohibited invidious discrimination in 1993 and gender identity in 2007.9 This expansion indicates that the scope of prohibited invidious discrimination within a profession can change over time. Such changes correspond to changes in accepted views within and outside the profession about the scope of invidious discrimination and, arguably, appropriately limit conscience-​based refusals.10 Is moral disapproval of patients an acceptable reason for refusing to care for them? It can be argued that insofar as clinicians are members of professions committed to health, well-​being, and respect for autonomy and dignity, moral disapproval of a patient, even if justifiable, is not an appropriate reason to withhold medical services. Clearly, if a clinician refuses to provide medical services to racists and convicted serial killers, rapists, and child molesters due to their alleged unethical conduct and lack of moral virtue, the refusal cannot be attributed to an unjustified prejudice or bias. Nevertheless, it can be maintained that a refusal would violate a practitioner’s professional obligations.

100   Mark R. Wicclair In response, it might be claimed that even though a virtuous practitioner would never refuse to provide care due to moral disapproval of a patient, as long as another clinician will provide medically necessary care in a timely fashion, a particular practitioner may refuse to treat racists, and convicted serial killers, rapists, and child molesters without violating her professional obligations. To be sure, this claim is debatable. Even if accepted, however, the same principle does not apply to refusals based on discrimination that is indisputably invidious, such as those based on race or ethnicity. Refusals based on prejudice or unjustified bias, even if conscience based and even if another practitioner is available to provide medically necessary services in a timely fashion, violate a general ethical constraint against invidious discrimination and a special obligation of health professionals. Generally, the value of affording clinicians moral space in which to maintain their integrity precludes placing additional constraints on acceptable conscience-​ based refusals. However, there are two exceptions. First, a provider’s conscience-​based refusal has little, if any, moral weight if it is undeniably incompatible with the goals of his profession. Suppose, for example, after experiencing a religious conversion, an internist no longer will provide pain medication to terminal cancer patients or refer them to a palliative care service. He now believes that pain is God’s punishment for sin and promoting pain relief would thwart God’s justice. Arguably, the internist’s conscience-​based objection to facilitating pain relief does not merit any moral weight. In such very unlikely contexts, the no-​exemption approach applies, and it can be said that someone who has a conscience-​based objection to relieving pain should not enter disciplines, such as internal medicine, palliative care, or nursing, that are committed to that goal. Second, conscience-​based refusals that are based on demonstrably false clinical beliefs merit little, if any, moral weight. Such cases are likely to be extremely rare. One example is a pharmacist whose conscience-​based objection to dispense emergency contraception (EC) is based on mistaken beliefs about its mechanism of action. A study of South Dakota pharmacists reported that 36.6 percent of the respondents did not correctly identify the mechanism of action of EC, and 19  percent incorrectly identified it as most similar to that of the abortifacient mifepristone (Van Riper and Hellerstedt 2005). Another study reported that 35.8 percent of New Mexico pharmacists surveyed mistakenly believed that “[o]‌ral emergency contraception is also known as RU-​486” (Borrego et al. 2006, 37). Such mistaken beliefs about EC are not limited to pharmacists. For example, similar findings are reported for family medicine physicians and nurses (Wallace et al. 2004). If such mistaken beliefs about the mechanism of action of EC are essential to a clinician’s conscience-​based objection to dispense it, no accommodation is warranted. It might be claimed that a conscience-​based refusal relies on demonstrably false clinical beliefs if it is based on a clinician’s skepticism about the whole brain (neurological) criterion of death. Arguably, however, this claim is not justified. To be sure, the whole brain criterion is a widely accepted legal and clinical standard of death. Nevertheless, problems and unanswered questions remain (President’s Council on Bioethics 2008; Shewmon 2009; Youngner, Arnold, and Schapiro 1999). Accordingly, if a physician

Conscientious Objection   101 refuses to rely on the brain death protocol to certify death because she believes that patients who satisfy the whole brain criterion might not be “really dead,” denying an accommodation cannot be justified on the grounds that her refusal is based on a demonstrably false belief. However, there may be other valid reasons to deny accommodation. For example, accommodation may place excessive burdens on the institution, the department or unit, and other clinicians. A similar point applies to conscience-​based objections to participating in DCDD. There is persistent controversy about whether patients who are declared dead subsequent to withdrawing life support by means of the circulatory criterion are “really dead” (Bernat 2010; Bernat et al. 2010; Menikoff 1998; Troug and Miller 2010). Accordingly, it is not justified to claim that a conscience-​based refusal to participate in DCDD relies on a demonstrably false clinical belief if the objection is based on a clinician’s belief that patients may not satisfy the dead donor rule (i.e., the rule that organ donors must be dead before vital organs may be removed) if their death is determined by the institution’s DCDD protocol). Although it may not be justified to deny an accommodation on the grounds that a refusal to participate in DCDD is based on a demonstrably false clinical belief, there may be other valid reasons to deny accommodation, such as the expected impact on the institution, department, unit, and/​or staff.

An Objection to Reasonable Accommodation It might be objected that the reasonable accommodation guidelines are unsatisfactory because they may require health professionals to compromise their moral integrity. Undeniably, depending on the specific circumstances, the guidelines may require health professionals to compromise their moral integrity. For example, a physician with a conscience-​based objection to offering palliative sedation to unconsciousness or writing orders to forgo MPNH might be required to compromise her moral integrity if she cannot be exempted without placing an excessive burden on the other clinicians who would be responsible for providing these services, or if an accommodation would prevent timely disclosure or access. However, this objection fails to consider that individuals acquire special obligations when they enter a health profession. Hence, depending on the circumstances, fulfilling one’s professional obligations may require a willingness to compromise moral integrity. Herein is a grain of truth in Savulescu’s previously cited claim that people who are “not prepared to offer legally permitted, efficient, and beneficial care to a patient because it conflicts with their values … should not be doctors” (2006, 94). Individuals who plan to enter a health profession can minimize the risk of being in situations that threaten their moral integrity by judiciously selecting practice disciplines and specialties or subspecialties. For example, pediatric residents who plan to pursue a fellowship and who have conscience-​based objections to offering parents a full range of legal and professionally accepted end-​of-​life options should consider fellowships in areas other than critical care. Clinicians can further minimize the risk of finding themselves in situations

102   Mark R. Wicclair that compromise their moral integrity by a careful choice of practice environments. For example, a physician or nurse who refuses to care for patients who refuse MPNH should not practice in a hospice setting. Similarly, clinicians with conscience-​based objections to a medical service may find a more accommodation-​friendly environment in a large urban medical center than in a small rural community hospital.

Conscience Clauses: Law and Public Policy Legal rules and regulations that protect the exercise of conscience are commonly referred to as “conscience clauses.” The Church Amendment (42 U.S.C. § 300a-​ 7(a–​b)), enacted in 1973, was the first U.S.  federal health-​care conscience clause. Although its scope initially was limited to abortion and sterilization, sections were added later that significantly expand the scope of its protections. For example, one additional section prohibits “discrimination” against a clinician for refusing to perform or assist in the performance of “any lawful health service or research activity on the grounds that his performance or assistance in the performance of such service or activity would be contrary to his religious beliefs or moral convictions” (42 U.S.C. § 300a-​7(c)). Additional federal conscience clause legislation includes the Coats Amendment (42 U.S.C. § 238n) and the Weldon Amendment (121 Stat. 1844, 2209). Both are limited to abortion. The Coats Amendment extends the protection of conscience to postgraduate physician training programs that refuse to provide abortion training as well as trainees who refuse to receive abortion training. The Weldon Amendment denies funding to federal agencies or programs and state or local governments that require individuals or institutions “to provide, pay for, provide coverage of, or refer for abortions.” In December 2008, in the final days of the presidency of George W.  Bush, the Department of Health and Human Services (HHS) issued a Final Rule that provided broad and comprehensive protection for practitioners and institutions that refuse to perform or assist in the performance of activities that violate their ethical or religious beliefs (45 CFR § 88).11 The HHS claimed that the Final Rule systematized previously enacted laws and regulations. However, this claim was challenged by those who maintained that it substantially broadened the scope of federal conscience protection legislation. In part in response to the criticism that the HHS Final Rule had inappropriately broadened the scope of conscience protection at the expense of patients’ access to health services, it was rescinded shortly after President Obama’s first inauguration. The regulation was short-​lived since its effective date was January 20, 2009, inauguration day. The U.S. health-​care reform legislation enacted in March 2010, the Patient Protection and Affordable Care Act (Public Law No. 111–​148), includes provisions that protect

Conscientious Objection   103 refusals related to abortion.12 Beyond these abortion-​related provisions, a subsection of the act protects clinicians who refuse to provide “any health care item or service furnished for the purpose of causing, or for the purpose of assisting in causing, the death of any individual, such as by assisted suicide, euthanasia, or mercy killing” (§ 1553(a)). The Act explicitly states that this subsection shall not “be construed to apply to, or to affect, any limitation relating to” withholding or withdrawing medical treatment, including nutrition and hydration, abortion, or “the use of an item, good, benefit, or service furnished for the purpose of alleviating pain or discomfort, even if such use may increase the risk of death, so long as such item, good, benefit, or service is not also furnished for the purpose of causing, or the purpose of assisting in causing, death, for any reason” (§1553(a)). Most states also have enacted conscience clause legislation, and most of these laws explicitly apply to abortion. Some explicitly protect other services, such as sterilization, contraceptives, family planning, artificial insemination, human embryonic stem cell research, human embryonic cloning, physician-​assisted suicide, and euthanasia. Two states, Illinois (Ill Rev Stat ch 745, § 70 (1998)) and Mississippi (Miss Code § 41–​107 (2004)), have broad conscience protection laws that provide blanket protection for conscience-​based refusals of any health procedure or service. In addition, state laws that authorize advance directives typically provide exemptions for health-​care professionals who have ethical or religious objections to implementing instruction directives and/​or the decisions of designated surrogates (Meisel and Cerminara 2009). Strictly speaking, not all laws and regulations that protect refusals by practitioners and institutions and are commonly classified as “conscience clauses” warrant that characterization. To be correctly classified as a conscience clause, a legal rule or regulation should specify that the refusal is based on ethical or religious beliefs. However, several laws and regulations commonly classified as conscience clauses do not include this requirement. On a federal level, the protections afforded by the Weldon and Coats Amendments are not limited to refusals based on ethical or religious beliefs. Neither are some of the protections provided by the Patient Protection and Affordable Care Act. Similarly, many state statutes that protect refusals related to abortion are not limited to those based on ethical or religious beliefs and are not properly classifiable as conscience clauses. A Minnesota statute even explicitly states that the reason for abortion-​related refusals is not relevant: No person and no hospital or institution shall be coerced, held liable or discriminated against in any manner because of a refusal to perform, accommodate, assist or submit to an abortion for any reason. (Minn Stat § 145.414 (1995); emphasis added)

On both the federal as well as the state level, a major shortcoming of conscience clauses is a failure to consider contextual factors. The federal Church and Weldon Amendments, the rescinded HHS Final Rule, and the Patient Protection and Affordable Care Act include no consideration of the impact of conscience-​based refusals on

104   Mark R. Wicclair patients, institutions, or other clinicians. Typically, the same is true of state conscience clauses, and only a few include exceptions for emergency situations. Typically, conscience clauses are either silent or exempt providers from a requirement to disclose, refer, or facilitate a transfer. Louisiana, Arkansas, and Illinois are among the states that provide such explicit exemptions in their conscience clauses. Maryland’s conscience clause considers some contextual factors and is an exception. It states that the protections of the conscience clause do not apply to referral: if the failure to refer a patient to a source for any medical procedure that results in sterilization or termination of pregnancy would reasonably be determined as: (1) The cause of death or serious physical injury or serious long-​lasting injury to the patient; and (2) Otherwise contrary to the standards of medical care. (Md Code Ann § 20–​ 214(d) (1991))

Similar to many state conscience clauses, the protections of the federal Church and Weldon Amendments apply to conscience-​based refusals “to perform or assist in the performance of ” (specified or unspecified) procedures or activities. Neither includes an explanation of what it means to “assist in the performance of.” The rescinded HHS Final Rule provided an extremely broad definition that was said to include “counseling, referral, training, and other arrangements for the procedure, health service, or research activity” (45 CFR § 88.2). No provision was made in the Final Rule for a consideration of contextual factors that might determine whether a refusal will interfere with a patient’s or surrogate’s timely access to information or medical services. Typically, both state and federal conscience clauses limit the extent to which administrators and supervisors can consider contextual factors when deciding whether to accommodate clinicians with conscience-​based objections. Maine’s abortion-​related conscience clause is representative of such limitations: No physician, nurse or other person, who refuses to perform or assist in the performance of an abortion, shall, because of that refusal, be dismissed, suspended, demoted or otherwise prejudiced or damaged by a hospital, health care facility, firm, association, professional association, corporation or educational institution with which he or she is affiliated or requests to be affiliated or by which he or she is employed. (Title 22 Me Rev Stat Ann § 1591 (1977))

If a similar approach were adopted in relation to end-​of-​life care, then even if accommodating a critical-​care nurse with a conscience-​based objection to participating in palliative sedation to unconsciousness were to impose an excessive burden on the hospital (e.g., require hiring another nurse) or other nurses (e.g., require changing shifts when it is excessively burdensome), the conscience clause would prevent a supervisor from taking any action against the nurse for refusing to perform her assigned duties. As a result, the supervisor would be powerless to enforce a denial of the nurse’s request for accommodation. In effect, then, the conscience clause requires the supervisor to grant an accommodation, no matter how great the burden to the hospital and other nurses.

Conscientious Objection   105 In view of the failure to consider contextual factors, most current conscience clause legislation provides a poor model for public policy in relation to conscientious objection in relation to end-​of-​life care. Although Title VII of the 1964 Civil Rights Act (42 USCS § 2000e et seq. (2005)) applies only to “employers” and is limited in scope, it exemplifies a more suitable contextual approach. It requires employers to “reasonably accommodate” conscience-​based objections of employees unless it would result in an “undue hardship” on the employer. This approach is preferable to conscience clause legislation that fails to consider how accommodation would likely affect employers, health-​care institutions, administrators, patients, and surrogates. It is arguable that there is another reason why most current conscience clause legislation provides a poor model for public policy. Typically, conscience-​based objections are protected only if practitioners refuse to provide a service that violates their ethical or religious beliefs. However, there are situations in which clinicians believe they have a moral obligation to provide a prohibited service. For example, some states substantially restrict the ability of physicians to implement the advance directives of pregnant women. Pennsylvania Act 69 is representative: Notwithstanding the existence of a living will, a health care decision by a health care representative or health care agent or any other direction to the contrary, life-​sustaining treatment, nutrition and hydration shall be provided to a pregnant woman who is incompetent and has an end-​stage medical condition or who is permanently unconscious unless, to a reasonable degree of medical certainty as certified on the pregnant woman’s medical record by the pregnant woman’s attending physician and an obstetrician who has examined the pregnant woman, life-​sustaining treatment, nutrition and hydration: (1) will not maintain the pregnant woman in such a way as to permit the continuing development and live birth of the unborn child; (2) will be physically harmful to the pregnant woman; or (3) will cause pain to the pregnant woman that cannot be alleviated by medication. (20 Pa Cons Stat Ann § 5429 (a) (2006))

Significantly, Act 69 permits a physician with a conscience-​based objection to forgoing life support to opt out by facilitating a transfer to another physician. Yet it does not provide a similar accommodation for a physician who believes, as a matter of conscience, that she has an obligation to respect a patient’s documented wishes to forgo life-​ sustaining treatment. It is at least arguable that public policy should not give clinicians moral space in which to protect their moral integrity only when they refuse to provide a medical service. For all these reasons, then, most current conscience clause legislation provides a poor model for public policy. A suitable approach to managing conscientious objection in relation to end-​of-​life care must include a consideration of ethically relevant contextual factors, such as the impact on patients, surrogates, family members, other clinicians, and institutions. Whereas most current conscience clause legislation ignores such factors, the recommended reasonable accommodation guidelines require that they be considered.

106   Mark R. Wicclair

Notes 1. The term “moral beliefs” includes beliefs that are not religiously based as well as beliefs that are religiously based. 2. CEJA Report 5-​A-​08. Available online at http://​www.ama-​assn.org/​ama1/​pub/​upload/​ mm/​369/​ceja_​5a08.pdf;: accessed January 3, 2013. 3. Savulescu eventually qualifies this claim to permit conscientious refusals as long as they do not restrict patient access to health services. In addition, his primary concern is with physicians who are government employees, such as physicians within the British National Health Service. 4. Available online at http://​www.consciencelaws.org/​model-​statute.aspx; accessed January 4, 2013. The model statute does not define “indirect participation.” However, it explicitly protects health care professionals from being compelled to counsel or educate “persons in a manner which indicates that the activities [specified in the act] … are morally neutral or acceptable.” 5. Available online at www.pfli.org/​main.php?pfli=modelpharmacistcc; accessed January 4, 2013. 6. House of Delegates Health Policy statements are accessible online via the AMA “Policy Finder”: http://​www.ama-​assn.org/​ama/​pub/​about-​ama/​our-​people/​house-delegates/​policyfinder.page ; accessed January 6, 2013. 7. Available online at http://​www.ama-​assn.org/​resources/​doc/​ethics/​ceja_​6a07.pdf; accessed January 6, 2013. 8. At the time of writing, an American Thoracic Society working group was drafting an extensive policy statement on managing conscientious objection in the ICU that includes detailed guidelines. 9. The AMA Board of Trustees approved adding discrimination on the basis of sexual orientation in 1993 and the House of Delegates approved it five years later (Schneider and Levin 1999: 1287–​1288). The Board of Trustees approved the addition of gender identity in 2007 (BOT Report 11, E-​9.03, “Recommendations to Modify AMA Policy to Ensure Inclusion for Transgender Physicians, Medical Students, and Patients”). The Council on Ethical and Judicial Affairs approved it in the same year (CEJA Report 2-​I-​07). 10. An example outside the health-​care professions is the explicit prohibition of employment discrimination based on gender identity that was added to the U.S. federal jobs website in January 2010 (Knowlton 2010: sec. A, p. 15). 11. Federal Register 73, 245 (Friday, December 19, 2008), p. 78073. Available online at http://​ edocket.access.gpo.gov/​2008/​pdf/​E8-​30134.pdf; accessed January 11, 2013. 12. The Patient Protection and Affordable Care Act is available online at http://​www.gpo.gov/​ fdsys/​pkg/​PLAW-​111publ148/​pdf/​PLAW-​111publ148.pdf; accessed February 2, 2013.

References American Medical Association (AMA) Council on Ethical and Judicial Affairs. 2010. Code of Medical Ethics: Current opinions, 2010–​2011. Chicago: Author. American Thoracic Society. 1991. Withholding and withdrawing life-​sustaining therapy. Annals of Internal Medicine 115(6):478–​485. Beauchamp, T. L., and J. F. Childress. 2009. Principles of biomedical ethics. 6th ed. New York: Oxford University Press.

Conscientious Objection   107 Benjamin, M. 2004. Conscience. In Encyclopedia of bioethics, edited by Stephen G. Post. New York: Macmillan Reference, 513–​517. Bernat, J. L. 2010. Point: Are donors after circulatory death really dead, and does it matter? Yes and yes. Chest 138(1):13–​16. Bernat, J. L., A. M. Capron, T. P. Bleck, S. Blosser, S. L. Bratton, J. F. Childress, M. A. DeVita, G. J. Fulda, C. J. Gries, M. Mathur, T. A. Nakagawa, C. H. Rushton, S. D. Shemie, and D. B. White. 2010. The circulatory-​respiratory determination of death in organ donation. Critical Care Medicine 38(3):963–​970. Blustein, J. 1993. Doing what the patient orders: Maintaining integrity in the doctor-​patient relationship. Bioethics 7(4):289–​314. Borrego, M. E., J. Short, N. House, G. Gupchup, R. Naik, and D. Cuellar. 2006. New Mexico pharmacists’ knowledge, attitudes, and beliefs toward prescribing oral emergency contraception. Journal of the American Pharmaceutical Association 46(1):33–​43. Brock, D. 2008. Conscientious refusal by physicians and pharmacists: Who is obligated to do what, and why? Theoretical Medicine and Bioethics 29:187–​200. Dresser, R. 2005. Professionals, conformity, and conscience. The Hastings Center Report 35(6):9–​10. Fowler, M. D. M. 2008. Guide to the Code of Ethics for Nurses: Interpretation and application. Silver Spring, MD: American Nurses Association. Guidelines on the termination of life-​sustaining treatment and the care of the dying. 1987. Briarcliff Manor, NY: Hastings Center. Hepler, C. D. 2005. Balancing pharmacists’ conscientious objections with their duty to serve. Journal of the American Pharmacists Association 45(4):434–​436. Knowlton, B. 2010. U.S. job site bans bias over gender identity. The New York Times, January 6, A–​15. Meisel, A. 1992. The legal consensus about forgoing life-​sustaining treatment: Its status and its prospects. Kennedy Institute of Ethics Journal 2(4):309–​345. Meisel, A., and K. L. Cerminara. 2009. The right to die: The law of end-​of-​life decisionmaking. 3rd ed. Austin: Wolters Kluwer. Menikoff, J. 1998. Doubts about death: The silence of the institute of Medicine. Journal of Law, Medicine and Ethics 26(2):157–​165. Moralists at the pharmacy. 2005. The New York Times, April 3, 12. Pellegrino, E. D., and D. C. Thomasma. 1993. The virtues in medical practice. New York: Oxford University Press. Pitts, L., Jr. 2005. Find another job: Pharmacists who refuse to sell contraceptives should take a hike. Pittsburgh Post-​Gazette, April 22, 13. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1983. Deciding to forego life-​sustaining treatment: Ethical, medical, and legal issues in treatment decisions. Washington, DC: U.S. Government Printing Office. President’s Council on Bioethics. 2008. Controversies in the determination of death. Washington, DC: Author. Rhodes, R. 2006. The ethical standard of care. American Journal of Bioethics 6(2):76–​78. Savulescu, J. 2006. Conscientious objection in medicine. British Medical Journal 332:294–​297. Schneider, J. S., and S. Levin. 1999. Uneasy partners: The lesbian and gay health care community and the AMA. JAMA 282(13):1287–​1288. Shewmon, D. A. 2009. Brain death: Can it be resuscitated? The Hastings Center Report 39:18–​24.

108   Mark R. Wicclair Skorupski, J. 2010. Conscience. In The Routledge companion to ethics, edited by John Skorupski. London: Routledge, 550–​561. Stein, R. 2005. Pharmacists’ rights at front of new debate: Because of beliefs, some refuse to fill birth control prescriptions. Washington Post, March 28, 1. Sulmasy, D. P. 2008. What is conscience and why is respect for it so important? Theoretical Medicine and Bioethics 29:135–​149. Troug, R. D., and F. G. Miller. 2010. Counterpoint: Are donors after circulatory death really dead, and does it matter? No and not really. Chest 138(1):16–​18. Van Riper, K. K., and W. L. Hellerstedt. 2005. Emergency contraceptive pills: Dispensing practices, knowledge and attitudes of South Dakota pharmacists. Perspectives on Sexual and Reproductive Health 37(1):19–​24. Wallace, J. L., J. Wu, J. Weinstein, D. W. Gorenflo, and M. D. Fetters 2004. Emergency contraception:  Knowledge and attitudes of family medicine providers. Family Medicine 36(6):417–​21. Wear, S., S. LaGaipa, and G. Logue. 1994. Toleration of moral diversity and the conscientious refusal by physicians to withdraw life-​sustaining treatments. Journal of Medicine and Philosophy 19(2):147–​159. Wicclair, M. 2013. Conscience. In International encyclopedia of ethics, edited by Hugh LaFollette. Hoboken, 1009-​1020. NJ: John Wiley & Sons. Wicclair, M. R. 2011. Conscientious objection in health care:  An ethical analysis. Cambridge: Cambridge University Press. Youngner, S. J., R. M. Arnold, and R. Schapiro. 1999. The definition of death: Contemporary controversies. Baltimore: Johns Hopkins University Press.

Chapter 7

C ontinu ou s Se dat i on at the End of L i fe Sigrid Sterckx and Kasper Raus

Introduction It is a well-​known fact that the way we die has undergone considerable changes in the last century or so. Increased knowledge of what constitutes a healthy lifestyle and advances in both preventive and curative medicine have resulted in an increased life expectancy. Regarding the cause of death, there has been a shift from infectious diseases to more degenerative conditions such as cancer and Alzheimer’s, meaning that an increasing number of people experience a longer dying trajectory (see, e.g., Seale, 2000, for an interesting overview of these changes). Research indicates that in Belgium two-​thirds of all patients who died in 2007 did not die suddenly (Bilsen et al., 2009), while in the Netherlands in 2010 an end-​of-​life decision was taken in 58% of all deaths (Onwuteaka-​Philipsen et al., 2012). Some commentators suggest that besides changing the manner in which we die, we are also changing our attitudes toward the end of life and perhaps our entire ars moriendi (Holm, 2013; Battin, 2013). There has, for example, been an attitude shift from “cure” to “care” for terminal patients who can no longer be cured (e.g., Saunders, 2001). Another shift has to do with the fact that although it is sometimes claimed that suffering at the end of life is meaningful (e.g., Sacred Congregation for The Doctrine of the Faith, 1980), it is now often recognized that “[n]‌ot everyone finds a ‘salvific meaning’ in suffering” (Smith, 1998: 382). Furthermore, many people no longer consider pain to be an inevitable element of dying, and pain management is increasingly being recognized as a basic human right (e.g., Brennan et al., 2007). Treating and controlling suffering (both physical and mental) has become a cornerstone of modern palliative care, and due to advances in pain medicine, a pain-​free death has become possible for many patients (e.g., Seymour et al., 2005).

110    Sigrid Sterckx and Kasper Raus However, the need remains great. Research by Fainsinger et al. (2000) in four palliative care wards has shown that more than 90% of patients required symptom control or management. In more recent research, Steindal et al. (2011: 771) noted that “pain is a highly prevalent symptom among dying hospitalized patients.” While much suffering can be relieved nowadays, patients’ suffering can still be so grave as to require far-​reaching interventions to relieve or control it. The administration of high dosages of pain-​relieving drugs to severely suffering patients, even doses so high that life-​shortening may occur, is now widely accepted. In most countries it is also allowed to withhold or withdraw life-​sustaining treatments when it is clear that they are futile and/​or on patient request. More controversial are practices such as euthanasia (legal in the Netherlands, Belgium, and Luxembourg) and physician-​assisted suicide (legal in, for example, the Netherlands, Switzerland, and the U.S. states of Oregon and Washington). In this chapter, however, we focus on a different medical practice for relieving intense suffering, namely continuous sedation at the end of life (hereafter “continuous sedation”), which involves the administration of sedative drugs so as to reduce a patient’s consciousness until he or she dies and to ensure that he or she no longer experiences any suffering.

Seeming Consensus and Guidelines There seems to be some degree of consensus that continuous sedation can be an acceptable way to relieve severe suffering at the end of life. Witness to that consensus is a study by Simon et al. (2007: 1) among German experts in medical ethics where “98% of the respondents regarded terminal sedation in dying patients with treatment-​refractory physical symptoms as acceptable.” The practice of continuous sedation also raises considerably less legal debate than euthanasia and physician-​assisted suicide. Though the legality of continuous sedation is sometimes questioned (e.g., Gevers, 2004; Delbeke, 2013), most commentators believe that continuous sedation is legal and even covered by current legal frameworks that allow the administration of high doses of medication even at the risk of life-​shortening. This can, for example, be seen in the United States. In her concurring opinion in Washington v. Glucksberg, an influential ruling of the U.S. Supreme Court, Justice O’Connor noted that she did not believe assisted suicide should be legalized as a patient who is suffering from a terminal illness and who is experiencing great pain has no legal barriers to obtaining medication, from qualified physicians, to alleviate that suffering, even to the point of causing unconsciousness and hastening death. (Washington v. Glucksberg, 1997: 736–​737)

With this ruling, the U.S. Supreme Court has been said to have explicitly endorsed continuous sedation (Orentlicher, 1997, 2013). Existing guidelines also see few problems with continuous sedation, as for example the Dutch guideline that maintains that

Continuous Sedation at the End of Life    111 it is “normal medical practice” (KNMG, 2009) and the Code of Medical Ethics of the American Medical Association (2008: 6), which goes as far as to claim that “It is the ethical obligation of a physician to offer palliative sedation to unconsciousness as an option for the relief of intractable symptoms.” Regarding how continuous sedation should be performed, many guidelines have been drafted and continue to be drafted. Examples include an influential Dutch guideline, issued by the Royal Dutch Medical Association (KNMG, 2009), a framework drafted by the European Association for Palliative Care (EAPC; Cherny and Radbruch, 2009), a Belgian guideline (Federatie Palliatieve Zorg Vlaanderen, 2012), an American guideline (American Medical Association, 2008), and a framework drafted by the Canadian Society for Palliative Care Physicians (Dean et al., 2012). The Dutch guideline is unique in having actual legal ramifications, and research has shown this guideline to be well known and widely followed by Dutch physicians according to their own reports (Hasselaar et al., 2009). All other guidelines are considered recommendations, and it is unknown to what degree they are known and followed in practice. Although at first sight these guidelines maintain that continuous sedation can be an acceptable practice, they differ in important respects when it comes to just how continuous sedation should be performed. For example, the Dutch guideline allows continuous sedation to be used for patients with a life expectancy of no more than two weeks, while the EAPC guideline recommends sedation only for patients in their last days or hours of life. The Dutch guideline also allows continuous sedation to be used for patients who are experiencing only existential suffering, something that the American Medical Association’s guideline clearly forbids. Existing guidelines also differ in their recommendations on how continuous sedation should be monitored once it is started; some guidelines recommend basic patient observation, while others recommend the use of sedation scales. Accordingly, underlying the seeming consensus is a considerable amount of debate. Indeed, there is hardly a single aspect of continuous sedation that is not under discussion. At the very heart of the debate lie the issues of terminology and definition.

Terminology and Definition Many terms have been used for referring to continuous sedation. “Terminal sedation” was the term that was coined first by Enck (1991) in an influential article that is sometimes claimed to have started the debate on continuous sedation (e.g., Materstvedt and Bosshard, 2009). Although this term continues to be used, it has been criticized for suggesting that continuous sedation has to do with “terminating” a patient’s life (Morita et al., 2002). An alternative term that is now widely used is “palliative sedation,” but this in turn has been criticized for being overly “euphemistic” and for begging the ethical question by suggesting that continuous sedation is by its very definition part of palliative care (van Delden, 2007). In research contexts, it is common to label the practice “continuous (deep) sedation,” as this term has the advantage of being purely descriptive (Rietjens et al., 2008).

112    Sigrid Sterckx and Kasper Raus A more fundamental issue is that of definition. As mentioned earlier, there is by no means a consensus on how continuous sedation should be defined, and this has resulted in a large number of definitions (e.g., Morita et al., 2001, 2002; Jones, 2013). Many of the existing definitions differ significantly from each other, and there is little that can be said to be shared by all. Some definitions are particularly broad and inclusive. A nice example comes from van Delden (2007: 187) who defines sedation as “sedation until death follows,” thereby capturing a broad spectrum of practices. Many other definitions, however, are more exclusive, for example limiting what is considered to be continuous sedation to only those cases where sedation is achieved intentionally. Tännsjö (2004: 15) clearly does this when he defines continuous sedation as “a procedure where through heavy sedation a terminally ill patient is put into a state of coma, where the intention of the doctor is that the patient should stay comatose until he or she is dead.” Sedation as a side-​effect (e.g., due to high dosages of a pain-​relieving agent such as morphine) is thereby excluded and not considered to be continuous sedation. Other authors go further and also include within their definitions the indications for which continuous sedation should be used. This is well illustrated by a definition from Rousseau (2000: 164) that limits continuous sedation to those cases where sedation is administered intentionally and for refractory symptoms: “PS [palliative sedation] can be defined objectively as the intention of purposely inducing and maintaining a sedated state, but not deliberately causing death, in specific clinical circumstances complicated by refractory symptoms.” Some definitions further mention that continuous sedation is, by definition, used only on patients with a short life expectancy. The American Medical Association (2008: 6), for example, defines sedation to unconsciousness as “the administration of sedative medication to the point of unconsciousness in a terminally ill patient.” The choice of which elements to include in the definition is, of course, far from neutral. Some commentators, such as van Delden, argue against narrow and exclusive definitions. The indications for which continuous sedation can be used as well as a patient’s life expectancy play a role in judging the practice’s ethical justifiability, so by including these as elements in the definition one is in a sense steering the ethical debate towards acceptance. Van Delden argues that we should employ a broad definition on the basis of which we can then start the ethical debate on what use of continuous sedation is acceptable. Others argue for the exact opposite and aim at including even more elements in the definition. Jones (2013), for example, pleads for a manifold definition of sedation containing as many as seventeen elements that are intended to capture every possible form of continuous sedation.

State of the Art Despite disputes over the precise definition, an increasing amount of research into the practice of continuous sedation gives us valuable insights into how often it is used in

Continuous Sedation at the End of Life    113 various countries and how it is performed. Research has shown that considerable differences exist between countries, for example between Belgium, the Netherlands, and the UK (Anquinet et al., 2012), as well as within the same country (e.g., Chambaere et al., 2010; Seale, 2010). In a 2008 review of all published research on continuous sedation, it became clear that the reported prevalence of continuous sedation in palliative care units differs significantly from study to study, ranging from 3.1% of deaths in one study to 51% in another (Claessens et al., 2008). Determining the precise rate of incidence for continuous sedation is highly difficult, due in part to differences in population and methodology but due mostly to differences in definition. Using a broader definition often results in a higher incidence being found than would be found using a narrow one. A Belgian study in 2007 using a broad definition reported the practice to occur in 14.5% of all deaths (Chambaere et al., 2010). Other research, using a more narrow definition, reported an incidence of 7.5% between 2004 and 2005 (Claessens et al., 2011). In the Netherlands, continuous sedation was reported as occurring in 12.3 % of all deaths in 2010 (Onwuteaka-​Philipsen et al., 2012), while for the UK in 2007–​2008 the incidence was estimated at 16.5% of all deaths (Seale, 2009). Somewhat older data shows that, in 2001–​2002, continuous sedation occurred in 2.5% of deaths in Denmark and 8.5% of deaths in Italy (Miccinesi et al., 2006). For many other countries, such as the United States, the incidence of continuous sedation is currently unknown. Despite difficulties in determining the exact incidence of continuous sedation, it is clear from research that the practice is far from uncommon. The reasons why this practice is so common are unknown. It has been suggested that continuous sedation fits a commonly held conception of what it means to have a “good death,” as the practice resembles or even mimics that image, namely of dying peacefully in one’s sleep (e.g., Seymour, 1999; Seymour et al., 2007). Death under continuous sedation is also often described as a “natural death,” something that is then identified as being an advantage over practices such as euthanasia or physician-​assisted suicide where death is clearly not natural. Such a portrayal has been criticized (Raus et al., 2012). The published research also gives us insight into the way in which continuous sedation is currently performed as well as the types of patients on which it is used. Studies suggest that continuous sedation is mostly used on patients with cancer or cardiovascular diseases (Anquinet et al., 2012) and generally only a short time before death occurs. In a Dutch study performed in 2005, 94% of all patients receiving continuous deep sedation died within one week after initiation (Rietjens et al., 2008). In accordance with recommendations made in guidelines (e.g., KNMG, 2009; Federatie Palliatieve Zorg, 2012) and by experts (e.g., Porta-​Sales, 2013), the most commonly used drugs for continuous sedation are benzodiazepines (particularly midazolam), often in combination with pain-​relief medication, such as opioids. In some cases, only opioids are used to initiate and maintain continuous sedation (Chambaere et  al., 2010). However, many experts consider opioids to be analgesics rather than sedatives, so continuous sedation using only morphine, for example, is often discouraged, and the Dutch National guideline even “regards the use of morphine as a sedative as bad practice” (KNMG, 2009: 11).

114    Sigrid Sterckx and Kasper Raus Not infrequently, continuous sedation is combined with withdrawing or withholding of artificial nutrition and hydration (ANH), but how often this occurs varies strongly between countries and settings. In Italy in 2001–​2002, ANH was withheld in 35% of all continuous sedation cases, whereas for Denmark in the same period the figure was 64% (Miccinesi et al., 2006). Given that the administration of food and fluids is such a controversial issue, this variation is not wholly unexpected. Many current guidelines discuss the issue of nutrition and hydration, and most guidelines recommend viewing the decision to withdraw or withhold ANH as a completely separate decision from the decision to sedate. Accordingly, when ANH is withdrawn or withheld, this has to be justified in its own right. Arguments commonly given for the stopping of all ANH for continuously sedated patients are its futility for these dying patients and the potential harms it might cause to the patients. These arguments are subject to criticism (Holm, 2013). Finally, a remarkable finding is that although the importance of patient consent is stressed in all sedation guidelines, and autonomy is frequently invoked as a justification for continuous sedation (discussed later), it is far from always obtained. In Belgium in 2007, there was a request for or a consent to continuous deep sedation in only 30% of all continuous sedation cases (Chambaere et  al., 2010). Similar research from the Netherlands shows that although continuous sedation is more frequently discussed with the patient, it is far from always discussed (Rietjens et al., 2004). This is a surprising finding, especially since the Dutch study looked only at continuous deep sedation without ANH (i.e., the most extreme form of continuous sedation). It is sometimes suggested that the low incidence of actual patient consent might be due to the fact that continuous sedation is considered only for patients with a very short life expectancy who are in a bad physical condition and who are therefore more likely to have become incompetent or less than fully competent when continuous sedation is considered.

Commonly Invoked Justifications As mentioned previously, some degree of consensus exists that continuous sedation can be an acceptable way to relieve unbearable suffering at the end of life. Regarding the ethical acceptability of the practice, the same justifications are commonly invoked by various commentators, but they are also often questioned.

The Doctrine of Double Effect Perhaps the most commonly invoked justification for continuous sedation is the doctrine of double effect. This doctrine distinguishes the intended effects of one’s action from its foreseen but unintended side-​effects and states that one may bring about as a side-​effect certain bad effects that one could not bring about intentionally. The doctrine, for example, states that it is allowed to administer to a patient high doses of

Continuous Sedation at the End of Life    115 pain-​relieving agents even when there is a risk of shortening life, provided that one’s intention is to relieve pain (a good effect) and that any potential life shortening (a bad effect) is a mere side-​effect. Euthanasia, on the other hand, is forbidden by the doctrine of double effect, as in euthanasia, whereby a physician intends to kill the patient (a bad effect) in order to achieve pain relief (a good effect). Continuous sedation is sometimes suggested to be more similar to the administration of pain relief than to euthanasia and therefore to be allowed by the doctrine of double effect. As a legal principle, the doctrine of double effect is sometimes recognized in relation to continuous sedation. A double-​effect type of reasoning was famously used in Britain for acquitting Dr. John Bodkin Adams, who was accused of killing an eighty-​one-​year-​ old patient using high doses of pain relief. The judge in this case noted that a doctor “is entitled to do all that is proper and necessary to relieve pain and suffering, even if the measures he takes may incidentally shorten life” (cited in Davies, 1998: 347). Although this trial did not deal with continuous sedation as such, this idea became an established legal principle in England and is frequently said to apply to continuous sedation. The U.S. Supreme Court made explicit reference to the doctrine of double effect as a justification for continuous sedation when in Vacco v. Quill it stated that “the concept of sedating pharmacotherapy is based on informed consent and the principle of double effect” (Vacco v. Quill, 1997: 807). To mention another example, in 2012 in the Netherlands a male nurse was put on trial for attempted murder for using continuous sedation on his elderly father. The nurse was acquitted on the grounds that he “did not have the intention to end his father’s life, nor did he intend to harm his father’s health, but instead acted with the intention of lessening his suffering” (Rechtbank’s Gravenhage Court, 2012; authors’ translation from Dutch). The Dutch court thus clearly used a double-​effect type reasoning. Nevertheless, the legal value of the doctrine in cases of continuous sedation at the end of life is sometimes questioned (e.g., Delbeke, 2013), and some others acknowledge the doctrine’s weaknesses but propose to retain it as a form of compromise (Huxtable and Horn, 2013). Apart from its use in legal contexts, the double-​effect doctrine is often referred to as an ethical justification for continuous sedation, justifying sedation while forbidding euthanasia and physician-​assisted suicide. However, the application of the doctrine of double effect to continuous sedation is also often criticized. Thus its application is far from simple as it has different interpretations and there are different ways of applying it to continuous sedation (e.g., Marquis, 1991; Raus et al., 2013). Moreover, as the doctrine of double effect forbids every action where a bad effect is used as a means to achieve a good effect and as some authors consider the permanent removal of consciousness to constitute a harm in itself (e.g., Holahan et al., 2013; van Delden, 2013), double-​effect reasoning would forbid continuous sedation for using a bad effect (removing consciousness) to achieve a good effect (pain relief). Finally, the line between what one intends to achieve and what one merely allows to occur as side-​effect is often very thin. First, physicians may have multiple intentions (Quill et al., 1997), and second, empirical research shows that physicians using continuous sedation often do intend to hasten death (Chambaere et al., 2010). For these not uncommon cases, double effect cannot serve as a justification.

116    Sigrid Sterckx and Kasper Raus

Last Resort and Refractory Suffering It seems generally agreed that continuous sedation is only indicated when a patient suffers unbearably due to refractory symptoms. Since a refractory symptom is a symptom that cannot be relieved using standard ways of treatment, this means that continuous sedation can be indicated only as a last resort (i.e., when no other means of treating a patient’s symptoms exist). As such, continuous sedation is sometimes justified by claiming that it concerns a far-​reaching practice that is the only way to relieve otherwise intractable suffering and “should only be implemented in the rarest of circumstances” (American Medical Association, 2008: 18–​19). Qualitative research indicates that some physicians and nurses do indeed justify their own involvement in continuous sedation by stating that it was a last resort and that, therefore, there was no alternative choice (Raus et al., 2014). This idea is supported by a quantitative study involving a large number of continuous sedation cases in 2007. In 82% of these cases, the responsible physician indicated that, according to him or her, there was no alternative to continuous sedation (Chambaere et al., 2010). Whether continuous sedation is in fact used as a “last resort” is at times questioned in view of its high incidence. In view of the considerable advances that have been made in palliative care, it seems unlikely that there was no medical alternative for continuous sedation reported in 14.5% of all deaths in Belgium (Chambaere et al., 2010) and 12.3% of all deaths in the Netherlands (Onwuteaka-​Philipsen et al., 2012). However, whether a symptom is refractory or not depends on whether or not a standard method of treating that symptom is available, and it can thus vary, for example, from setting to setting and even from physician to physician. A certain method of treatment might, for example, not be available in the home-​care setting and yet available in a hospital setting, making some symptoms refractory at home but not in the hospital. A technique might also not be known or mastered by the patient’s physician, causing a symptom that would not be refractory with a different physician to be de facto refractory. In the light of these issues, it becomes next to impossible to estimate what would be a “normal” level of refractory symptoms.

Autonomy and Patient Consent The practice of continuous sedation is sometimes justified by referring to the autonomous request of a patient not wanting to suffer anymore. Indeed, professional guidelines emphasize that continuous sedation should be preceded by patient consent (when this is possible) or by a surrogate such as an advance directive or proxy consent. Available qualitative research also indicates that obtaining patient consent often makes it easier for physicians and nurses to initiate continuous sedation (Raus et al., 2014). Research findings (discussed earlier), however, show that patient consent is not always sought and obtained by the physician, even in those cases where consent could be obtained. The patient’s family, by contrast, is almost always involved either by consenting to continuous sedation or even by requesting it (Chambaere et al., 2010).

Continuous Sedation at the End of Life    117 Some commentators are critical of whether continuous sedation indeed respects autonomy (e.g., Holm, 2013; Battin, 2013). They claim that although continuous sedation is often justified by referring to autonomy, and patients can make a request for continuous sedation, it is the physician who determines whether the conditions for receiving it are fulfilled (i.e., whether the patient has a short enough life expectancy and whether his or her symptoms are no longer treatable). Moreover, continuous sedation leaves a patient sedated and thus unable to further exercise her autonomy. In this regard, Margaret Battin (2013: 249) writes Terminal sedation involves the administration of pain medication in the hands of others, and, once the process has begun, the patient no longer has a way of controlling the course of events on the way to death: what had been a gain in patient autonomy is erased.

As a result, continuous sedation might in fact be less autonomy driven than are euthanasia and physician-​assisted suicide and more concerned with the potentially paternalistic goal of “relieving pain.” Continuous sedation is thus criticized for being an overly medicalized or technicalized way of dealing with health-​related problems. One might respond by saying that although a patient may no longer be autonomous, a patient undergoing continuous sedation is at least free from suffering. However, in most people’s conception of what constitutes a good death, being free from pain is only one element. A study involving seriously ill patients and their family, physicians, and care providers showed that their idea of a good death also involves elements such as “being kept clean,” “having someone who will listen,” and “maintaining one’s dignity,” all of which were rated higher than “being free of pain” (Steinhauser et al., 2000). Some patients might therefore prefer being conscious and in pain over being unconscious and free from suffering (e.g., Heyse-​Moore, 2003). In conclusion, while continuous sedation at the end of life might be highly successful at providing a pain-​free death, it is sometimes claimed to run the risk of ignoring the other elements of what patients perceive as relevant for achieving a good death. If continuous sedation really is autonomy driven, we should also respect a patient’s wish to remain conscious for as long as possible.

Proportionality Continuous sedation is also frequently justified by stating that it is a proportional reaction to grave suffering. Indeed, all guidelines of which we are aware take care to stress that physicians administering continuous sedation should do so proportionally in relation to the symptoms a patient is suffering from (i.e., making sure a patient is neither too lightly nor too deeply sedated). Deep sedation is then indicated only for severe suffering. In view of the importance given to the proportionality condition, it is striking to note that very few guidelines deal with the issue of how proportionality (i.e., depth of sedation) should be monitored. This is particularly noteworthy as it is known from

118    Sigrid Sterckx and Kasper Raus anaesthesia research that some patients can appear to be calm and peaceful and yet be fully aware, suggesting the possibility that some seemingly sedated dying patients are actually suffering (Mashour, 2010). When continuous sedation is monitored, the most common means is by basic clinical assessment, where the physician watches whether the patient is peaceful or whether he or she shows signs of being awake. This strategy is recommended by a Belgian guideline (Federatie Palliatieve Zorg Vlaanderen, 2012) and a Canadian framework for continuous sedation (Dean et al., 2012). However, this reduces “consciousness” to “responsiveness to stimuli,” and we know from research that these are only imperfectly correlated (Noreika et al., 2011). A different technique for measuring consciousness (and thus proportionality) is using so-​called sedation scales. This is recommended by, for example, the Dutch national guideline (KNMG, 2009) and the EAPC guideline (Cherny and Radbruch, 2009). When using sedation scales, physicians administer stimuli (e.g., calling the patient’s name, prodding, shaking, a painful trapezius squeeze, etc.) and score the patient’s reaction. A global score is then given to determine the depth to which a patient is sedated. However, these scales, like basic clinical assessment, rely on patients responding to stimuli, which can often be misleading (e.g., Alkire et al., 2008). Moreover, using scales can be invasive, as physicians are often required to administered stimuli that would be painful to a conscious patient. An alternative might be to use an EEG to determine a patient’s degree of consciousness. However, an EEG is complicated and also produces much data irrelevant to consciousness; thus monitors have been developed for anaesthesia that reduce the complex EEG to a more simple value. One example is the BIS monitor, which converts a single EEG to a readily usable number between zero (inert) and 100 (fully awake) indicating a patient’s consciousness. BIS monitoring involves placing an electrode patch on the patient’s forehead but requires no further stimulation. Moreover, BIS monitors, unlike sedation scales, allow for continual monitoring of consciousness. This technique is not used in palliative care, but more research seems to be needed if the requirement of proportionality is to have real substance, as the currently used monitoring techniques are not necessarily the most effective or the least invasive. A research project on the use of BIS monitors on continuously sedated patients is currently taking place at the University Hospital in Ghent.

Debates on Continuous Sedation Continuous sedation is at times criticized for being a slow way to perform euthanasia and is therefore sometimes referred to as “slow euthanasia,” a term coined by Billings and Block (1996). This idea has led to fierce debates over whether or not continuous sedation can or does shorten a patient’s life. Many authors and guidelines maintain that, when properly performed, continuous sedation does not shorten life, and various studies have been performed that seem to support this claim (e.g., Ventafridda et al., 1990;

Continuous Sedation at the End of Life    119 Sykes and Thorns, 2003; Morita et al., 2005). When comparing the time until death of patients receiving continuous sedation with patients not receiving continuous sedation, these studies found no significant difference and therefore concluded that life-​ shortening was absent (or minimal) in continuous sedation. Others question the results of these studies by arguing that their methodology is such that they are not absolutely conclusive, as there might be important differences between sedated and nonsedated patients that are missed and that invalidate the research (Stone et al., 1997). Moreover, in these types of studies, the definition one uses is crucial and likely to affect one’s results. If one looks only at cases of continuous sedation where life is less likely to be shortened (e.g., sedation of patients in their last hours or days of life), it is not surprising that no life-​shortening is seen. Cases where life-​shortening is most likely to occur are often missing from empirical research into the potential life-​shortening effect of continuous sedation. Moreover, showing that there is no statistically significant difference between sedated and nonsedated patient groups does not necessarily mean that life cannot be or have been shortened in individual cases. Therefore, although some research has been done, this has not silenced the debate on whether or not continuous sedation shortens life. Another debated issue is whether the use of continuous sedation should be restricted to patients with a very short life expectancy. Although all guidelines recommend continuous sedation only for patients who are in their last days of life, some commentators (e.g., Cellarius, 2008) argue that this “imminence condition,” as Cellarius calls it, is rarely argued for and does not follow from the standard justifications of continuous sedation. It has been suggested that limiting the use of continuous sedation to imminently dying patients serves only to prevent life-​shortening from occurring (Janssens et al., 2012). If this is indeed the restriction’s purpose, it is not certain that it succeeds, as it has been suggested that life-​shortening is also possible for patients who are sedated for a shorter period of time. Moreover, requiring the treating physician to estimate the patient’s life expectancy before starting continuous sedation can be dangerous, as physicians are less than optimal at determining how long it will be before a patient dies (Wilson et al., 2005; Clarke et al., 2009). Physicians might therefore underestimate life expectancy (thereby sedating patients for a longer period of time than initially intended) or overestimate it (thereby denying continuous sedation to patients who do in fact meet the conditions for receiving it). Finally, although cancer patients often follow a more or less predictable disease trajectory (making it easier to estimate life expectancy), other diseases, such as cardiovascular diseases or degenerative diseases, follow less predictable trajectories (Murray et al., 2005). As such, patients suffering from diseases other than cancer might be disadvantaged when it comes to receiving continuous sedation. A related debate centers around ANH. Continuous sedation can occur with or without the administration of ANH. The latter form of continuous sedation is, of course, more controversial. As patients who are continuously sedated are no longer able to take in food and fluids themselves (due to reduced consciousness), withholding nutrition and hydration may cause them to die from the lack thereof, making continuous sedation similar to euthanasia. However, since all guidelines recommend using continuous

120    Sigrid Sterckx and Kasper Raus sedation only in a patient’s last days of life, it is usually claimed that patients receiving sedation without ANH will die before dehydration or starvation can have an effect (KNMG, 2009). It has been suggested that the main reason the Dutch guideline holds to the two-​week life expectancy as an important criterion is that this is the time it takes for a patient to die from lack of nutrition and hydration (Janssens et al., 2012). In this respect, however, it is sometimes argued that there might still be a life-​shortening effect even when the patient is sedated for only a few days. As mentioned previously, physicians are often bad at estimating life expectancy, making it possible for patients to live long enough for dehydration and starvation to have an effect. Second, though it might take two weeks for lack of food and fluids to be the sole cause of death, it might still have some effect before this time (van Delden, 2013). Commentators such as Holm (2013) argue that if proportionality is indeed what justifies continuous sedation and if pain relief can be achieved using continuous sedation with food and fluids administered, then one is obliged to use sedation with ANH as it is a less invasive way of obtaining the same effect. Withholding ANH can then only be justified when providing food and fluids is deemed futile. However, cases where this is the case are few, as Holm (2013: 233) argues: “If withdrawal gives rise to symptoms, then continuation [of ANH] is clearly not futile, because it would prevent the symptoms from occurring.” A final debate concerns existential or psychological suffering as an indication for continuous sedation. While continuous sedation for somatic suffering is generally accepted, opinions on sedation for purely psychological or existential suffering differ greatly. Available research indicates that the most common indications for continuous sedation in practice today are all somatic symptoms causing physical suffering, such as, for example, pain, delirium, and respiratory distress (e.g., Chater et al., 1998; Claessens et al., 2008). Other research among German experts in medical ethics has shown that the acceptance of continuous sedation for “mental suffering” is significantly lower than the acceptance of continuous sedation for physical suffering (Simon et al., 2007). Qualitative research involving physicians and nurses also indicates that some physicians would find it difficult to start up continuous sedation for patients with solely existential or psychological suffering, and some would not even consider using it (Raus et al., 2014). Some sedation guidelines, such as the EAPC and the Dutch National guidelines, do allow continuous sedation for such suffering but at the same time recommend that it be used only on patients with a very short life expectancy (no more than fourteen days) so that the patients receiving continuous sedation are also highly likely to also have at least some somatic suffering. Thus while these guidelines do allow it, the additional requirement that patients receiving continuous sedation should have a short life expectancy makes it unlikely to be much used in practice. Other guidelines do not allow continuous sedation for patients with existential suffering. An example is the Code of Ethics of the American Medical Association (2008: 7), which notes that “Palliative sedation is not an appropriate response to suffering that is primarily existential ( … ). Existential suffering is better addressed by other interventions.”

Continuous Sedation at the End of Life    121 Thus while existential or psychological suffering is often a factor in choosing continuous sedation, it is not surprising that sedation is rarely used for patients with only psychological suffering, as this is either not allowed/​recommended or allowed/​recommended only within a narrow margin.

Conclusion Continuous sedation at the end of life is often seen as an acceptable way to relieve severe suffering at the end of life and is often justified by referring to the doctrine of double effect, last-​resort considerations, autonomy, and proportionality. Nevertheless, many aspects of continuous sedation are still the topic of intense debate, for example whether the practice shortens life or whether it should be used only for imminently dying patients suffering from symptoms of somatic origin. A large part of the debate relates back to the lack of agreement on how to label and define continuous sedation, which allows divergent views on when continuous sedation is at issue. For some, continuous sedation refers to the use of a proportionate dose of sedative medication on dying patients suffering from severe refractory symptoms. Seen from this perspective, continuous sedation appears less problematic, as life-​shortening is less likely and any other use of sedation (e.g., on patients with a longer life expectancy) can be dismissed as not even deserving the label of continuous sedation. Others consider continuous sedation to refer to a broader spectrum of cases, such as those in which a disproportionate dose of medication is given, or where continuous sedation is used on patients with a longer life expectancy and no artificial nutrition and hydration are given. If one considers such cases to be genuine examples of continuous sedation, debates such as that on life-​shortening come to the forefront.

References Alkire, M. T., A. G. Hudetz, and G. Tononi. 2008. Consciousness and anesthesia. Science 322(5903):876–​880. American Medical Association, Council on Ethical and Judicial Affairs. 2008. Sedation to Unconsciousness in End of Life Care. CEJA Report 5-​A-​08.Chicago:  American Medical Association. http://​www.ama-​assn.org/​resources/​doc/​code-​medical-​ethics/​2201a.pdf. Anquinet, L., J. Rietjens, C. Seale, J. Seymour, L. Deliens, andA. van der Heide. 2012. The Practice of Continuous Deep Sedation Until Death in Flanders (Belgium), The Netherlands, and the U.K.:  A  Comparative Study. Journal of Pain and Symptom Management 44(1):33–​43. Battin, M. 2013. Terminal sedation: recasting a metaphor as the ars moriendi changes. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 240–​249. Billings, J. A., and S. D. Block. 1996. Slow euthanasia. Journal of Palliative Care 12(4):21–​30.

122    Sigrid Sterckx and Kasper Raus Bilsen, J., J. Cohen, K. Chambaere, P. Pousset, B. D. Onwuteaka-​Philipsen, F. Mortier, and L. Deliens. 2009. Medical end-​of-​life practices under the euthanasia law in Belgium. The New England Journal of Medicine 361(11):1119–​1121. Brennan, F., D. B. Carr, and M. Cousins. 2007. Pain management: a fundamental human right. Anesthesia & Analgesia 105(1):205–​221. Cellarius, V. 2008. Terminal sedation and the imminence condition. Journal of Medical Ethics 34:69–​72. Chambaere, K., J. Bilsen, J. Cohen, J. A. Rietjens, B. D. Onwuteaka-​Philipsen, F. Mortier, and L. Deliens. 2010. Continuous deep sedation until death in Belgium: a nationwide survey. Archives of Internal Medicine 170(5):490–​493. Chater, S., R. Viola, J. Paterson, and V. Jarvis. 1998. Sedation for intractable distress in the dying—​a survey of experts. Journal of Palliative Medicine 12:255–​269. Cherny, N. I., and L. Radbruch. (Board of the European Association for Palliative Care). 2009. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care. Journal of Palliative Medicine 23(7):581–​593. Claessens, P., J. Menten, P. Schotsmans, and B. Broeckaert. 2008. Palliative sedation: a review of the research literature. The Journal of Pain and Symptom Management 36(3):310–​333. Claessens, P., J. Menten, P. Schotsmans, and B. Broeckaert. 2011. Palliative sedation, not slow euthanasia: a prospective, longitudinal study of sedation in Flemish palliative care units. The Journal of Pain and Symptom Management 41(1):14–​24. Clarke, M. G., P. Ewings, T. Hanna, L. Dunn, T. Girling, and A. L. Widdison. 2009. How accurate are doctors, nurses and medical students at predicting life expectancy? The European Journal of Internal Medicine 20(6):640–​644. Davies, M. 1998. Textbook on Medical Law. 2nd ed. London: Blackstone. Dean, M., V. Cellarius, B. Henry, D. Oneschuk, and L. Librach. 2012. Framework for continuous palliative sedation therapy in Canada. Journal of Palliative Medicine 15(8):870–​879. Delbeke, E. 2013. The legal permissibility of continuous deep sedation at the end of life: a comparison of laws and a proposal. In:  S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 132–​148. Enck, R. E. 1991. Drug-​induced terminal sedation for symptom control. The American Journal of Hospice & Palliative Medicine 8(5):3–​5. Fainsinger, R. L., A. Waller, M. Bercovici, K. Bengston, W. Landman, M. Hoskings, J. M. Nunez-​ Olarte, and D. deMoissac. 2000. A multicenter international study of sedation for uncontrolled symptoms in terminally ill patients. Journal of Palliative Medicine 14(4):257–​265. Federatie Palliatieve Zorg Vlaanderen 2012. Richtlijn Palliatieve Sedatie. http://​www.pallialine. be/​accounts/​129/​docs/​richtlijn_​palliatieve_​sedatie.pdf. Gevers, S. 2004. Terminal sedation:  a legal approach. The European Journal of Health Law 10(4):359–​367. Hasselaar, J., S. Verhagen, A. Wolff, and Y. Engels. 2009. Changed patterns in Dutch palliative sedation practices after the introduction of a national guideline. Archives of Internal Medicine 169(5):430–​437. Heyse-​Moore, L. 2003. Terminal restlessness and sedation:  a note of caution. Journal of Palliative Medicine 17: 469. Holahan, T., T. Carroll, C. Gonzalez, and T. E. Quill. 2013. Palliative sedation, consciousness and personhood. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​ Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 202–​217.

Continuous Sedation at the End of Life    123 Holm, S. 2013. Terminal sedation and euthanasia: the virtue in calling a spade what it is. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 228–​239. Huxtable, R., and R. Horn. 2013. Continuous deep sedation at the end of life: balancing benefits and harms in England, Germany and France. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 160–​176. Janssens, R., J. J. M. van Delden, and G. A. M. Widdershoven. 2012. Palliative sedation: not just normal medical practice. Ethical reflections on the Royal Dutch Medical Association’s guideline on palliative sedation. Journal of Medical Ethics 38(11):664–​668. Jones, D. A. 2013. Death by equivocation: a manifold definition of terminal sedation. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives, Cambridge: Cambridge University Press: pp. 47–​64. KNMG (Koninklijke Nederlandse Maatschappij ter bevordering van de Geneeskunde. 2009. KNMG-​ Guideline for Palliative Sedation (2009). http://​knmg.artsennet.nl/​Publicaties/​ KNMGpublicatie/​Guideline-​for-​palliative-​sedation-​2009.htm. Marquis, D. B. 1991. Four versions of double effect. Journal of Medicine and Philosophy. 16:515–​544. Mashour, G. A., ed. 2010. Consciousness, Awareness, and Anesthesia. Cambridge: Cambridge University Press. Materstvedt, L. J., and G. Bosshard. 2009. Deep and continuous palliative sedation (terminal sedation): clinical-​ethical and philosophical aspects. Lancet Oncology 10:622–​627. Miccinesi, G., J. Rietjens, L. Deliens, E. Paci, G. Bosshard, T. Nilstun, M. Norup, and G. van der Wal. 2006. Continuous deep sedation: physicians’ experiences in six European countries. The Journal of Pain and Symptom Management 31(2):122–​129. Morita, T., Y. Shinone, M. Ikenaga, M. Miyoshi, T. Nakaho, K. Nishitateno, M. Sakonji, Y. Shima, K. Suenaga, C. Takigawa, H. Kohara, K. Tani, Y. Kawamura, T. Matsubara, A. Watanabe, Y. Yagi, T. Sasaki, A. Higuchi, H. Kimura, H. Abo, T. Ozawa, Y. Kizawa, and Y. Uchitomi. 2005. Efficacy and safety of palliative sedation therapy: a multicenter, prospective, observational study conducted on specialized palliative care units in Japan. The Journal of Pain and Symptom Management 30(4):320–​328. Morita, T., S. Tsuneto, and Y. Shima. 2001. Proposed definitions for terminal sedation. Lancet 358(9278):335–​356. Morita, T., S. Tsuneto, and Y. Shima. 2002. Definition of sedation for symptom relief: a systematic literature review and a proposal of operational criteria. The Journal of Pain and Symptom Management 24(4):447–​453. Murray, S. A., M. Kendall, K. Boyd, and A. Sheikh. 2005. Illness trajectories and palliative care. British Medical Journal 330 (7498):1007–​1011. Noreika, V., L. Jylhänkangas, L. Móró, K. Valli, K. Kaskinoro, R. Aantaa, H. Scheinin, and A. Revonsuo. 2011. Consciousness lost and found: Subjective experiences in an unresponsive state. Brain and Cognition 77:327–​334. Onwuteaka-​Philipsen, B. D., A. Brinkman-​Stoppelenburg, C. Penning, G. J. de Jong-​Krul, J. J. van Delden, and A. van der Heide. 2012. Trends in end-​of-​life practices before and after the enactment of the euthanasia law in the Netherlands from 1990 to 2010: a repeated cross-​ sectional survey. Lancet 380(9845):908–​915. Orentlicher, D. 1997. The Supreme Court and physician-​assisted suicide—​rejecting assisted suicide but embracing euthanasia. The New England Journal of Medicine 337(17):1236–​1239.

124    Sigrid Sterckx and Kasper Raus Orentlicher, D. 2013. Principle and practice for palliative sedation: gaps between the two. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 116–​131. Porta-​Sales, J. 2013. Palliative sedation: clinical, pharmacological and practical aspects In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 65–​85. Quill, T. E., R. Dresser, and D. Brock. 1997. The rule of double effect—​a critique of its role in end-​of-​life decision making. The New England Journal of Medicine 337(24):1768–​1771. Raus, K., L. Anquinet, J. Rietjens, L. Deliens, F. Mortier, and S. Sterckx. 2014. Factors that facilitate or constrain the use of continuous sedation at the end of life by physicians and nurses in Belgium: results from a focus group study. Journal of Medical Ethics 44(4):230–​234. Raus, K., S. Sterckx, and F. Mortier. 2012. Continuous deep sedation at the end of life and the “natural death” hypothesis. Bioethics 26(6):329–​336. Raus, K., S. Sterckx, and F. Mortier. 2013. Can the doctrine of double effect justify continuous deep sedation at the end of life? In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives. Cambridge: Cambridge University Press: pp. 177–​201. Rechtbank’s Gravenhage Court. 2012. 09/​925370-​08. LJN BV6482 Rietjens, J., J. J. van Delden, B. Onwuteaka-​Philipsen, H. Buiting, P. van der Maas, and A. van der Heide. 2008. Continuous deep sedation for patients nearing death in the Netherlands: descriptive study. British Medical Journal 336(7648):810–​814. Rietjens, J., A. van der Heide, A. M. Vrakking, B. Onwuteaka-​Philipsen, P. J. van der Maas, and G. van der Wal. 2004. Physician reports of terminal sedation without hydration or nutrition for patients nearing death in the Netherlands. Annals of Internal Medicine 141:178–​185. Rousseau, P. 2000. The ethical validity and clinical experience of palliative sedation. Mayo Clinical Proceedings 75:1064–​1069. Sacred Congregation for the Doctrine of the Faith. 1980. Declaration on Euthanasia—​Iura et bona. AAS 72, 1:542–​552, Documenta 38, 5 May 1980. http://​www.vatican.va/​roman_​curia/​ congregations/​cfaith/​documents/​rc_​con_​cfaith_​doc_​19800505_​euthanasia_​en.html. Saunders, C. 2001. The evolution of palliative care. Journal of the Royal Society of Medicine 94: 430–​432. Seale, C. 2000. Changing patterns of death and dying. Social Science and Medicine 51: 917–​930. Seale, C. 2009. End-​of-​life decisions in the UK involving medical practitioners. Palliative Medicine 23:198–​204. Seale, C. 2010. Continuous deep sedation in medical practice: a descriptive study. Journal of Pain and Symptom Management 39(1):44–​53. Seymour, J. 1999. Revisiting medicalisation and “natural’ death.” Social Science and Medicine 49:691–​704. Seymour, J., D. Clark, and M. Winslow. 2005. Pain and palliative care: the emergence of new specialties. Journal of Pain and Symptom Management 29(1):2–​13. Seymour, J., R. Janssens, and B. Broeckaert. 2007. Relieving suffering at the end of life: practitioners’ perspectives on palliative sedation from three European countries. Social Science and Medicine 64(8):1679–​1692. Simon, A., M. Kar, J. Hinz, and D. Beck. 2007. Attitudes towards terminal sedation: an empirical survey among experts in the field of medical ethics. BMC Palliative Care 6: doi: 10.1186/​ 1472-​684X-​6-​4.

Continuous Sedation at the End of Life    125 Smith, G. P. 1998. Terminal sedation as palliative care: revalidating a right to a good death. Cambridge Quarterly of Healthcare Ethics 7:382–​387. Steindal, S. H., I. S. Bredal, L. W. Sørbye, and A. Lerdal. 2011. Pain control at the end of life: a comparative study of hospitalized cancer and noncancer patients. Scandinavian Journal of Caring Sciences 25(4):771–​779. Steinhauser, K. E., E. C. Clipp, M. McNeilly, N. A. Christakis, L. McIntyre, and J. A. Tulsky. 2000. In search of a good death: observations of patients, families, and providers. Annals of Internal Medicine 132(10):825–​832. Stone, P., C. Phillips, O. Spruyt, and C. Waight. 1997. A comparison of the use of sedatives in a hospital support team and in a hospice. Palliative Medicine 11:140–​144. Sykes, N., and A. Thorns. 2003. Sedative use in the last week of life and the implications for end-​ of-​life decision making. Archives of Internal Medicine 163: 341–​344. Tännsjö, T. 2004. Terminal sedation: a substitute for euthanasia?, in T. Tännsjö, ed., Terminal Sedation: Euthanasia in Disguise? Dordrecht: Kluwer Academic, pp. 15–​30.

Vacco v. Quill. 1997. 521 U.S. 793 (U.S. Supreme Court).

van Delden, J. J. 2007. Terminal sedation: source of a restless ethical debate. Journal of Medical Ethics 33(4):187–​188. van Delden, J. J. M. 2013. The ethical evaluation of continuous sedation at the end of life. In: S. Sterckx, K. Raus, and F. Mortier, eds., Continuous Sedation at the End of Life—​Clinical, Legal and Ethical Perspectives, Cambridge: Cambridge University Press: pp. 218–​227. Ventafridda, V., C. Ripamonti, F. De Conno, M. Tamburini, and B. R. Cassileth. 1990. Symptom prevalence and control during cancer patients’ last days of life. Journal of Palliative Care 6(3):7–​11.

Washington v. Glucksberg. 1997. 521 U.S. 702 (US Supreme Court).

Wilson, J., M. G. Clarke, P. Ewings, J. D. Graham, and R. MacDonagh. 2005. The assessment of patient life-​expectancy: how accurate are urologists and oncologists? British Journal of Urology 95:794–​798.

Chapter 8

The Ethics of Me di c a l ly Assisted Nu t ri t i on a nd Hy dration at t h e E nd of Li fe Separating the Wheat from the Chaff Daniel P. Sulmasy

Food and water are basic necessities for animal life. In ordinary human life, these animal necessities assume a complex set of social and even spiritual meanings that present significant challenges for palliative medical practice surrounding the issues of nutrition and hydration. The nurturance and care that are symbolized by an infant feeding at its mother’s breast affect all our eating and drinking. Food and drink are a means of interpersonal communication and social bonding as well as nutrition, and sharing meals has important familial, social, and often even religious significance. When someone is sick, the first thing a family often does is to offer a cup of water. As an old medical school professor of mine used to tell his patients, “Food is medicine. Eat!” Yet the development of new technologies for providing enteral and parenteral nutrition has complicated matters enormously. For countless millennia, as human beings became sick from chronic illnesses, they lost their appetites as part of the natural history of these disorders, calorie consumption dropped, and dehydration gradually ensued. These were viewed as phenomena intrinsically and uncontroversially linked to the dying process. Similarly, acute neurological injuries prevented eating and drinking, and this was just part of what happened when someone died of a stroke or head trauma. The advent of new technologies for supplying nutrition and hydration, however, raises questions about how one ought wisely to use these technologies in caring for those who are dying. These questions are medically and morally vexing for clinicians, patients, and families. In this chapter, I will first review each of the available treatment modalities, discussing techniques, indications, benefits, and risks. I will then propose a classification of four

The Ethics of Medically Assisted Nutrition and Hydration    127 critical decisional junctures along the trajectory of disease progression, points at which questions regarding nutrition and hydration arise across all life-​threatening diagnoses. I will then discuss several disabling, chronic, progressive, and eventually fatal diagnoses with differing trajectories toward death and discuss the pros and cons of providing medically assisted nutrition and hydration in the care of patients with these disorders at each of these four critical decisional junctures. Finally, I will discuss a number of specific ethical controversies in medically assisted nutrition and hydration, including religious issues, the moral and psychological difference between stopping and starting nutritional therapy, advance care planning around these modalities, and the ethics of voluntarily stopping eating and drinking.

Treatment Modalities For centuries, medicine and surgery failed to appreciate the enormous impact that adequate nutrition has on recovery from disease. Even in the second half of the twentieth century, patients who were unable to eat were often treated for weeks with no nutrition other than intravenous (IV) 5% dextrose in a solution of ¼ normal saline to which a small amount of potassium was added. Now, however, when patients are unable to eat and or drink, or are unable to digest what they eat or drink, or are unable to consume sufficient calories for healthy living, or when oral intake poses a problem for their health or interferes with treatment, contemporary medicine can provide alternative means for maintaining hydration and nutrition. These include enteral modalities, such as nasogastric (NG) or percutaneous feeding tubes, and parenteral modalities such as IV and subcutaneous means. Before discussing the ethics of their use in palliative medicine, I will first review the medical aspects of these techniques one by one.

Nasogastric Tube Feeding Enteral treatment is generally preferred to parenteral. As the saying goes, “if the gut works, use it.” There are fewer complications with enteral feeding compared with parenteral feeding, it is cheaper, and the adequacy of nutrition is generally better (Daly 1999). NG tubes are tubes placed through the nose down into the stomach. The use of large-​ bore tubes, such as those used for gastric lavage, has now largely been supplanted by the use of smaller bore, flexible feeding tubes, which are much more comfortable for patients. They work well and can be used indefinitely if the stomach and intestines are functional. These tubes are not without complications, however. They can be misplaced in the lungs (up to 3% of the time when done at the bedside “blindly” without radiological support); may not be placed far enough into the stomach to function properly; can lead to aspiration pneumonia, nose bleeds, or sinus infections; or can perforate the gut

128   Daniel P. Sulmasy or cause the intestines to become tangled and the blood supply cut off (Baskin 2006; Bullock, Waltrip, Price, and Glandiuk 2004; de Aguilar-​Nascimentio and Kudsk 2007; Metheny, Meert, and Clouse 2007; Stayner, Bhatnagar, McGinn, and Fang 2012). To diminish the chances of aspiration, these tubes are sometimes advanced into the proximal small intestine, but this increases the risks of perforation or bowel entanglement. To diminish the complications associated with tube placement, endoscopic and radiographic techniques are increasingly being used to guide tube placement (Sorokin and Gottlieb 2006).

Percutaneous Endoscopic Gastrostomy Tube Feeding The other “enteral” option is to place a tube through the skin into the stomach. There are two techniques for placing such tubes. In the nonsurgical approach, a gastroenterologist passes a flexible fiberoptic device called an esophagogastroduodenoscope through the mouth of a sedated patient, into the stomach, and, in a darkened procedure room, shines a light anteriorly so that it can be seen on the skin surface. The gastroenterologist then uses a large needle to insert a feeding tube through the skin at the point of the light so that it passes directly into the stomach, at which point a set of “bumpers” are set in place (one in the stomach and one on the skin surface) to keep the tube in position (Kurien, McAlindon, Westaby, and Sanders 2010). This technique, called percutaneous endoscopic gastrostomy or PEG tube placement, has been used widely since the 1980s and is generally considered safe and effective. Complications include cardiac or pulmonary complications from the sedation, perforation of the stomach, infection or bleeding at the site of percutaneous insertion, a real but lesser chance of bowel entanglement compared with flexible NG tubes, and aspiration pneumonia due to food being placed in the stomach going up the esophagus and down the trachea.

Surgical Gastrostomy or Jejunostomy Tube Feeding Before the advent of PEG techniques, surgery was necessary to place an indwelling feeding tube, and surgical placement still finds some uses today. Tubes can be placed either in the body of the stomach, the duodenum, or the beginning of the small intestine (jejunum). Since the 1990s, surgical feeding tube placement has been done laparoscopically (Murayama, Schneider, and Thompson 1995). Some data, however, suggest that an open technique has fewer complications (Bankhead, Fisher, and Rolandelli 2005). Although PEG is generally the technique of choice, surgical placement is sometimes necessary, as when patients cannot have endoscopy because of tumors of the head and neck or esophagus. Many also prefer open or laparoscopic surgical techniques if the tube is to be placed in the jejunum because the use of endoscopy to do so is new and technically difficult to perform (Nagle and Murayma 2004). Complications of the tube itself are similar to those of tubes placed via endoscopy, except that jejunostomy tubes appear to carry

The Ethics of Medically Assisted Nutrition and Hydration    129 a lower risk of aspiration (Panagiotakis, DiSario, Hilden, Ogara, and Fang 2008). The additional complications are those of surgery itself: general anesthesia and the invasion of the peritoneal (intra-​abdominal) cavity.

Proctoclysis Although not a technique for feeding, patients can be hydrated by the insertion of a tube into the rectum, and water or normal saline can be infused and systemically absorbed. This last enteral technique has been available since the early twentieth century (Needham 1995). It is safe and effective and has been used in palliative care for patients who are unable to eat or drink, or who have limited IV access (such as when veins have been scarred by chemotherapy and other IV treatments), or for those whose skin is in such poor condition as to preclude hypodermoclysis (Bruera, Pruvost, Scholler, Montejo, and Watanabe 1998; Bruera, Schoeller, and Pruvost 1994).

Intravenous Hydration and Nutrition IV hydration was first introduced for the treatment of the dehydration associated with cholera in the nineteenth century (Lewins 1832). However, technical problems such as bleeding and infection precluded widespread use of the technique. By the late 1930s, sterile saline was being poured through glass funnels into rubber tubes attached to steel intravenous needles as an alternative to proctoclysis, but it was not until World War II that anything like the modern notion of IV therapy became widespread, safe, and effective (Millam 1996). This technique is most useful for restoring fluid volume in the form of water and electrolytes but is limited by the fact that the amount of sugar or other carbohydrates, protein, fat, and other vital nutrients that can be delivered by peripheral vein is restricted because these substances are caustic to small veins such as those on the arms. Although generally safe and effective, complications include fluid overload from too much salt and water, electrolyte abnormalities, infection, irritation or clotting at the vein site, bruising and bleeding at the site, and swelling due to the needle becoming dislodged from the vein or transfixing the vein, with the fluid subsequently extravasating into the subcutaneous space. Although many of these complications are common (inflammation of the vein or phlebitis may occur, at least in mild form, in as many as 50% of patients), these complications are rarely severe or life-​threatening (Dychter, Gold, Carson and Haller 2012).

Total Parenteral Nutrition Parenteral nutrition can be supplied at a level nearly commensurate with that of enteral nutrition by cannulating a large-​diameter vein such as the superior vena cava either by

130   Daniel P. Sulmasy accessing the jugular vein or the subclavian vein and inserting a large-​bore catheter. These veins are able to withstand the potentially caustic effects of highly concentrated nutrients and can support what is called total parenteral nutrition (TPN; see Bozzetti 2007). Using this technique, one can supply amino acids, lipids, high concentrations of sugar, and essential vitamins and nutrients that cannot be supplied via a peripheral vein. To help protect against accidental dislodging or infection, the catheter can be “tunneled” underneath the skin with “ports” that are left dangling outside the skin and can be repeatedly disinfected and punctured by a needle connected to the nutritional solution (or medications). This technique can be used when patients are expected to require nutritional support for more than a few days but the enteral route cannot be used (e.g., due to blockage by a tumor) or when there are medical reasons to “rest” the bowel due to effects wrought either by the disease or by its therapy. The technique is very effective and remarkably safe for short-​and medium-​term use, but complications tend to limit long-​ term use. Complications associated with the insertion of the tube (often performed, because of its invasiveness, by a surgeon or other specially trained physician) include bleeding or puncture of the lung (pneumothorax). Long-​term complications include infection, clot formation (thrombosis and embolism), and irritation of the large veins (phlebitis).

Partial Parenteral Nutrition Best conceived of as a sort of “compromise” between IV hydration and nutrition and TPN is a technique known as partial parenteral nutrition (PPN). This technique allows one to supply more nutrition than is possible using a small vein for IV hydration but less than is possible using TPN. PPN utilizes medium-​size veins, generally accessed through a peripherally inserted central catheter (PICC) line. This technique is less invasive and less risky than TPN, but, although supplying more nutrition than is possible with a simple IV line, falls far short of meeting the patient’s full nutritional requirements. PPN is sometimes used to supplement very poor short-​term oral intake or to provide better nutritional support than could be accomplished with a simple IV. Generally, it is only used for, at most, a week or two with the expectation that the patient will soon recover and be able to eat and drink regularly.

Hypodermoclysis Sometimes patients are close to death but cannot receive IV therapy, either owing to the poor condition of their veins (e.g., as a result from scarring from chemotherapy) or (e.g., in resource-​poor countries) due to a lack of training in IV cannulation on the part of the caregivers. Fluids can then be administered under the skin using a technique called hypodermoclysis (Sasson and Shvartzman 2001). Some medicines can be given this way, but no real nutrients, and one cannot expect to fully rehydrate a patient using

The Ethics of Medically Assisted Nutrition and Hydration    131 this technique, only to restore the patient partially. More commonly used in the early twentieth century, it fell out of favor with improvements in IV techniques (Berger 1984). However, there has been a renewed interest in the technique in geriatrics and palliative medicine in recent years (Bruera, Legris, Kuehn, and Miller 1990; Di Giulio et al. 2008; Fonzo-​Christe, Vukasovic, Wasilewski-​Rasca, and Bonnabry 2005; Hays 1985; Nasri 2000, Ronchon et al. 1997). The technique is generally safe, and it is effective if one’s goals are modest. Complications include bleeding and swelling under the skin if the infusion is too fast or the subcutaneous space is limited. It cannot be used for more than a few days, and its use in palliative care typically is limited to patients who are very close to death.

Critical Decisional Junctures Sadly, there are any number of disabling, chronic, progressive, eventually fatal diseases, and each has a different trajectory toward death (Lynn 2001). There are, however, roughly speaking, four critical decisional junctures that can be thought about with respect to the natural history of each disease and the intended therapeutic use of any of the various techniques of medically assisted nutrition and hydration. These may be categorized as follows: (1) nutrition and hydration as adjunctive to the treatment of the underlying disease, (2)  nutrition and hydration as life-​prolonging therapy, (3)  nutrition and hydration as palliative in the terminal phase of the disease (final months), and (4) nutrition and hydration as palliative in the agonal phase of the disease (final days to weeks). These “critical junctures” are obviously broadly drawn, idealized conceptual notions. It is well-​known, for example, that motives can be multiple and that prognostication is often limited. Yet isolating these critical decisional junctures helps to highlight the clinical and moral “moments” that patients, their loved ones, and practitioners face with respect to decisions about nutrition and hydration and therefore seem useful for the purposes of this chapter.

General Ethical Considerations Patients generally have moral obligations to maintain their health and to avail themselves of medical interventions when they are sick. Such obligations can be justified on the basis of enlightened self-​interest, on the grounds of beneficence (because one is limited in one’s ability to do good for others, such as family, friends, and community, when one is sick), and on the grounds of social utility because persons who are healthy contribute, on the whole, to advancing the net social welfare. However, such obligations are “imperfect” in the sense that they do not oblige everyone everywhere and under all circumstances and apply only up to a point. Generally speaking, such obligations

132   Daniel P. Sulmasy cease when the medical intervention is likely to be ineffective or when the burdens of pursuing the intervention outweigh the benefits. Although it is common to speak of a legal “right” to refuse treatment, one need not invoke the concept of rights to justify forgoing treatment under such circumstances (Sulmasy 2002). All one need recognize is that positive moral precepts, such as the duty to give to charity, only apply up to a point. Where the point is at which the obligation ceases will differ from individual to individual. With respect to medical treatment, it will differ depending on such things as the individual person’s tolerance for pain, the value he or she places on continued bodily life, individual moral and psychological resources, personal religious and philosophical views, and other personal concerns, capacities, and convictions. Importantly, this general principle applies just as readily to medical nutritional interventions as it does to other medical interventions. Although the law refrains from making any judgments whatsoever about when an individual might be morally mistaken in deciding to forgo a treatment, ethics can judge such decisions as morally justifiable or unjustifiable. For example, whereas an eighteen-​ year-​old maintains a legal right to refuse insulin therapy for newly diagnosed type 1 diabetes, were an eighteen-​year-​old boy to forgo insulin based on the belief that the burdens of a lifetime of insulin injections outweighed the benefits for him, ethics could judge that such a decision is outside the reasonable bounds of normal human variation and that he has made a moral mistake. Although one could not legally force him to take insulin, his refusal to do so would seem to be a violation of the general moral duty to maintain health. Likewise, some decisions to forgo medical interventions for nutrition and hydration might be judged morally wrong even if the patient has a legal right to refuse such care. Importantly, decisions to forgo potentially life-​sustaining therapies that are either ineffective or have been judged to be more burdensome than beneficial have, for centuries, been distinguished from euthanasia and assisted suicide. To forgo ventilator support because one has judged that it is more burdensome than beneficial in the end stages of emphysema is not an act of suicide. Here, the law follows traditional moral thinking in upholding this distinction between killing and allowing to die (Vacco v. Quill 1997), while many philosophers, especially libertarians, political liberals, and utilitarians, have questioned the meaningfulness of the distinction (Dworkin, Nagel, Nozick, Rawls, and Thomson 1997). Properly formulated, however, the distinction is coherent (Sulmasy 1998). To “kill” a patient is to act with the specific intention-​in-​acting of making the patient dead by way of one’s intentional act of establishing a new lethal pathophysiological state; to allow to die is to act with the specific intention-​in-​acting of withholding or withdrawing an intervention that is forestalling the effects of an underlying lethal pathophysiological state, either intending to make the patient dead or not. If one’s intention-​ in-​acting is to make the patient dead, then even one’s act of allowing to die can be judged morally equivalent to euthanasia and one could be held culpable for the act. If one’s intention is merely to concede that the patient has no further moral obligation to continue medical intervention, and one’s intention-​in-​acting is fulfilled when the patient is no longer receiving the medical intervention, then one’s allowing the patient to die is

The Ethics of Medically Assisted Nutrition and Hydration    133 justified on traditional moral grounds, providing that the judgment that the burdens of continued treatment outweigh the benefits is reasonable. This formulation applies to IV hydration and feeding tubes every bit as much as it does to ventilator treatment. This is not, of course, an argument to establish that euthanasia is wrong. That would require a separate argument beyond the scope of this chapter. Rather, it is to say that if the withholding or withdrawing of life-​sustaining treatment is distinct from euthanasia, and these actions can often be morally justified—​as the conservative moral position just presented suggests—​then these considerations apply to the withholding and withdrawing of medically assisted nutrition and hydration as much as they do to other life-​sustaining interventions.1

The Ethics of Nutrition and Hydration at Critical Decisional Junctures Adjunctive Therapeutic Use of Nutrition and Hydration The use of medically assisted (artificial) means of supplying nutrition and hydration as an adjunct to treatment with a curative intent is relatively uncontroversial. This is most common in the treatment of malignancies. Patients with head and neck or esophageal cancer who are unable to eat seem to do better if they are given nutritional support by PEG tube (Wiggenraad et al. 2007). Recent data seem to suggest that NG tube feeding is just as good (Nugent, Lewis, and O’Sullivan 2013). Similarly, both NG and PEG feeding seem equally efficacious in supporting patients in the acute phase of stroke that renders them at least temporarily unable to eat and who have a reasonable chance of recovery (Norton, Homer-​Ward, Donnelly, Long, and Holmes 1996). Although the principle holds true that enteral feeding is preferred if the gut “works,” patients undergoing intensive chemotherapy who are unable to be fed enterally due to gastrointestinal toxicity of the drugs and patients who have undergone extensive resections of gastrointestinal cancers may require TPN (Mercadante 1996). Ordinarily, if there are good reasons to pursue therapy with curative intent and such nutritional support is adjunctive to that therapy, it would be morally good and right to accept such nutritional support, all other things being equal.

Nutrition and Hydration as Life-​Prolonging Therapy The use of medically assisted (artificial) means of supplying nutrition and hydration as a life-​prolonging measure in patients with chronic, eventually fatal illnesses who either have permanently lost the ability to eat or who are ambulatory and at high risk for aspiration with oral intake but have a reasonable prognosis for living for an extended time if

134   Daniel P. Sulmasy they accept such support should also not be morally controversial. For example, patients with head and neck or esophageal malignancies who have incurable metastatic disease but for whom control of the cancer requires a treatment that has made normal eating impossible or inadvisable may benefit from protracted use of a feeding tube, which could potentially extend length of life and provide a reasonable quality of life. Likewise, it is occasionally the case that swallowing is affected early in the course of a progressive, eventually fatal, neurodegenerative disease. Particularly if the disease does not (or has not yet) adversely affected cognitive capacity and the patient remains ambulatory, the risk for aspiration due to the feeding tube (as a consequence of esophageal reflux of the tube feeding and subsequent aspiration) is low. Ordinarily, in such circumstances, it would be morally good and right to pursue such nutritional support, all other things being equal. For example, in amyotrophic lateral sclerosis (ALS; especially the bulbar as opposed to the spinal form of the disease), the use of feeding tubes in this situation, early or in the middle course of the disease, has come to be regarded as the standard of care and may prolong life by as much as six months (Borasio et al. 2001; Spataro, Ficano, Piccoli, and La Bella 2011).

Nutrition and Hydration as Palliative Care in the Terminal Phase of Disease When a patient enters the terminal phase of a chronic, eventually fatal disease, with only months to live, nutrition and hydration are sometimes offered with palliative intent. As a general rule, this is both medically and morally unwise. From a medical point of view, the best available data suggest that such measures are either more burdensome than beneficial for patients or, at best, of no proven benefit. If the latter is the case, then the involved expense to patient and society and even the small burdens for the patient of undergoing bothersome invasive procedures with even small risks of complications, done for no good reason, seem morally unwarranted. For example, in patients with advanced malignancies who are given PEG tubes, there is a 20% complication rate and 75% one-​year mortality (Keung, Liu, Nuzhad, Rabinowits, and Patel 2012). Similarly, TPN has little effect on outcomes in terminally ill cancer patients (Dy 2006). When a chronic, progressive neurological condition such as Alzheimer’s disease reaches a terminal phase in which patients are bed-​bound, unable to eat, and unable to recognize family members, the best evidence suggests that medical interventions to provide nutrition and hydration do little to affect length or quality of life compared with those who do not receive tube feeding (Callahan et al. 2000; Finucane, Christmas, and Travis 1999). Median survival rates after tube placement in patients with Alzheimer’s disease have ranged from two to six months (Brett 2001; Meier, Ahronheim, Morris, Baskin-​Lyons, and Morrison 2001; Murphy and Lipman 2003). Some have urged caution regarding the security with which one can make claims about lack of benefit based on these data (Glick and Jotkowitz 2013) and have cited a few small studies

The Ethics of Medically Assisted Nutrition and Hydration    135 (reported in the body of an editorial and in a letter) that appear to show modest benefit (Lindemann and Nikolaus 2001; Shapiro and Friedmann 2006). However, systematic reviews, taking the literature as a whole, concur that there is a lack of demonstrable benefit (Alagiakrishnan, Bhanji, and Kurain 2013; Sampson, Candy, and Jones 2009), even though, for obvious ethical reasons, no randomized controlled trials have been conducted to offer definitive proof. Despite these data, the use of PEG tubes in patients with Alzheimer’s has increased, and 10% of all feeding tubes in older patients are placed because of Alzheimer’s disease (Mendiratta et al. 2012). Perhaps this is a paradigmatic example of what has been dubbed “the technological imperative”—​the notion that “can” implies “ought.” It is true, for example, that aspiration pneumonia is common in patients with advanced Alzheimer’s disease (Kalia 2003). Nonetheless, NG tubes and PEG tubes do not provide a resolution to this problem in this patient population at this stage of the disease because, as explained earlier, food placed in the stomach via a tube is prone to reflux up the esophagus and then to be aspirated down the trachea. This is especially true for bed-​bound patients for whom the protection against aspiration provided by gravity and the upright position are absent. The comfort and concern symbolized by providing food and water are sometimes offered as reasons morally necessitating medical interventions to provide nutrition and hydration when patients are unable to eat. However, even beyond the fact that it is unlikely to provide biomedical benefit, proponents of this line of reasoning fail to appreciate that the symbolic comfort they aspire to provide cannot be accomplished by these means. Patients fed by vein or enterally placed tubes do not experience the pleasure of taste and do not experience the human interaction that surrounds a human meal. Moreover, because tube feeding is generally performed by continuous rather than bolus provision of the liquid nutritional mixture, the patient is generally alone while being nourished and never experiences the stomach distention that signals satiety. One might argue that the feeling of satiety triggered by stomach distention could be accomplished by bolus feeding, but bolus feeding significantly increases the risk of reflux and aspiration, which is generally the reason that tube feeding was, however misguidedly, initiated in the first place. Somehow, it also seems that patients with progressive, chronic, eventually fatal degenerative neurological diseases such as Alzheimer’s are treated differently from patients with other progressive, chronic, eventually fatal diseases such as cancer. In patients with cancer or incurable infections, such as multidrug-​resistant tuberculosis, or other diseases of “the body,” it is accepted that “the body is shutting down,” and it is therefore thought acceptable to forgo life-​sustaining treatments such as tube feedings. Somehow, popularly, Alzheimer’s disease is thought to be different, as if it only affected “the mind,” and it is therefore not accepted that, at the end stage, the bodies of such patients are “shutting down.” In fact, however, there is evidence that Alzheimer’s disease is also a systemic disease and that the circulating factors associated with diminution in appetite in diseases like cancer (such as cachexin) are also abnormal in the end stages of Alzheimer’s disease (Hoffer 2006; Intebie et al. 2002–​2003). Whereas some of the loss of interest in eating in Alzheimer’s disease is due to the disease’s direct effects on cognition, some is also due to these systemic factors. At the end stage, the bodies of patients with

136   Daniel P. Sulmasy this disorder are also “shutting down,” and the forgoing of medical interventions to provide nutrition and hydration should not be thought on this basis to differ, morally, from the forgoing of nutrition and hydration in patients dying of cancer. In other diseases, such as cancers, these considerations apply a fortiori. Yet, clinical practice still has not caught up with either ethics or the evidence. For example, even though a recent study showed that 14% of patients dying of cancer in the hospital had TPN in the last week of life (Cheng, Sham, Chan, Li, and Au 2015), multiple studies have failed to show evidence of any benefit in length or quality of life by the use of either TPN or enteral feeding in cancer patients who have entered the terminal phases of their malignancies (Dy 2006). Complication rates and in-​hospital mortality rates are high (20%) for patients with terminal cancer treated with feeding tubes (Keung, Liu, Nuzhad, Rabinowits, and Patel 2012). What are the alternatives to enteral or parenteral nutrition and hydration in terminally ill patients? First, one could accept the risk of aspiration and continue feeding, by hand if necessary, as much as the patient desires. This would be justified under the rule of double effect (Sulmasy 2007a; Sulmasy and Pellegrino 1999) because the feeding is itself good; one foresees, but does not intend the aspiration; aspiration is not the cause of the nutrition one does intend; and the benefits of feeding by mouth could easily be construed as proportional to the burdens of possible aspiration under these circumstances. A decision to accept the risks of oral feeding could be coupled with a decision to forgo treatment of the pneumonia on the grounds that such treatment needlessly prolonged an inevitable death. This strategy has been called “comfort feeding only” (Palecek et al. 2010). Second, were loved ones to remain fearful of aspiration, but ready to forgo the feeding tube, the patient could be fed very small amounts of a substance such as honey, placed via a spoon under the tongue, so that the patient would experience the sweet taste and the enjoy, at whatever level of cognition remains, the interpersonal interaction of being fed by hand but with a very low risk of aspiration. Third, one could choose not to feed at all. Under such a plan, one should pay especially careful attention to oral comfort, using mouth swabs and similar measures. In all these cases, the justification would be the same: that tube feeding is either of no benefit or the burdens outweigh the benefits. Those who would counterargue that such patients experience grave discomfort by forgoing nutrition and hydration fail to take account of the indirect evidence against this thesis. First, as noted, patients at the end stages of almost all chronic diseases lose their appetites as part of the natural history of these disorders, and the biochemical concomitants of this terminal anorexia appear to be present in Alzheimer’s disease just as they are in chronic infections and malignancies. Second, third-​party observations suggest that such patients do not appear to be in pain or discomfort (Pasman et al. 2005; Rapoport, Shaheed, Newman, Rugg, and Steele 2013). Third, to the extent that any such patients might experience discomfort, this can be treated adequately with analgesics and conservative measures such as good mouth care. Those who would counterargue that nutrition and hydration are “basic care” that cannot, morally, be withheld or withdrawn from anyone, appear to be making one of two possible mistakes (Sulmasy 2006). The first might be that they are categorizing medically assisted nutrition and hydration as somehow nonmedical. Even the preferred

The Ethics of Medically Assisted Nutrition and Hydration    137 phrase of morally conservative observers, “medically assisted,” however, makes this claim seem, at best, linguistically confused. No one without a medical license is permitted to insert a PEG tube or a subclavian vein catheter for TPN. Whereas providing food and beverage for normal eating and drinking can be counted as basic care (and food and water ought always to be offered to patients), the medical modalities for providing nutrition and hydration are absolutely not the same, experientially or mechanically, as eating and drinking. There is no reason to think that being nourished through a tube forcing food into one’s stomach is morally or medically different from being ventilated through a tube forcing air into one’s lungs. Certainly U.S. law considers medically assisted nutrition and hydration a medical treatment (Cruzan v. Director 1990). The second possible mistake would be to think that a patient cannot, morally, forgo even what might ordinarily seem like “basic care” on the grounds that it has become more burdensome than beneficial. It is not how many whistles and bells an intervention has that renders it optional, but the burdens relative to the benefits. Even morally conservative observers, for instance, recognize that one ought not force-​feed a patient who, in the throes of death, has lost appetite due to the ravages of the underlying disease. Similarly, whereas blankets might also be considered basic care, if a patient were febrile and the room chilly, one would not force the patient to accept a blanket simply because the room was chilly without taking account its lack of usefulness to the patient in these particular circumstances. Underlying these attempts to argue in favor of always providing medically assisted nutrition and hydration often lurks yet another subtle but important moral mistake. The moral point of view assumed by most proponents of nutrition and hydration for the terminally ill, although noble, distorts the picture. They tend to see the issue of nutrition and hydration as a question of what duties the family, the medical staff, and the community have toward the sick and the dying rather than assuming the perspective of the patient and asking how far the patient’s duty to preserve his or her life must extend (Sulmasy 2005a, 2006). Viewed from the perspective of caregivers, it might, at first, seem reasonable to conclude that the caregivers have a duty to provide sustenance because it is relatively simple, even if medical assistance is required. Viewed from the perspective of the patient, however, even when the patient cannot give voice to that perspective, the picture looks different. It is often quite reasonable to judge that the patient is justified in refusing yet one more invasive treatment, which is of dubious benefit anyway, and has already sufficiently fulfilled the basic duty to preserve his or her life. The proper perspective, morally speaking, is the perspective of the patient.

Nutrition and Hydration as Palliative Care in the Agonal Phase of Disease Because, as argued earlier, medically assisted nutrition and hydration is generally either ineffective or more burdensome than beneficial once the disease has progressed to the terminal phase, a fortiori the same considerations would obtain regarding the ethics of

138   Daniel P. Sulmasy forgoing these means in the agonal phase. Available data suggest, for example, that there is no impact on in-​hospital mortality for cancer patients in the agonal stage of their diseases provided with PEG tube feedings compared with patients not receiving PEG tubes (Raijmakers et al. 2011). The old hospice nurse’s aphorism is that it is better to die “dry” than to die “wet.” Overhydration increases the work of breathing for the patient and exacerbates the “death rattle,” which can be unsettling for loved ones who often have never seen (or heard) anyone die before. Once again, good oral care is important for comfort. It would seem, however, unnecessarily intrusive and uncomfortable to insert an NG tube or an IV in the final hours or days unless one or the other is needed to deliver a drug with a specific and important palliative purpose. There are, however, two situations in which medically assisted nutrition and/​or hydration have been thought to have a possible palliative role even in the agonal phase when the patient is expected to die within hours or days. The first is in treatment of bowel obstruction and the second in the treatment of terminal delirium. I will consider each of these in turn. Bowel obstruction in the agonal phase of a disease often results in severe pain, nausea, and vomiting and constitutes a palliative care emergency. Gynecologic and gastrointestinal cancers are especially common causes, particularly in the setting of peritoneal metastases or of an unresected primary intestinal mass. Because definitive surgical treatment is generally not indicated, opioid drugs can exacerbate nausea and impair gastrointestinal motility, and decompression with a large-​bore NG tube seems harsh from a palliative perspective, the care of such patients is not easy. It is sometimes suggested that parenteral nutrition has a palliative role in this situation. If the cause, however, is genuine, complete obstruction rather than ileus or obstipation; there is no real prospect for reversing the situation; and the patient’s life span is short, then there is almost no role for TPN or IV hydration (Soriano and Davis 2011). Patients tend to “third-​space” much of the fluid, and vigorous hydration leads to anasarca and pulmonary edema, adding to the patient’s discomfort and disfigurement as death approaches. Such results are hardly palliative. There might be a palliative role for a PEG tube under such circumstances, not to provide nutrition, but to “vent” the obstruction without interposing a NG tube between the patient and loved ones and thus circumventing the discomfort associated with an NG tube (Issaka et al. 2013). If the patient is not quite yet in the agonal phase of the disease, the obstruction has been vented with a tube or is only partial, and the patient and/​or loved ones see value in the possibility of prolonging life by several weeks, then TPN may play a modest palliative role. Median survival under such circumstances was fifty-​five days in one study (August, Thorn, Fisher, and Welchek 1991). For most patients, however, these fifty-​five days will require bearing with a host of other symptoms that will make this time seem more burdensome than beneficial. For several decades, the role of hydration in the treatment of terminal delirium has been controversial. Dehydration has frequently been cited as a cause for terminal delirium and gentle rehydration recommended (Centeno, Sanz, and Bruera 2004; Dalal and Bruera 2004; Moyer 2011). The evidence for an association between delirium and

The Ethics of Medically Assisted Nutrition and Hydration    139 dehydration was sparse, but suggestive (Bruera et al. 1995, 2005; de Stoutze, Bruera, and Suarez-​Almazor 1995). Several narrative reviews pointed out that the numbers in these studies were small, the data largely observational, and prospective trials lacking (Bavin 2007; Plonk and Arnold 2005). One observational study showed that the univariate association between delirium and dehydration disappeared in multivariate modeling, however, suggesting that dehydration was not causally associated (Lawlor et al. 2000). Now, two prospective trials have been completed, both of which failed to show any benefit from IV hydration (Bruera et al. 2013; Morita, Tei, and Inoue 2003). Another recent trial failed to substantiate any benefit from PEG tubes in preventing delirium in cancer patients in the last week of life (Raijmakers et al. 2011). Whereas the optimal treatment of terminal delirium remains to be determined, the best current evidence suggests that neither enteral nor parenteral hydration is indicated for this purpose. Thus, the nursing aphorism that “it is better to die dry than wet” still holds, and there is little medical or moral warrant for medically assisted nutrition and hydration in the last days of life.

Special Ethical Considerations Regarding Nutrition and Hydration in Palliative Care Religious Issues Several religious traditions have developed special positions regarding this form of medical intervention and have deemed the provision of medically assisted nutrition and hydration a moral imperative. These positions ought to be respected by clinicians, just as they ought to respect any other seriously held moral views. The arguments that I have evinced in this chapter are permissive, not prescriptive. That is to say, the moral stance of practitioners ought to be that it is morally permissible for patients to forgo medically assisted nutrition and hydration, not that patients are morally obligated to forgo such treatment. The standard for unilaterally withholding or withdrawing potentially life-​sustaining medical interventions ought to be one of biomedical futility—​that is to say—​that the treatment either will not work, or the patient will die in a very short period of time even if the treatment is provided (Sulmasy 1997). Although there might be occasions when this standard is met (e.g., multisystem organ failure in a patient with refractory malignancy already on maximal life support), it is generally not outside the bounds of reasonable credibility for a patient or a patient’s loved ones to judge the benefits of medically assisted nutrition and hydration commensurate with the burdens. If so, then the treatment ought to be provided for such patients, even if the practitioners would not weigh the burdens and the benefits in the same way themselves.

140   Daniel P. Sulmasy I will discuss two particular religious traditions in some detail, even though some other religious communities have similar positions. Orthodox Jews, for example, generally require medically assisted nutrition and hydration in patients who are unable to eat by themselves (Rosner and Abramson 2009). Although some Jewish physicians have tried to offer halachic arguments in favor of the permissibility of withholding medically assisted nutrition and hydration (Gillick 2001), this has not been accepted by halachic decisors or the Orthodox rabbinical community (Kupfer and Tessler 2000). The Conservative movement is split on the matter (Ain 2003). The most prominent and convincing Conservative Jewish voice in favor of the permissibility of sometimes withholding or withdrawing medically assisted nutrition and hydration is Eliot Dorff (1998). Reform Judaism is generally more favorably disposed toward a permissive stance but with a palpable sense of reluctance (Wahofsky 2005). Israelis have shown outcome data similar to that produced in the United States revealing that long-​term survival rates are equivalent whether feeding tubes are used or not in patients with end-​stage dementia (Jaul, Singer, and Calderon-​Margalit 2006). It is probably the case, however, that the Jewish practice of providing medically assisted nutrition and hydration in patients who are unable to eat is based on conceptions of duty and observance of religious law rather than consequences. In comparative studies, rates of tube feeding in demented elderly patients are higher among Jewish patients than among Gentiles, higher in Israel than in Canada, and, within Canada, higher in Jewish than in secular institutions. (Clarfield et al. 2006). Among Christians, mainstream Protestants have not generally made a distinction between feeding tubes and other potentially life-​extending treatments. The Roman Catholic Church, however, has recently taken a position that makes the use of a feeding tube, other things being equal, morally obligatory in certain clinical circumstances. This position should not be overstated or misinterpreted. The Roman Catholic Church invented the term “extraordinary means” 500 years ago and has consistently applied its centuries-​old casuistry to new technologies, permitting the withholding and withdrawing of such technologies as the ventilator since the 1950s (Pius XII 1958). In general, this also includes forgoing medically assisted nutrition and hydration (Sulmasy 2005b). A recent exception, however, concerns the use of this modality for patients in the permanent vegetative state (PVS). In caring for patients suffering from this condition (or similar conditions of permanent, stable, severe cerebral injury, short of brain death), provided there are no complications associated with the treatment itself, if the patient is not otherwise dying, there are no other significant comorbidities or obvious sources of suffering, and there are no socioeconomic constraints such as a lack of availability of the means or excessive cost, the Church now officially teaches that medically assisted enteral feeding must be considered “ordinary” and morally obligatory. The exact text of U.S. Ethical and Religious Directives for Catholic Health Care Services, Number 58, is as follows: In principle, there is an obligation to provide patients with food and water, including medically assisted nutrition and hydration, for those who cannot take food orally. This obligation extends to patients in chronic and presumably irreversible

The Ethics of Medically Assisted Nutrition and Hydration    141 conditions (e.g., the “persistent vegetative state”) who can reasonably be expected to live indefinitely if given such care. Medically assisted nutrition and hydration become morally optional when they cannot reasonably be expected to prolong life or when they would be “excessively burdensome for the patient or [would] cause significant physical discomfort, for example resulting from complications in the use of the means employed.” For instance, as a patient draws close to inevitable death from an underlying progressive and fatal condition, certain measures to provide nutrition and hydration may become excessively burdensome and therefore not obligatory in light of their very limited ability to prolong life or provide comfort. (Ethical and Religious Directives 2009)

This position obviously does not entail, as some commentators have suggested, that the Catholic Church now teaches that we must “all die with a feeding tube” (Caplan 2004). This moral directive to use feeding tubes does not apply to the end stages of malignancies or chronic incurable infections. Nor does it apply to end-​stage dementia or other progressive neurological conditions such as Parkinson’s disease, because, as described earlier, data demonstrating that these modalities actually extend life in patients suffering from such conditions are scant, these patients are irreversibly dying of progressive underlying neurological conditions, and they often are suffering from multiple other comorbidities (Sulmasy 2007b). Furthermore, even for those patients with PVS who are dependent on tube feeding, should complications arise, the Catholic teaching does not require the continued use of the feeding tube. This is a new teaching for the Catholic Church. It is unclear to many Catholic theologians exactly how it squares with centuries of traditional teaching (Hardt 2008; Shannon and Walter 2005). The scope and importance of this teaching, however, should not be exaggerated. It appears to apply primarily to rare conditions such as PVS, based on the notion that such patients are not dying, but are instead disabled persons deserving of our social resources. It is not the case that the Catholic Church now teaches that everyone must die with a feeding tube. Other religious traditions may have special views about feeding tubes as well, but an in-​depth discussion of each is beyond the scope of this chapter. Suffice it to reiterate that such views should be respected every bit as much as nonreligious views should be. It is helpful, however, in providing morally good palliative care, to understand the rationale behind such views and to have a clear understanding of the precise scope of these religious moral teachings regarding medically assisted nutrition and hydration.

Withholding Versus Withdrawing Nutrition and Hydration It is a standard teaching in bioethics that withdrawing a medical intervention does not require more moral justification than withholding that intervention. The same holds for medically assisted nutrition and hydration. One needs to justify a decision to withhold

142   Daniel P. Sulmasy just as much as one needs to justify a decision to withdraw medically assisted nutrition and hydration. The grounds for such justification are the same in either case: either the treatment cannot achieve its goals, or it is more burdensome than beneficial. Sometimes, in fact, it is morally preferable to attempt medically assisted nutrition and hydration, as when one is uncertain whether the patient will recover. In such situations of uncertainty, it can, in fact, be more medically justifiable to withdraw nutrition and hydration support after an unsuccessful trial than it would have been to have withheld it from the beginning when one did not know whether it might work. Nonetheless, feeding tubes seem to raise special concerns. Logically speaking, the concern that feeding tubes are, by the nature of the therapy, morally different from other life-​sustaining treatments is unfounded. One might argue that in stopping a feeding tube one knows with certainty that the patient will die, whereas this is not true in other cases. But this claim is false. If a patient is apneic, one can be quite certain that the patient will die, quickly, if one turns off a ventilator. If it is permissible to discontinue ventilator support when one is reasonably certain that the patient cannot breathe and will die, then it ought to be permissible to discontinue nutrition and hydration when one is reasonably certain that the patient cannot eat and will die. Moreover, the claim that one can always be apodictically certain that a patient will not eat is itself false. There are cases in which one withdraws a feeding tube, and the patient regains appetite or the underlying disease state improves and the patient begins to eat. Although this is uncommon, it cannot be claimed that one is apodictically certain that a patient will die after discontinuing nutrition and hydration, and this cannot be the grounds for prohibiting the capping or pulling of a feeding tube. Furthermore, the active/​passive distinction does no real work in distinguishing withdrawing a feeding tube from withdrawing other treatments because discontinuing a feeding tube is an active motor act, just as discontinuing a ventilator is an active motor act. And neither requires more moral justification than a decision to withhold the therapeutic intervention (Sulmasy 1998). Despite these philosophical objections, certain religious groups do maintain a moral distinction between withholding and withdrawing feeding tubes. Prominent among these are many Jewish groups. Schostak, for instance, acknowledges the philosophical judgment but points out that this is a matter of religious law, writing, “both legally and ethically, there is no distinction between withholding or withdrawing medical treatment. In halacha, however, the difference between the two is extremely significant” (Schostak 1994). Nonetheless, individual rabbis and patients may differ on this matter, and each case should be decided individually. It is important to point out that neither Roman Catholicism nor any other mainstream Christian denomination holds the view that the distinction between withholding and withdrawing treatment is morally significant. Some smaller Christian groups do hold this view, however, and, as argued earlier, as a matter of justice, these religious views should be respected even if one does not accept the patient’s rationale. Religion and philosophy aside, it does seem psychologically more difficult for patients, families, and physicians to discontinue treatment than it does for them to withhold treatment. People seem to feel more causally responsible if they discontinue

The Ethics of Medically Assisted Nutrition and Hydration    143 a treatment than if they never started it in the first place, even if they cannot articulate a rational reason for that intuition. Because of this, it is useful to discuss the potential psychological burdens of discontinuing medically assisted nutrition and hydration in consent discussions undertaken before this therapy is initiated, particularly if undertaken explicitly as a time-​limited trial. Families may either be able to prepare themselves better for this eventuality or decide that, for instance, given the low likelihood of success of a particular trial and their anticipated psychological difficulties in authorizing its discontinuation, it would be better not to start medically assisted nutrition and hydration in the first place.

Advance Care Planning Regarding Nutrition and Hydration Because this therapy involves such medical, moral, psychological, and legal complexities compared with other potentially life-​sustaining treatments, it is probably best to be explicit about one’s views regarding medically assisted nutrition and hydration in advance care planning. For instance, some jurisdictions require “clear and convincing evidence” of a patient’s prior wishes before loved ones can authorize the discontinuation of medically assisted nutrition and hydration, and this high evidentiary standard has been upheld by the U.S. Supreme Court (Cruzan v. Director 1990). Advance care planning can provide such evidence. Those who have someone they can trust to speak for them can appoint a durable power of attorney for health care and should discuss their thoughts, feelings, values, and preferences regarding nutrition and hydration with their health care power of attorney, including any particular religious views. Because medical situations can be quite difficult to predict, and the circumstances are always unique, I believe it is best to state explicitly that the agent has the power to make decisions regarding nutrition and hydration while informing the agent about one’s values and beliefs about nutrition and hydration as a guide to decision-​making at the bedside. It is best to avoid sweeping statements such as “no feeding tubes” or “under no circumstances should my nutritional support be stopped.” Those who cannot appoint a power of attorney for health care may choose to express their preferences through a treatment directive such as a living will, aiming for enough specificity to be able to guide decision-​ making while also trying to avoid extreme statements.

Voluntarily Stopping Eating and Drinking One of the most controversial issues in medically assisted nutrition and hydration at the end of life is the question of voluntarily stopping eating and drinking at the end of life. Some have suggested that patients who report an “intolerable” level of suffering and no longer wish to live, but have no physical or physiological reason that they cannot eat, may, by reason of a willful decision, refuse food and water as a way of hastening

144   Daniel P. Sulmasy their own deaths. It has been argued that (1) it is uncertain that this is a form of suicide, (2) it is morally appropriate for physicians sometimes to suggest this course of action to patients, (3)  physicians ought never provide medically assisted nutrition and hydration to such patients, and (4) physicians have moral obligations to support patients in their decisions not to eat and to provide, as indicated, hospitalization or hospice care and treatment for any symptoms such patients might experience, including, if requested, palliative sedation to unconsciousness (Quill and Byock 2000; Quill and Lo 1997; Schwartz 2007). These views have been justified on the basis of physician duties to respect patient autonomy, to relieve suffering, and not to abandon patients. There are, however, significant logical and moral problems with the justifications offered for this practice (Jansen and Sulmasy 2002). First, if there is no underlying pathological reason why a person cannot eat (such as swallowing difficulties, anorexia such as that associated with high levels of circulating cachexin in chronic diseases, etc.), then this cannot be a case of “allowing to die.” By definition, “allowing to die” involves the withholding or withdrawing of a life-​sustaining treatment that interferes with the natural progression of an underling pathological condition (Sulmasy 1998). Because the intention is to bring about death, and there is no underlying pathophysiological condition that has altered appetite or eating such that death would ensue but for a medical intervention such as tube feeding,2 the act must be categorized as an act of killing. Because the person wills this upon him-​or herself with the specific intention-​in-​acting of making him-​or herself dead, then the act would be one of “self-​killing” or suicide (Selbstmord in German). No amount of linguistic redescription of the act would change its essential features. Voluntarily stopping eating and drinking when there is no medical reason that one cannot eat or drink in the normal way is suicide, plain and simple. And, if suicide is morally wrong, then voluntarily stopping eating and drinking is morally wrong. Second, to suggest this course of action to a patient who goes on to carry out this plan is to assist that patient in suicide. Assistance is not merely material. One can assist another person by providing ideas and/​or information. So, by definition, if a physician were to provide assistance by suggesting the possibility or by instructing the patient in how to do so, the act would be physician-​assisted suicide. The traditional Hippocratic Oath, which forbids euthanasia, explicitly forbids assistance in suicide by means of a recommendation to the patient: “I will neither give a deadly drug to anybody if asked for it, nor will I make a suggestion to this effect” (Hippocrates 1967). And if physician-​ assisted suicide is wrong, then it is wrong for a physician to suggest this course of action to a patient. Third, it is true that the conditions under which one might consider it morally permissible to force-​feed a patient are limited. For instance, in cases of anorexia nervosa in which a patient has lost decisional capacity, one might do so, but this would be under the supposition that the will of the patient has been distorted by the psychiatric disease, thus depriving the patient of the decisional capacity to refuse. The same would be true of interventions for suicidal patients. In these cases, medically assisted nutrition and hydration would be last resorts, and one would appeal first to any residual, undistorted

The Ethics of Medically Assisted Nutrition and Hydration    145 will to eat and survive as the first course of action. Supposing that a patient with intact decision-​making capacity voluntarily stops eating and drinking with no pathophysiological cause, one ought not to initiate force-​feeding. This does not mean, however, that the physician must accept this act of voluntarily stopping eating and drinking as morally justified. One could, of course, appeal to the patient’s intact autonomous will and reason as one’s first resort, trying to persuade the patient otherwise. This would not be paternalistic or an abuse of the physician’s Aesculapian power. There is a difference between indifference and respect for autonomy. It can be indifferent to say “you do your thing and I’ll do mine.” True respect sometimes entails offering a challenge to a patient’s decision, testing its depth, authenticity, and probity. In the end, if the patient persists in the decision voluntarily to stop eating and drinking, assaulting the patient’s autonomy by force-​feeding surely would represent an unjustifiable violation of that patient’s dignity. Nonetheless, this does not mean that the decision must be ratified and accepted. Respect for the patient’s dignity entails, first and foremost, respect for the intrinsic dignity of the patient. Respect for a patient’s interests implies, logically and morally, respect for the patient whose interests they are. If the patient is eliminated, the grounds for respecting the patient’s interests are eliminated. Moreover, some patients who float the idea of assisted suicide are only “testing the waters” by making the suggestion, seeing if anyone cares enough to try to convince them not to do so. Thinking that one is being respectful of such a patient’s rights by agreeing with the plan may only confirm his or her deepest fear—​that no one cares enough to try to stop him or her from following through—​thus reinforcing the resolve to carry out the suicide plan. Fourth, providing a space within which to carry out the plan or aiding in symptom control while the patient is dying as a result of voluntarily stopping eating and drinking is a form of moral complicity in the act. If assisted suicide is wrong, and voluntarily stopping eating and drinking is a form of suicide, it follows that making the process easier is a form of assistance in that suicide, which would also be morally wrong. It would be disingenuous to claim that one is not aiding the patient in suicide, but only (1) respecting the autonomy of the patient by not forcing medically assisted nutrition and hydration and (2) not abandoning the patient at a time of need. If one knows an act is morally wrong under a full and complete description, one cannot claim to engage in that act under a partial description under which it would not be morally wrong, then one does not thereby escape culpability. “Oedipus marries his mother” and “Oedipus marries Jocasta” name the same event. But if Oedipus knows that Jocasta is his mother, then Oedipus cannot claim only to be marrying a woman named Jocasta and thereby justifiably claim not to be guilty of incest. Of course, if one does not believe that either suicide or physician-​assisted suicide is wrong, then neither suggesting nor aiding in such a course of action would be wrong. Space precludes a direct discussion of the ethics of assisted suicide and euthanasia in this chapter. All that I have argued here is that voluntarily stopping eating and drinking is a form of suicide and that for a physician to suggest this course of action, to condone this course of action, or to provide a safe haven in which to perform the action, all constitute forms of physician-​assisted suicide. It is disingenuous to argue otherwise.

146   Daniel P. Sulmasy If the patient raises the option, those who are morally opposed to physician-​assisted suicide (or fearful that it might be an illegal action in jurisdictions in which physician-​ assisted suicide is illegal) should explore the patient’s reasons for the action, address unmet needs that might be prompting the request, explore alternatives, and, if the patient remains resolute, politely explain their opposition and offer to help the patient in any way they can short of assistance in this course of action. They should show compassion and concern and firmly leave the door open for the patient to return should the patient change his or her mind.

Conclusion I have discussed the medical techniques for assistance with nutrition and hydration, and the benefits and risks of these techniques. I have argued that the duty of a patient to make use of medically assisted nutrition and hydration can be assessed according to the same criteria that are used to assess whether one may forgo any medical intervention and that medically assisted nutrition and hydration become morally optional whenever they either will not work or when the burdens outweigh the benefits, as judged by the patient (or the patient’s loved ones in the event that the patient becomes unable to decide for him-​or herself). I then argued that there are four critical decisional junctures that can be described along the course of a number of different disease-​specific trajectories toward death. I considered the pros and cons of medically assisted nutrition and hydration at each of these junctures. I argued that the benefits tend to outweigh the burdens when medically assisted nutrition and hydration is used as an adjunctive treatment or as a life-​prolonging treatment, but that the case for using these techniques is very weak in the terminal and agonal phases of most chronic, disabling, progressive, and eventually fatal diseases. I described several particular religious positions regarding medically assisted nutrition and hydration and argued that these positions need to be better understood by clinicians and, as a general rule, respected. I argued that there is no philosophical reason that decisions to withdraw medically assisted nutrition and hydration should require more moral justification than withholding these interventions, but I also described several religious positions that make a moral distinction in this regard and argued that these positions should be respected. Nonetheless, given the difference in psychological reaction patients and family members often experience when asked to authorize withdrawing versus withholding, I counseled that clinicians should explicitly take account of this difference whenever they are eliciting informed consent for a time-​limited trial of medically assisted nutrition and hydration. Finally, I argued that although one should never force medically assisted nutrition and hydration on a patient who has voluntarily stopped eating and drinking, if the patient has no medical reason rendering him or her unable to eat, this action is a form of suicide, and physician participation in such a plan is a form of physician-​assisted suicide.

The Ethics of Medically Assisted Nutrition and Hydration    147 Food and drink are so basic to animal existence and so fraught with symbolic and interpersonal meaning for human beings that the ethical issues surrounding the forgoing of medically assisted nutrition and hydration seem to be more contentious than the issues surrounding forgoing other medical interventions. This makes clear thinking about the issues vital if we are to get the ethics right. It is my hope that this chapter contributes to such clarification.

Notes 1. The traditional moral position holds that it is acting with the specific intention-​in-​ acting of making the patient dead that is morally wrong. To the extent that this view is justified (and such justification is beyond the scope of this chapter), then, according to this view, killing would always be wrong (except in cases of self-​defense or rescue), whereas allowing to die would sometimes be morally permissible. Thus, although this traditional moral position would hold that forgoing a life-​sustaining treatment with the specific intention-​in-​action of making a patient dead is morally wrong, it must also be acknowledged that it would be difficult, if not impossible, to enact a policy that absolutely prohibited the forgoing of treatments with the intention of making the patient dead. Rather, a reasonable approach for a society that held this conservative moral view (which is, after all, the view of most of the world’s societies) would be to prohibit all cases in which the intentions are clear (i.e., all cases of direct killing such as euthanasia and assisted suicide), while leaving judgment open about cases of allowing to die. Killing functions under such a moral system as a pathognomic sign of the agent’s intention to make the patient dead. If a physician injects 150 mEq of potassium chloride into a patient’s left ventricle, or gives a patient a prescription for 2,000 mg of secobarbital, which the patient ingests under the physician’s guidance and support, then the physician’s intentions are clear. In cases of allowing to die, one would have more moral work to do. It is almost universally agreed that when the patient or the surrogate of a decisionally incapable patient judges the burdens of a treatment to outweigh the benefits, this judgment is within the broad bounds of reasonability, and the intention-​in-​acting is to stop the treatment rather than to make the patient dead, then this action is morally acceptable. It is also universally agreed that, in at least some cases, the withholding or withdrawing of life-​sustaining treatments with the specific intention-​in-​acting of making the patient dead is morally wrong (e.g., if a physician were to do so without the consent of a patient in order to marry the patient’s spouse, with whom the physician is having an affair). Societies that hold the conservative moral view can then adopt reasonable policies that prohibit euthanasia and assisted suicide but are permissive about cases of allowing to die, recognizing that most cases of forgoing life-​sustaining treatment will meet the moral standard of not intending death while reserving the possibility of making judgments of moral blameworthiness in extreme cases, such as when the intention is clearly to make the patient dead as a means to a nefarious end. Consequently, some cases of allowing to die, whether by forgoing ventilator support, nutrition and hydration, or other modalities, in which death is the outcome intended by both the patient and the physician would need to be tolerated because the burdens of trying to prohibit such acts would be excessive.

148   Daniel P. Sulmasy 2. Presuming that the patient is not suicidally depressed, which would represent a pathophysiological condition affecting appetite and eating, but also pathologically engendering a suicidal intention.

References Ain, S. October 31, 2003. “Rabbis split on feeding tube case.” The Jewish Week. http://​www. thejewishweek.com/​news/​new_​york/​rabbis_​split_​feeding_​tube_​case. Last accessed Dec. 10, 2015. Alagiakrishnan, K., R. A. Bhanji, and M. Kurian. January-​February 2013. “Evaluation and management of oropharyngeal dysphagia in different types of dementia: A systematic review.” Archives of Gerontology and Geriatrics 56(1): 1–​9. August, D. A., D. Thorn, R. L. Fisher, and C. M. Welchek. 1991. “Home parenteral nutrition for patients with inoperable malignant bowel obstruction.” Journal of Parenteral and Enteral Nutrition 15: 323–​327. Bankhead, R. R., C. A. Fisher, and R. H. Rolandelli. December 2005. “Gastrostomy tube placement outcomes: Comparison of surgical, endoscopic, and laparoscopic methods.” Nutrition in Clinical Practice 20(6): 607–​612. Baskin, W. N. February 2006. “Acute complications associated with bedside placement of feeding tubes.” Nutrition in Clinical Practice 21(1): 40–​55. Bavin, L. September 2007. “Artificial rehydration in the last days of life:  Is it beneficial?” International Journal of Palliative Nursing 13(9): 445–​449. Berger, E. Y. March 1984. “Nutrition by hypodermoclysis.” Journal of the American Geriatric Society 32(3): 199–​203. Borasio, G. D., P. J. Shaw, O. Hardiman, A. C. Ludolph, M. L. Sales Luis, V. Silani, and European ALS Study Group. September 2001. “Standards of palliative care for patients with amyotrophic lateral sclerosis: Results of a European survey.” Amyotrophic Lateral Sclerosis and Other Motor Neuron Disorders 2(3): 159–​164. Bozzetti, F. December 2007. “Total parenteral nutrition in cancer patients.” Current Opinions in Supportive and Palliative Care 1(4): 281–​286. Brett, A. S. October 22, 2001. “Dementia, gastrostomy tubes, and mortality.” Archives of Internal Medicine 161(19): 2385–​2386. Bruera, E., J. J. Franco, M. Maltoni, et al. 1995. “Changing pattern of agitated impaired mental status in patients with advanced cancer: Association with cognitive monitoring, hydration, and opioid rotation.” Journal of Pain and Symptom Management 10: 287–​291. Bruera, E., D. Hui, S. Dalal, et  al. January 1, 2013. “Parenteral hydration in patients with advanced cancer:  A  multicenter, double-​ blind, placebo-​ controlled randomized trial.” Journal of Clinical Oncology 31(1): 111–​118. Bruera, E., M. A. Legris, N. Kuehn, and M. J. Miller. 1990. “Hypodermoclysis for the administration of fluids and narcotic analgesics in patients with advanced cancer.” Journal of Pain and Symptom Management 5: 218–​220. Bruera, E., M. Pruvost, T. Schoeller, G. Montejo, and S. Watanabe. April 1998. “Proctoclysis for hydration of terminally ill cancer patients.” Journal of Pain and Symptom Management 15(4): 216–​219. Bruera, E., R. Sala, M. A. Rico, et al. 2005. “Effects of parenteral hydration in terminally ill cancer patients: A preliminary study.” Journal of Clinical Oncology 23: 2366–​2371.

The Ethics of Medically Assisted Nutrition and Hydration    149 Bruera, E., T. Schoeller, and M. Pruvost. 1994. “Proctoclysis for hydration of terminal cancer patients.” Lancet 344: 1699. Bullock, T. K., T. J. Waltrip, S. A. Price, and S. Galandiuk. September 2004. “A retrospective study of nosocomial pneumonia in postoperative patients shows a higher mortality rate in patients receiving nasogastric tube feeding.” American Surgery 70(9): 822–​826. Callahan, C. M., K. M. Haag, M. Weinberger, et al. 2000. “Outcomes of percutaneous endoscopic gastrostomy among older adults in a community setting.” Journal of the American Geriatric Society 48(9): 1048–​1054. Caplan, A. April 6, 2004. “Must we all die with a feeding tube?” MSNBC.com. http://​www. nbcnews.com/​id/​4669899/​ns/​health-​health_​care/​t/​must-​we-​all-​die-​feeding-​tube/​ Last accessed Dec. 10, 2015. Centeno, C., A. Sanz, and E. Bruera. April 2004. “Delirium in advanced cancer patients.” Palliative Medicine 18(3): 184–​194. Cheng, B. H., M. M. Sham, K. Y. Chan, C. W. Li, and H. Y. Au. 2015. “Intensive palliative care for patients with hematological cancer dying in hospice: Analysis of the level of medical care in the final week of life.” American Journal of Hospital Palliative Care. 32: 221–​225. Clarfield, A. M., J. Monette, H. Bergman, et al. 2006. “Enteral feeding in end-​stage dementia: A comparison of religious, ethnic, and national differences in Canada and Israel.” Journal of Gerontology, A Biological Sciences Medical Sciences 61: 621–​627. Cruzan v. Director, Missouri Department of Health, 497 U.S. 261 (1990). Dalal, S., and E. Bruera. November-​December 2004. “Dehydration in cancer patients: To treat or not to treat.” Journal of Supportive Oncology 2(6): 467–​479, 483. Daly, J. M. 1999. “The evolution of surgical nutrition: Nutrient and anabolic interventions.” Annals of Surgery 22: 19–​20. de Aguilar-​Nascimento J. E., and K. A. Kudsk July-​August 2007. “Clinical costs of feeding tube placement.” Journal of Parenteral and Enteral Nutrition 31(4): 269–​273. de Stoutz, N. D., E. Bruera, and M. Suarez-​Almazor. 1995. “Opioid rotation for toxicity reduction in terminal cancer patients.” Journal of Pain and Symptom Management 10: 378–​384. Di Giulio, P., F. Toscani, D. Villani, C. Brunelli, S. Gentile, and P. Spadin. September 2008. “Dying with advanced dementia in long-​term care geriatric institutions:  A  retrospective study.” Journal of Palliative Medicine 11(7): 1023–​1028. Dorff, E. N. 1998. Matters of life and death:  A  Jewish approach to modern medical ethics. Philadelphia: The Jewish Publication Society, 208–​217. Dworkin, R., T. Nagel, R. Nozick, J. Rawls, and J. J. Thomson. March 27, 1997. “Assisted suicide: The philosopher’s brief.” New York Review of Books: 41–​47. Dy, S. M. 2006. “Enteral and parenteral nutrition in terminally ill cancer patients: A review of the literature.” American Journal of Hospital Palliative Care 23: 369–​377. Dychter, S. S., D. A. Gold, D. Carson, and M. Haller. March-​April 2012. “Intravenous therapy: A review of complications and economic considerations of peripheral access.” Journal of Infusion Nursing 35(2): 84–​91. Ethical and religious directives for Catholic health care services, 5th ed. 2009. Directive 58. Washington, DC:  United States Conference of Catholic Bishops. http://​www.usccb.org/​ issues-​and-​action/​human-​life-​and-​dignity/​health-​care/​upload/​Ethical-​Religious-​Directives-​ Catholic-​Health-​Care-​Services-​fifth-​edition-​2009.pdf Last accessed Dec. 10, 2015. Finucane, T. E., C. Christmas, and K. Travis. 1999. “Tube feeding in patients with advanced dementia:  A  review of the evidence.” Journal of the American Medical Association 282(14): 1365–​1370.

150   Daniel P. Sulmasy Fonzo-​Christe, C., C. Vukasovic, A. F. Wasilewski-​Rasca, and P. Bonnabry. April 2005. “Subcutaneous administration of drugs in the elderly: Survey of practice and systematic literature review.” Palliative Medicine 19(3): 208–​219. Gillick, M. R. February 2001. “Artificial nutrition and hydration in the patient with advanced dementia:  Is withholding treatment compatible with traditional Judaism?” Journal of Medical Ethics 27(1): 12–​15. Glick, S. M., and A. B. Jotkowitz. 2013. “Feeding dementia patients via percutaneous endoscopic gastrostomy.” Annals of Long-​Term Care: Clinical Care and Aging 21(1): 32–​34. Hardt, J. A. January 28, 2008. “Church teaching and my father’s choice.” America 198 (2): 11–​16. Hays, H. 1985. “Hypodermoclysis for symptom control in terminal cancer.” Canadian Family Physician 31: 1253–​1256. Hippocrates. 1967. “The Oath.” In The Hippocratic oath: Text, translation, and interpretation. In Ancient medicine: Selected papers of Ludwig Edelstein, edited by O. Temkin and C. L. Temkin. Baltimore: Johns Hopkins University Press, 3–​63. Hoffer, L. J. 2006. “Tube feeding in advanced dementia: The metabolic perspective.” British Medical Journal 333: 1214–​1215. Intebi, A. D., L. Garau, I. Brusco, M. Pagano, R. C. Gaillard, and E. Spinedi. 2002–​2003. “Alzheimer’s disease patients display gender dimorphism in circulating anorectic adipokines.” Neuroimmunomodulation 10(6): 351–​358. Issaka, R. B., D. M. Shapiro, N. D. Parikh, M. F. Mulcahy, S. Komanduri, J. A. Martin, and R. N. Keswani. December 24, 2013. “Palliative venting percutaneous endoscopic gastrostomy tube is safe and effective in patients with malignant obstruction.” Surgical Endoscopy 28(5): 1668–​1673. Jansen, L. A., and D. P. Sulmasy. 2002. “Sedation, hydration, alimentation, and equivocation:  Careful conversation about care at the end of life.” Annals of Internal Medicine 136: 845–​849. Jaul, E., P. Singer, and R. Calderon-​Margalit. December 2006. “Tube feeding in the demented elderly with severe disabilities.” Israeli Medical Association Journal 8(12): 870–​874. Kalia, M. 2003. “Dysphagia and aspiration pneumonia in patients with Alzheimer’s disease.” Metabolism 52: 36–​38. Keung, E. Z., X. Liu, A. Nuzhad, G. Rabinowits, and V. Patel. December 2012. “In-​hospital and long-​term outcomes after percutaneous endoscopic gastrostomy in patients with malignancy.” Journal of the American College of Surgeons 215(6): 777–​786. Kupfer, Y., and S. Tessler. June 8, 2000. “Rethinking the role of tube feeding in patients with advanced dementia.” New England Journal of Medicine 342(23): 1755–​1756. Kurien, M., M. E. McAlindon, D. Westaby, and D. S. Sanders. May 7, 2010. “Percutaneous endoscopic gastrostomy (PEG) feeding.” British Medical Journal 340: c2414. Lawlor, P. G., B. Gagnon, I. L. Mancini, J. L. Pereira, J. Hanson, M. E. Suarez-​Almazor, and E. D. Bruera. March 27, 2000. “Occurrence, causes, and outcome of delirium in patients with advanced cancer: A prospective study.” Archives of Internal Medicine 160(6): 786–​794. Lewins, R. 1832. “Injection of saline solutions in extraordinary quantities into the veins in cases of malignant cholera.” Lancet 2: 243–​244. Lindemann, B., and T. Nikolaus. 2001. “Outcomes of percutaneous endoscopic gastrostomy in dementia patients.” Journal of the American Geriatric Society 49(6): 838–​839. Lynn, J. 2001. “Perspectives on care at the close of life. Serving patients who may die soon and their families: The role of hospice and other services.” Journal of the American Medical Association 285: 925–​932.

The Ethics of Medically Assisted Nutrition and Hydration    151 Meier, D. E., J. C. Ahronheim, J. Morris, S. Baskin-​Lyons, and R. S. Morrison. February 26, 2001. “High short-​term mortality in hospitalized patients with advanced dementia: Lack of benefit of tube feeding.” Archives of Internal Medicine 161(4): 594–​599. Mendiratta, P., J. M. Tilford, P. Prodhan, K. Curseen, G. Azhar, and J. Y. Wei. December 2012. “Trends in percutaneous endoscopic gastrostomy placement in the elderly from 1993 to 2003.” American Journal of Alzheimers Diseases and Other Dementias 27(8): 609–​613. Mercadante, S. January 1996. “Nutrition in cancer patients.” Supportive Care of Cancer 4(1): 10–​20. Metheny, N. A., K. L. Meert, and R. E. Clouse. March 2007. “Complications related to feeding tube placement.” Current Opinions in Gastroenterology 23(2): 178–​182. Millam, D. January-​February 1996. “The history of intravenous therapy.” Journal of Intravenous Nursing 19(1): 5–​14. Morita, T., Y. Tei, and S. Inoue. August 2003. “Agitated terminal delirium and association with partial opioid substitution and hydration.” Journal of Palliative Medicine 6(4): 557–​563. Moyer, D. D. February 2011. “Review article: Terminal delirium in geriatric patients with cancer at end of life.” American Journal of Hospital Palliative Care 28(1): 44–​51. Murayama, K. M., P. D. Schneider, and J. S. Thompson. 1995. “Laparoscopic gastrostomy: A safe method for obtaining enteral access.” Journal of Surgical Research 58: 1–​5. Murphy, L. M., and T. O. Lipman. June 9, 2003. “Percutaneous endoscopic gastrostomy does not prolong survival in patients with dementia.” Archives of Internal Medicine 163(11): 1351–​1353. Nagle, A. P., and K. M. Murayama. 2004. “Laparoscopic gastrostomy and jejunostomy.” Journal of Long Term Effects of Medical Implants 14(1): 1–​11. Nasri, F. 2000. “Subcutaneous hydration by hypodermoclysis. A practical and low cost treatment for elderly patients.” Drugs and Aging 16(4): 313–​319. Needham, D. 1995. “Proctoclysis for hydration.” Lancet 345: 596. Norton, B., M. Homer-​Ward, M. T. Donnelly, R. G. Long, and G. K. Holmes. January 6, 1996. “A randomised prospective comparison of percutaneous endoscopic gastrostomy and nasogastric tube feeding after acute dysphagic stroke.” British Medical Journal 312(7022): 13–​16. Nugent, B., S. Lewis, and J. M. O’Sullivan. January 31, 2013. “Enteral feeding methods for nutritional management in patients with head and neck cancers being treated with radiotherapy and/​or chemotherapy.” Cochrane Database of Systematic Reviews 1: CD007904. Palecek, E. J., J. M. Teno, D. J. Casarett, L. C. Hanson, R. L. Rhodes, and S. L. Mitchell. March 2010. “Comfort feeding only: A proposal to bring clarity to decision-​making regarding difficulty with eating for persons with advanced dementia.” Journal of the American Geriatric Society 58(3): 580–​584. Panagiotakis, P. H., J. A. DiSario, K. Hilden, M. Ogara, and J. C. Fang. April-​May 2008. “DPEJ tube placement prevents aspiration pneumonia in high-​risk patients.” Nutrition in Clinical Practice 23(2): 172–​175. Pasman, H. R., B. D. Onwuteaka-​Philipsen, D. M. Kriegsman, M. E. Ooms, M. Ribbe, and G. van der Wal. 2005. “Discomfort in nursing home patients with severe dementia in whom artificial nutrition and hydration is forgone.” Archives of Internal Medicine 165 1729–​1735. Pius XII. Spring 1958. “The prolongation of life: An address to an international congress of anesthesiologists, Nov. 24, 1957.” The Pope Speaks 4(4): 393–​398. Plonk, W. M. Jr., and R. M. Arnold. October 2005. “Terminal care:  The last weeks of life.” Journal of Palliative Medicine 8(5): 1042–​1054. Quill, T. E., and I. R. Byock. 2000. “Responding to intractable terminal suffering: The role of voluntary refusal of food and fluids.” Annals of Internal Medicine 132: 402–​414.

152   Daniel P. Sulmasy Quill, T. E., and D. W. Lo. 1997. “Palliative options of last resort: A comparison of voluntarily stopping eating and drinking, terminal sedation, physician-​assisted suicide, and voluntary active euthanasia.” Journal of the American Medical Association 278: 2099–​2104. Raijmakers, N. J., L. van Zuylen, M. Costantini, et al. OPCARE9. July 2011. “Artificial nutrition and hydration in the last week of life in cancer patients. A systematic literature review of practices and effects.” Annals of Oncology 22(7): 1478–​1486. Rapoport, A., J. Shaheed, C. Newman, M. Rugg, and R. Steele. 2013. “Parental perceptions of forgoing artificial nutrition and hydration during end-​of-​life care.” Pediatrics 131(5): 861–​869. Rochon, P. A., S. S. Gill, J. Litner, M. Fischbach, A. J. Goodison, and M. Gordon. May 1997. “A systematic review of the evidence for hypodermoclysis to treat dehydration in older people.” Journal of Gerontology A, Biological Sciences and Medical Sciences 52(3): M169–​M176. Rosner, F., and N. Abramson. March 2009. “Fluids and nutrition: Perspectives from Jewish Law (Halachah).” Southern Medical Journal 102(3): 248–​250. Sampson, E. L., B. Candy, and L. Jones. April 2009. “Enteral tube feeding for older people with advanced dementia.” Cochrane Database of Systematic Reviews 15(2): CD007209. Sasson, M., and P. Shvartzman. 2001. “Hypodermoclysis: An alternative infusion technique.” American Family Physician 64: 1575–​1578. Schostak, Z. 1994. “Jewish ethical guidelines for resuscitation and artificial nutrition and hydration of the dying elderly.” Journal of Medical Ethics 20: 93–​100. Schwartz, J. 2007. “Exploring the option of voluntarily stopping eating and drinking within the context of a suffering patient’s request for a hastened death.” Journal of Palliative Medicine 10: 1288–​1297. Shannon, T. A., and J. A. Walter. 2005. “Assisted nutrition and hydration and the Catholic tradition.” Theological Studies 66: 651–​662. Shapiro, D. S., and R. Friedmann. 2006. “To feed or not to feed the terminal demented patient: Is there any question?” Israeli Medical Association Journal 8(7): 507–​508. Soriano, A., and M. P. Davis. March 2011. “Malignant bowel obstruction: individualized treatment near the end of life.” Cleveland Clinic Journal of Medicine 78(3): 197–​206. Sorokin, R., and J. E. Gottlieb. September-​October 2006. “Enhancing patient safety during feeding-​tube insertion: A review of more than 2,000 insertions.” Journal of Parenteral and Enteral Nutrition 30(5): 440–​445. Spataro, R., L. Ficano, F. Piccoli, and V. La Bella. May 15, 2011. “Percutaneous endoscopic gastrostomy in amyotrophic lateral sclerosis: Effect on survival.” Journal of Neurological Science 304(1–​2): 44–​48. Stayner, J. L., A. Bhatnagar, A. N. McGinn, and F. C. Fang. December 2012. “Feeding tube placement: Errors and complications.” Nutrition in Clinical Practice 27(6): 738–​748. Sulmasy, D. P. 1997. “Futility and the varieties of medical judgment.” Theoretic Medicine 18: 63–​78. Sulmasy, D. P. 1998. “Killing and allowing to die: Another look.” Journal of Law and Medical Ethics 26: 55–​64. Sulmasy, D. P. 2002. “Informed consent without autonomy.” Fordham Urban Law Journal 30: 207–​220. Sulmasy, D. P. 2005a. “Double effect reasoning and care at the end of life: Some clarifications and distinctions.” Vera Lex 6(1–​2): 107–​145. Sulmasy, D. P. 2005b. “Terri Schiavo and the Roman Catholic tradition of forgoing extraordinary means.” Journal of Law and Medical Ethics 33: 359–​362. Sulmasy, D. P. July-​August 2006. “End of life care revisited.” Health Progress 87(4): 50–​56.

The Ethics of Medically Assisted Nutrition and Hydration    153 Sulmasy, D. P. 2007a. “ ‘Re-​inventing’ the rule of double effect.” In The Oxford handbook of bioethics, edited by B. Steinbock. Oxford: Oxford University Press, 114–​149. Sulmasy, D. P. December 7, 2007b. “Preserving life? The Vatican and PVS.” Commonweal CXXXIV (21): 16–​18. Sulmasy, D. P., and E. D. Pellegrino. 1999. “The rule of double-​effect: Clearing up the double talk.” Archives of Internal Medicine 159: 545–​550. Vacco v. Quill, 521 U.S. 793 (1997). Washofsky, M. Fall 2005. “A Jewish guide to the moral maze of hi-​tech medicine.” Reform Judaism. http://​reformjudaismmag.org/​Articles/​index.cfm?id=1048. Last accessed Dec. 10, 2015. Wiggenraad, R. G., L. Flierman, A. Goossens, et al. 2007. “Prophylactic gastrostomy placement and early tube feeding may limit loss of weight during chemoradiotherapy for advanced head and neck cancer, a preliminary study.” Clinical Otolaryngology 32: 384–​390.

Chapter 9

Disorders of C onsciousne s s a nd N euro-​P alliat i v e  C a re  Toward an Expanded Scope of Practice for the Field Joseph J. Fins and Maria G. Master

Severe Brain Injury and Palliative Care Two decades ago, when Daniel Callahan responded to those advocating a right to physician-​assisted suicide, he argued that to appeal to that right was an illusory attempt to control the timing and manner of one’s death. More precisely, he asserted that the will to control nature, to micromanage one’s existential fate, was “self-​determination run amok” (Fins and Pohl, in press).1 It was, in Callahan’s view, an atomistic vision of the self within the broader context of patients, families, and biology. Despite the legalization of physician-​assisted suicide in some jurisdictions, Callahan’s argument about biology and the limits of control still ring true when considering severe brain injury, the topic of this chapter. Unlike the dichotomous outcomes that inform Hollywood portrayals (Fins 2007), heroic efforts to save the lives of patients with severe brain injury result in intermediate outcomes. There are many form fruste recoveries of patients who have survived but done so with impairments that most would not choose if there had been the choice (Fins and Schiff 2010), if there somehow had been the possibility to prospectively avoid an unwanted fate after injury or illness struck. In the forthcoming book, Rights Come to Mind: Brain Injury, Ethics and the Struggle for Consciousness (Fins 2015b) one of us (JJF) recounts how families newly confronted with the existential threat posed by severe brain injury make medical choices. In more than fifty Internal Review Board (IRB)-​approved interviews with more than fifty families

Disorders of Consciousness and Neuro-Palliative Care    155 conducted under the aegis of the Consortium for the Advanced Study of Brain Injury (CASBI) at Weill Cornell Medical College and The Rockefeller University (Consortium for the Advanced Study of Brain Injury), we have begun to understand how families confront the sudden challenge of brain injury. Their initial overriding concern is survival. Worries about quality of life follow weeks and months later, often after patients are no longer dependent on life support but are left with a profound symptom burden. Through these narratives, we have uncovered a crisis in social compassion and care, both for the families and their marginalized loved ones. While advanced technology is publicly heralded for saving lives through novel and heroic means, its aftermath has been neglected and calls for a response. These patients have not chosen the condition that technology has left them to endure. Many have been warehoused and ignored without social consequence. Their families and surrogates struggle to define the reality of their loss in a medical and social system that has neither offered them closure nor made vigorous efforts at supportive and rehabilitative care. This degree of neglect has profound implications for the palliative care community, which is well-​positioned to make a difference. It is critical that those who work in palliative medicine expand their mission, scope of care, and legacy of service. Patients in liminal states of consciousness may perceive pain and potentially suffer. Families, tethered to the bedside, endure grief and sometimes guilt about the choices that have been made which have so burdened their loved ones and their family (Fins and Hersh 2011). Our health-​care system is inhospitable to these patients and families and their needs. No longer with acute care requirements after hospital discharge, and too often beyond the reach of rehabilitative benefits for which they either are deemed not qualified or which are too quickly consumed, these patients and families end up sequestered in chronic care, often suffering from inadequate medical and supportive care (Fins 2003a). Like the dying patient in an acute care hospital in need of hospice, patients with severe brain injury find themselves displaced in the care system in part because of the role their stories have played in making American law and establishing the right to die (Fins 2006a). That legacy, ensconced in Quinlan, Cruzan, and Schiavo, and the ideological battles that followed over the right to die, have politicized severe brain injury on both the right and left, leading to unfortunate misconstruals about patient and family needs and, in some cases, abandonment. The centrality of brain injury to the right to die began with the landmark case of Karen Ann Quinlan over the right of surrogates to withdraw life-​saving therapy on behalf of a vegetative patient. In Quinlan, Chief Judge Hughes agreed to the request of Karen’s parents to remove her ventilator. The decision was significantly influenced by the expert testimony of the court-​appointed neurologist, Fred Plum. Plum attested that Karen’s condition was irreversible and that additional treatment would be futile to restore the neurological function underlying her ability to think and feel. Accordingly, Hughes concluded that if her medical providers and the hospital ethics committees decided that “there is no reasonable possibility of Karen’s ever emerging from her present comatose condition to a cognitive, sapient state … [then] the present life-​support system may be

156    Joseph J. Fins and Maria G. Master withdrawn”2 This notion of futility linked to one’s cognitive sapient state became the moral warrant for the establishment of a right to die. Over the ensuing decades, of course, the right to die expanded. Although it started with severe brain injury, it has been generalized to all patients or their surrogates who express a preference to withhold or withdraw life-​sustaining therapy. But, despite this expansion, the association of the establishment of a right to die continues to be linked to the vegetative state and the legacy of the Quinlan case and successor cases like Cruzan and Schiavo (Annas 1996). Unfortunately, patients with disorders of consciousness (DOC) suffer from extremes of “care” motivated by ideological commitments that are often inadequately reflected on. We need to preserve the right to die and affirm the right to care for patients and families who want that care (Fins 2006b). As Schiavo illustrated, vegetative patients may be prone to receiving futile care in violation of their preferences or virtually abandoned. Or, as we will discuss, sentient patients with the potential for recovery may be misdiagnosed as vegetative when they are in fact minimally conscious. We suggest that the same ethical rationale first developed in the Quinlan case—​ predicated on a potential for consciousness—​that allows the vegetative patient a right to die in fact exhorts us to support the right to care for those who are minimally conscious. These vulnerable patients (and their families) need less polemical debate and more focused attention on their care needs. They need what in Spanish is called a refugio, a refuge, which referred to the roadside stopping places, the primitive hospices, that lined the medieval pilgrimages of the sick to Santiago de Compostela. Both metaphorically and actually, patients with severe brain injury and their families need the reservoir of expertise and the deep well of compassion that exists in the palliative care community to comprehensively meet their needs and achieve what one of us (JJF) has previously described as neuro-​palliative care (Fins and Pohl in press). To realize this vision, this expanded scope of palliative care practice, we will define and review brain states constituting DOC, their differential biology, diagnostic challenges, social impact, and emerging therapeutic opportunities. We will then explore advances in technology that may assist us in meeting the needs of certain patients with DOC. And we will conclude by suggesting ways to integrate the needs of these patients in both the local and national palliative care infrastructure. Throughout, we will affirm the personhood of those with liminal states of consciousness and the moral obligation to reintegrate them into our communities and nexus of care.

The Diagnostic Challenge of the Minimally Conscious State Over recent decades, increased use of measures that decrease intraventricular pressure, such as intraventricular shunts and craniectomy, have increased the prevalence of severely injured patients who in the past would have likely become vegetative or died

Disorders of Consciousness and Neuro-Palliative Care    157 from herniation. Instead, these therapeutic measures have mitigated mortality risk and produced a class of patients experiencing a liminal condition known as the minimally conscious state (MCS), a state functionally above and distinct from the vegetative state, with which it can be mistakenly conflated. The need for an expanded scope of palliative care stems from the challenges posed by MCS and the changing epidemiology of severe brain injury associated with this condition. MCS formally entered the literature as a diagnostic category when the Aspen Criteria were published in Neurology in 2002 (Giacino et al. 2002). MCS patients can be easily confused with those who are vegetative, but there is a fundamental difference between these states: a MCS patient is conscious. A vegetative patient is not. Unlike the vegetative patient whose neurobiology is classically described as intact brain stem function without higher cortical function, MCS patients exhibit definite evidence of awareness of self, others, or the environment. In contrast to the vegetative patient’s “wakeful unresponsiveness,” first described by Jennett and Plum in their classic 1972 Lancet (Jennett and Plum, 1972) article describing a permanent vegetative state (PVS) as a “syndrome without a name,” MCS patients interact with their environment. They will show intention, attention, or memory. Sometimes they may say an occasional word, grasp for a ball, or track someone who enters the room. The trouble—​the profound challenge—​of MCS is that these patients demonstrate such behaviors intermittently and unreliably in response to commands. This makes these patients difficult to identify and properly diagnose, thus leaving them outside of clinical view.

The Differential Diagnosis of DOC This omission can be rectified, in part by a better understanding of the underlying neurobiology of this disorder that distinguishes MCS from the vegetative state and leads to differing ethical obligations (Fins 2008, 2010).3 Experiments suggest important functional differences in the MCS brain that may indicate active internal networks that could form the basis for consciousness; this compels a reassessment of the ethical imperative for care and comfort. Positron emission tomography (PET) scans have afforded a window into the brain beyond the clinical bedside. Stephen Laureys and colleagues used this technique to examine the pain response in vegetative patients (Laureys et  al. 2002). That study compared cerebral activation in response to painful stimuli in fifteen vegetative state patients (twelve with nontraumatic insults and mean post-​insult time of one month) and fifteen normal volunteers. The results demonstrated preserved and robust activation of the brainstem, thalamus, and primary somatosensory cortex in all subjects, but, unlike the controls, whose responses activate a broad pain network widely distributed in the brain, vegetative patients were “functionally disconnected” from the broader pain network, suggesting a loss of functional integration (Fins and Schiff 2007). In contrast to the vegetative state, MCS patients have the capability for a network response to the environment. Their brains appear to show more widespread activation

158    Joseph J. Fins and Maria G. Master when receiving external stimulation. When assessed by neuroimaging methods, the MCS brain responds to auditory stimulation with integrated processing and preserved neural connectivity between low-​level sensory areas and the frontal parietal regions that are thought to be involved with the emergence of conscious perception. Unlike vegetative patients, MCS subjects experienced large-​scale network activation in response to hearing spoken narratives, much like normal control subjects (Demertzi, Laureys, and Boly 2009; Schiff et al. 2005). A useful analogy might be to imagine the brain as if it were an interstate highway system (Fins in press). In MCS, the roadways are serviceable but no one is traveling on them. In the vegetative state, the system is not intact. The potential for travel—​in our metaphor, the activation of widely disseminated networks—​is lost once the vegetative state becomes a PVS, but it remains in MCS. In MCS, these networks are underactivated but potentially full of unrealized potential if stimulated. To return to the metaphor, someone needs to drive on the highway. Recent studies have sought to activate these dormant networks through the rubric of what has since been described by Schiff and Posner as the mesocircuit (Schiff 2010a). These efforts have demonstrated the ability of drugs like zolpidem (Whyte and Myers 2009) and amantadine (Giacino et al. 2012) and interventions like thalamic deep brain stimulation to promote activation of the circuitry putatively underlying consciousness (Fins 2000, 2003b; Schiff and Fins 2007a; Schiff et al. 2006, 2007). These studies highlight a clinical dilemma: MCS patients present themselves as integrated when stimulated, yet when an MCS brain is dormant the patient is behaviorally—​ or overtly—​difficult to distinguish from that of a vegetative patient. What differentiates MCS from the vegetative state then is its potential for activation and integration. It is this potentiality that carries great ethical salience because this network response is the basis for consciousness. The response of MCS patients to language (Schiff et al. 2005)—​manifested through neuroimaging—​suggests that the MCS patient is, at some level, perceiving and responding to the world. This raises the possibility that patients in a MCS may additionally have the ability to appreciate their isolation and segregation from family, friends, and community. When MCS and vegetative patients are conflated, the consciousness of the former is devalued, and their potential to experience pain and perhaps higher degrees of isolation and suffering is ignored. Such errors, whether willful or through neglect, need to be rectified, and this imperative forms the basis for articulating and creating a neuro-​palliative ethic of care for this long-​neglected population (Fins 2005; Fins and Pohl in press).

The Evolution of DOC How might we come to identify those brain-​injured patients in need of neuro-​palliative care? Diagnostic assessment requires the proper use of an instrument specifically designed to assess brain disorders where consciousness but not motor control may be

Disorders of Consciousness and Neuro-Palliative Care    159 available. The JFK-​Johnson Coma Recovery Scale-​Revised (CRS-​R), devised by Giacino et al., captures the cognitive functions of a patient that may be missed by conventional instruments like the Glasgow Coma (GCS) and Outcomes Scales (GOS), which measure behavior and do not assess the presence of consciousness (Giacino, Kalmar, and Whyte 2004; Teasdale and Jennette 1974). The Glasgow scales penalize patients who may be conscious but have limited motor output. For example, a locked-​in patient, with normal cognitive function but no motor output below his spinal or brainstem injury, might be erroneously classified as unconscious when the problem is motor output and not the absence of consciousness. Similarly, the MCS patient with subtle but verifiable signs of consciousness can be missed by the Glasgow scales. Repeated exams are also necessary to capture the intermittent displays of awareness seen in MCS. An isolated exam will inevitably miss evidence of consciousness that waxes and wanes throughout the day. Testing ideally at different times of the day allows the observer to capture the patient’s highest level of function. As critical as multiple examinations to establish a diagnosis is the need to understand that DOC evolve over time. Clinicians need to appreciate this evolution in order to guide families through decision-​making. There is a trajectory of recovery followed by patients who have had an injury capable of producing a sustained loss of consciousness. Generally, this process begins with coma, which is an eyes-​closed state of unresponsiveness affecting the brainstem and both hemispheres. Unless medically induced, comas are self-​limited, lasting 10–​14 days. A coma may either lead to brain death (whole brain death involving brainstem and higher cortical function), resolve with complete recovery (as in anesthetic coma), or progress to the vegetative state, in which the brainstem begins to regain function in the absence of concomitant higher function (Schiff 2009). Patients may then progress to MCS. Drawing on data from the 1994 Multi-​Society Task Force, a patient is in a persistent vegetative state when vegetative for one month. When the persistent vegetative state becomes permanent depends on the etiology of injury: three months after anoxic injury and twelve months after traumatic injury (Multi-​Society Task Force on PVS 1994). Generally, patients with traumatic injury fare better than those who are anoxic, with 50% of traumatic brain injury (TBI) patients achieving an outcome of MCS or higher, whereas 77% of patients with anoxic comas will either die or remain permanently vegetative (Posner, Saper, Schiff, and Plum 2007). Of note, predictive models do not exist for those patients who have been treated with therapeutic hypothermia after cardiac arrest. Cooling the brain exerts a protective function that has made the Levy Criteria for post-​ arrest outcomes obsolete for this clinical scenario (Hypothermia after Cardiac Arrest Study Group 2002; Scirica 2013; Stevens and Sutter 2013). Before the permanence of the vegetative state sets in at three or twelve months, patients may transition to MCS and retain the potential for functional communication (Giacino 2005). If communication becomes consistent and reliable, the patient is said to have emerged from MCS. Emergence has been documented as late as 19 years after a TBI, as in the case of Terry Wallis (Schiff and Fins 2007b). How patients evolve out of the MCS is not well understood, and, to date, we are unable to predict who will embark

160    Joseph J. Fins and Maria G. Master on this next stage of a recovery trajectory (Schiff 2010b). Both the wide range of additional of recovery and the variability of these time frames makes prediction difficult, as Lammi’s data suggest (Lammi, Smith, Tate, and Taylor 2005). Because an MCS patient retains the prospect of recovery, even years and decades after the injury (Lammi et al. 2005), mistaking a patient’s brain state as vegetative when it is actually minimally conscious represents a serious ethical omission. Nonetheless, the literature reports unconscionable misdiagnosis rates (Andrews et al. 1996; Childs, Mercer and Childs 1993; Wilson et al. 2002). More recently, Schnakers et al. found that 41% of nursing home patients diagnosed as vegetative were, in fact, in MCS after more intense clinical evaluation (Schnakers et  al. 2009). The impact of these diagnostic errors on a patient’s lived experience and potential outcome cannot be overstated and ethically represent a serious ethical omission that needs to be understood in order to be addressed. Typically, the transition to MCS occurs after the patient has been discharged to a chronic care facility because the clinical exam did not reveal enough early evidence of improvement to qualify for rehabilitation (Fins, Master, Gerber, and Giacino 2007). As Schnakers et al. report, this transition can easily go unnoticed (Schnakers et al. 2009). The mix of a low-​acuity setting coupled with the episodic nature of behaviors that would mark a patient as minimally conscious versus vegetative present a challenge to practitioners. Families who report behaviors to clinical staff that then are not repeated upon command are ascribed to denial and wishful thinking and are ignored as evidence of the biology of MCS. The propensity to view observations in the chronic care setting in this way is reinforced by traditional medical hierarchies that reinforce the power of the discharge diagnosis to shape the perceptions of those following up with care within a nursing home setting. The preeminence of the acute care setting, with its technologically intensive diagnostic tools and heroic measures for intervention, casts a long shadow over subsequent caregivers who often labor to serve patients under less glorious, and typically less well funded, conditions. Who are they (usually geriatricians or internists) to challenge a diagnosis received from the tertiary care facility made by neurologists and neurosurgeons? After all, haven’t those physicians saved the patient’s life and provided skilled neurointensive care? The bias shifts away from direct observation in the nursing home setting when a patient arrives prepackaged and preemptively labeled as vegetative. Given the structure of our health-​care system and the nature of follow-​up for patients with severe brain injury after the acute phase of their injury, most patients will be far from the academic medical centers when and if they progress to MCS. This progression can be missed when patients are discharged from acute care medical centers with an authoritative diagnosis of being in the vegetative state and begin to emerge in less medically intensive venues. In those settings, the power of the initial—​authoritative academic—​diagnosis may be difficult to question or overcome in the face of episodic and inconsistent evidence of the patient’s awareness of self, environment, or others. Notations by hopeful families who observe evidence of awareness can be too easily

Disorders of Consciousness and Neuro-Palliative Care    161 dismissed as wishful thinking or denial. Brief observations by rotating physicians are likely to be inadequate to capture the intermittent but definite evidence of awareness of the MCS state. Conceptually, the discharge diagnosis is also not challenged because most receiving physicians do not appreciate that there is an evolution of brain states associated with DOC. Most receiving physicians assume that a discharge diagnosis of vegetative state would be permanent, as might a diagnosis of Alzheimer’s disease or heart failure. What clear and convincing evidence must be presented, and by whom, to allow a clinician to begin to appreciate an evolution of brain states when the academic medical center had not even considered the possibility? Moreover, most clinicians have limited skills and experience in diagnosing and caring for the MCS patient. Many clinicians still do not appreciate that the persistent vegetative state does not become permanent before three or twelve months. Often the closest person to the patient in a nursing home setting is a nurse who may not be institutionally supported in questioning the authority of the academic medical center that discharged the patient from acute care. But the VS diagnosis is only fixed three to twelve months after the precipitating insult. Short of that time frame, the patient’s brain state is syndromic and does not have a fixed diagnosis. Before permanence sets in, patients have the potentiality to move into MCS. We now know that an accurate and timely diagnosis of MCS is a critical step in the trajectory of the patient’s life. Capturing the potential of a patient to regain consciousness early on can signal hope for the family and energize rehabilitation efforts in an otherwise bleak situation. The neurobiological capacities of the MCS potentially include the ability to experience a range of emotions, needs, pains, and pleasures. Once identified, MCS patients might even be returned to their family and community. Our moral obligations to a sentient being, one able to interact with and be aware of herself and others (an MCS patient), should be different from what is owed to the patient who is unconscious (a vegetative patient) (Fins 2009). The difference may appear negligible at first glance, until we focus on the fact that the MCS patient is conscious and a vegetative patient is not. An accurate diagnosis of MCS is therefore an ethical imperative because it indicates a patient’s ability to experience pain and also a latent capability for emergence and fuller recovery.

Neuroimaging and Neuroethics Over the past two decades, advances in neuroimaging have prompted a growing awareness of the potential discordance between brain states as discerned by behavioral metrics or motor output and what might be manifested on a scan (Fins 2010).4 This poses a threat to established diagnostic criteria that are based in the clinical exam and also presents an opportunity to use neuroimaging as a neuroprosthetic tool for communication with those who lack motor activity (Banja and Fins 2013).

162    Joseph J. Fins and Maria G. Master For example, a case published by Owen and colleagues at the Universities of Cambridge and Liege showed that a patient considered vegetative could respond to active language paradigms through the guise of neuroimaging (Fins and Plum 2004; Owen et al. 2006). The case involved a twenty-​five-​year-​old woman who appeared behaviorally vegetative as a result of a TBI six months earlier. Investigators asked her a series of questions while measuring brain activity with functional magnetic resonance imaging (fMRI). She was asked to imagine playing tennis in her head, walking through her home, and disambiguating two linguistically similar words. Remarkably, despite a vegetative diagnosis based on behavioral criteria, fMRI revealed activity in motor, spatial, and language networks that was identical to normal subjects. Using this “active language” paradigm, Monti and colleagues used fMRI to link these tasks with Yes/​No responses (Monti et al. 2010). One patient with TBI who was diagnosed as vegetative by behavioral criteria was able to reliably answer questions using this method. By this means, fMRI may have opened a communication channel for someone otherwise thought to be inaccessible. Bardin et al. added a further wrinkle to observed discordance between behavioral findings and a diagnosis discerned by experimental neuroimaging methods (Bardin et al. 2011). Using an active paradigm similar to that used by Monti et al. (2010), Bardin and colleagues found that responses can be present but sometimes outside the anticipated region of interest (ROI) and delayed (Bardin et al. 2011; Fins 2012; Fins and Schiff 2006). These discoveries and paradoxical findings prompt the question of whether behavioral or neuroimaging data should be used to make a diagnosis. The same patient could be deemed vegetative under one method and minimally conscious under the other method. Should one form of assessment take precedence over the other? How ought we interpret the findings of neuroimaging as they may bear on our understanding of a patient’s consciousness and agency? Can the lens of neuroimaging provide the same grounds for diagnosis as behaviors currently do? To the degree that diagnosis yields prognosis, this dilemma suggests an evolving need for an ever more nuanced approach. As proposed in a paper in the Hastings Center Report, patients who are behaviorally vegetative but demonstrate active responses on neuroimaging may well be described in a new category of nonbehavioral MCS (Fins and Schiff 2006). As such, the subject’s highest level of function—​whether manifested behaviorally or via neuroimaging—​should dictate the patient’s diagnosis. Having called for a role for neuroimaging in parallel to behavioral assessment, it should be noted that behavioral evaluation remains the most sensitive and specific mode of assessment when the CRS-​R assessment tool is used to evaluate patients (Giacino et al. 2004, 2009). Indeed, the patient studied by Monti et al. was subsequently identified to be MCS by behavioral criteria using the CRS-​R (Giacino et  al. 2004), thus confirming the MCS diagnosis suggested by neuroimaging. For our purposes, it is important to appreciate the ethical obligation to identify consciousness when it is present by making use of both behavioral assessment and emerging neuroimaging methods.

Disorders of Consciousness and Neuro-Palliative Care    163

Clinical Challenges in Neuro-​Palliative Care Recognizing that MCS patients are conscious and potentially able to appreciate their environment and exert personal agency, we are faced with a complicated ethical mandate to reintegrate them back into the human community from which they were exiled as a consequence of injury (Fins 2015a). The palliative challenge to minimizing pain and suffering extends beyond the somatic and into the psychological. Meeting the psychological needs of these patient begins with engaging the family system of care that is so often strained by a severe brain injury.

Family Systems The essential point of contact between the patient with DOC and the larger community is often the family (Master and Fins, in progress).5 It is the family who characterizes the patient and provides the advocacy around which critical treatment plans evolve. And it is the family who is left to care for the patient with DOC for potentially years to come, often increasingly isolated and outside the view of professional health-​care providers. Moving then from an atomistic to a molecular approach of the patient experience, we need to begin with a consideration of families and MCS. The need for family intervention is amply supported by numerous clinical studies revealing correlations between psychiatric pathology in the family and a member with brain injury. Although any illness may expose vulnerabilities, families with a brain-​ injured member report more enduring patterns of family dysfunction than do families managing better described medical conditions (Curtiss, Klemz, and Vanderploeg 2002; Suppes and Fins 2013). This psychological burden may have multiple etiologies rooted in part within the shifting family system dynamics. Parents may lose the joy of an anticipated developmental trajectory for their young child or struggle under the regression of an adult child to a state of chronic dependency. An injured spouse may leave the survivor to care for the entire family while also shouldering the burden of a deep sense of loss and abandonment. Sibling dynamics, previously fueled by a healthy sense of rivalry balanced by shared responsibility, now are potentially offset by guilt, resentment, or diminished parental guidance and stamina. Recognizing these profound stressors, a nuanced formulation of family dynamics highlighting risk factors and opportunities for resilience can be a foundational basis for providing meaningful intervention and support. Families may also progress along a temporal trajectory from an initial stage of shock and relief at the survival of their loved one to a longer phase of ambivalence, where hope and anticipated grief fluctuate in an evolving attempt at psychic equilibrium (Curtiss et al. 2002; Suppes and Fins 2013). Shaping that trauma, MCS is often first discerned by the

164    Joseph J. Fins and Maria G. Master family members who are in closest connection with the patient and able to identify evidence of attending and intention, command following, or other purposeful behaviors contingently responsive to external stimuli (Wade et al. 1998; 2006). Yet families often report that their observations are minimized, rather than supported by health-​care providers (Fins 2015b; Fins and Hersh 2011). Where the family recognizes evidence of consciousness, however, the relationship to the patient may be more hopefully engaged and emotionally sustaining (Wade et al. 1998, 2006). Diagnostic accuracy of the MCS could thus validate a family’s hopeful perception of a meaningful relationship with their loved one (Fins 2015b). Supporting resilience and relationship dynamics shape expectations of recovery—​ and demands for care. Those families with the most robust sense of inclusion and connection with their injured family members may be more emotionally empowered to advocate for innovative and emerging treatments. They may be most resistant to the corrosive impact of depression, anxiety, or despair. Paradoxically, they may be at greatest risk of social isolation because they cannot engage in fully grieving their loss and moving away from the trauma (Suppes and Fins 2013; Shear and Shair 2005). These dynamics require further exploration in the palliative care context, with the goal of understanding how best to enhance resilience and reinforce meaning in the setting of loss. These efforts could directly impact patient outcome and the well-​being of the family unit of care.

Suffering In addition, clinicians should be at least hypothetically concerned that some patients in the MCS have the potential to experience psychological suffering and an awareness of their loss. As discussed earlier, MCS patients have the ability to process language and recognize the emotional salience of familiar narratives. Some MCS patients, especially those with a disproportionate motor versus cognitive deficit, may have an intact sense of self and thus a potential for suffering, as eloquently described by Eric Cassell, which is an existential threat to the self (Cassell, 1991). The emotional toll even under the best of circumstances and prognoses also cannot be underestimated. Family dynamics shift and remodel over time. Loved ones move forward in life to develop parallel and no longer shared experiences with the MCS patient. The emerging MCS patient may be gradually regaining the ability to engage the world only to face an unexpected, profound isolation and loss of social meaning (Fins 2015b). Meeting the supportive and psychotherapeutic needs of such patients is essential to a meaningful recovery. With limited prognostic models for the MCS patient, the clinician may need to intuitively consider the potential for psychic pain and suffering. This may require assuming that the patient can in fact hear the clinician at the bedside and shaping discussions to recognize the impact of any comments and discussions about the patient’s condition. And, more speculatively, it will require thinking about the psychological impact of isolation and loneliness and developing rehabilitative psychotherapy for these patients in tandem with their families.

Disorders of Consciousness and Neuro-Palliative Care    165

Pain and Symptom Management Finally, a fundamental concern after a TBI is that of pain management. How do we understand pain in a patient who is not able to speak for himself? How do we focus our efforts appropriately on the patient who is capable of awareness yet appears superficially to be as impervious to suffering as the vegetative patient? The MCS patient may be in unyielding pain yet be unable to seek relief, a problem only compounded by the prevailing rate of misdiagnosis that places these patients into the vegetative state, a nosological category defined as one unable to experience pain. Because the MCS patient is generally unable to communicate discomfort, whether physical or psychological, clinicians have substantially relied on the impressions of family members, with the hope that these projections reflect the subjective experience of the patient (Phipps, DiPasquale, Blitz, and White 1997; Phipps and Whyte 1999). Although most experts rely on the CRS-​R scale to measure levels of consciousness (Giacino et al. 2004), to date, we have no pain scales designed for the unique situation of the MCS patient. The CRS-​R does not elicit information about the quality of a patient’s emotional or physical distress, but it does provide vital information about a patient’s potential, as a conscious being, to experience these emotions. We do know that the trajectory of the MCS medical course suggests multiple vectors for pain or discomfort. The trauma that initiated the brain injury may also have left the patient with fractures, wounds, and other superficial or internal bodily injuries. Patients may have recurrent neurological seizures as a result of the brain insult. As efforts move forward to save the patient, he may further be subjected to invasive procedures, intubations, and central lines. And, as time progresses, these patients endure medical issues related to their prolonged immobility, such as muscle contractures, blood clots, feeding aspiration, or bedsores. Our challenge is to best provide palliative care in a situation in which the pain connected to these conditions may be unarticulated or unnoticed (Fins and Pohl in press).

Toward a Neuro-​palliative Ethic of Care Society has designated neither role nor status to the patient who remains in the MCS condition of intermittent awareness—​potentially for years or decades—​yet retains the ability to perceive self and to suffer or even know joy. With emerging technological advances, they may even be able to communicate their wishes for care. But, at this time, to the untrained eye, they appear irretrievably vegetative and lost to their community. And without the public will to uncover them, this class of citizens remain isolated, refused adequate rehabilitation, and denied their basic human rights.

166    Joseph J. Fins and Maria G. Master Instead of reiterating ideological positions around end-​of-​life care that do not meet the needs of patients or families, as was the case during the Schiavo debate (Fins 2006b, 2015a,b; Fins and Plum 2004; Fins and Pohl in press), we should strive to both preserve the right to die for those who are beyond hope of recovery while affirming the right to care for those who might benefit from coming advances in neuroscience. Distinguishing between the MCS and the VS, and beginning to clarify for families and their health-​care providers the potential prognostic range for these conditions, is an essential step. Palliative medicine, as an ethic and system of care, has a singular opportunity to bring its experience and expertise to patients and families whose lives have been inalterably changed by a disorder of consciousness (Fins 2005; Fins and Pohl in press). By further articulating a neuro-​palliative ethic of care consistent with the best traditions of the field (Fins 2006b), practitioners can bring relief and solace to a marginalized community in need of clinical engagement and compassion.

Notes 1. The discussion on diagnostic and ethical challenges draws from Fins and Pohl (in press). 2. In Re Quinlan 70 N.J. 10; 355 A.2d 647; 1976 N.J. LEXIS 181; 79 A.L.R.3d 205. 3. This section is drawn from Fins (2010). 4. This section is drawn from Fins (2010). 5. This section draws from Master and Fins (in progress).

References Andrews, K., Murphy, L., Munday, R., et  al. 1996. Misdiagnosis of the vegetative state: Retrospective study in a rehabilitation unit. British Medical Journal 313: 13–​16. Annas, G. 1996. “The ‘right to die’ in America: Sloganeering from Quinlan and Cruzan to Quill and Kevorkian.” Duquesne Law Review 34: 875–​897. Banja, J. D., and Fins, J. J. 2013. Ethics in brain injury medicine. In Brain injury medicine, 2nd edition, edited by N. Zasler, D. I. Katz, and R. D. Zafonte. New York: Demos Medical Publishing, 1374–​1390. Bardin, J. C., Fins, J. J., Katz, D. I., et al. 2011. “Dissociations between behavioral and functional magnetic resonance imaging-​based evaluations of cognitive function after brain injury.” Brain 134: 769–​782. Cassell, E. 1991. The nature of suffering. New York: Oxford University Press. Childs, N. L., Mercer, W. N., and Childs, H. W. 1993. “Accuracy of diagnosis of persistent vegetative state.” Neurology 43: 1465–​1467. Curtiss, G., Klemz, S., and Vanderploeg, R. D. 2002. “Acute impact of severe traumatic brain injury on family structure and coping responses.” Journal of Head Trauma Rehabilitation 15(5): 1113–​1122. Demertzi, A., Laureys, S. and Boly, M. 2009. Coma, persistent vegetative states and diminished consciousness. Encyclopedia of Consciousness. Vol. 1. Amsterdam, The Netherlands: Elsevier, 147–​156.

Disorders of Consciousness and Neuro-Palliative Care    167 Fins, J. J. 2000. “A proposed ethical framework for interventional cognitive neuroscience: A consideration of deep brain stimulation in impaired consciousness.” Neurological Research 22: 273–​278. Fins, J. J. 2003a. “Constructing an ethical stereotaxy for severe brain injury: Balancing risks, benefits and access.” Nature Reviews Neuroscience 4: 323–​327. Fins, J. J. 2003b. “From psychosurgery to neuromodulation and palliation: History’s lessons for the ethical conduct and regulation of neuropsychiatric research.” Neurosurgery Clinics of North America 14(2): 303–​319. Fins, J. J. 2005. “Clinical pragmatism and the care of brain injured patients: Towards a palliative neuroethics for disorders of consciousness.” Progress in Brain Research 150: 565–​82. Fins, J. J. 2006a. A palliative ethic of care: Clinical wisdom at life’s end. Sudbury, MA: Jones and Bartlett. Fins, J. J. 2006b. “Affirming the right to care, preserving the right to die: Disorders of consciousness and neuroethics after Schiavo.” Supportive & Palliative Care 4(2): 169–​178. Fins, J. J. 2007. “The portrayal of coma in contemporary motion pictures.” Neurology 68: 79–​80. Fins, J. J. 2008. “Neuroethics and neuroimaging:  Moving towards transparency.” American Journal of Bioethics 8(9): 46–​52. Fins, J. J. 2009. Neuroethics and disorders of consciousness: A pragmatic approach to neuropalliative care. In The neurology of consciousness, cognitive neuroscience and neuropathology, edited by S. Laureys and G. Tononi. Burlington, MA: Elsevier, 234–​244. Fins, J. J. 2010. “Neuroethics, neuroimaging and disorders of consciousness: Promise or peril?” Transactions of the American Clinical and Climatological Association 122: 336–​346. Fins, J. J. 2012. “Wait, wait … don’t tell me: Tuning into the injured brain.” Archives of Neurology 69(2): 158–​160. Fins, J. J. 2015a. From the persistent vegetative state to the minimally conscious state: Ethical implications of disorders of consciousness. In Routledge Companion to Bioethics, edited by J. Arras, R. Kukla, and E. Fenton. New York: Routledge, 474–​485. Fins, J. J. 2015b. Rights come to mind: Brain injury, ethics and the struggle for consciousness. New York: Cambridge University Press. Fins, J. J. In press. Trastornos de Conciencia y Los Derechos Humanos: Una Nueva Frontera Ética y Científica. Solemne Sesión de Toma de Posesión como Académico de Honor del Dr. Joseph J. Fins. Con Laudatio del Excmo. Sr. D. Diego Gracia Guillen. 18 noviembre 2014. Madrid: Instituto de Espana, Real Academia Nacional de Medicina. Fins, J. J., and Hersh J. 2011. Solitary advocates: The severely brain injured and their surrogates. In Patients as policy actors (Critical issues in health and medicine), edited by B. Hoffman, N. Tomes, R. Grob, and M. Schlessinger. Piscataway, NJ: Rutgers University Press, 21–​42. Fins, J. J., Master, M. G., Gerber, L. M., and Giacino, J. T. 2007. “The minimally conscious state: A diagnosis in search of an epidemiology.” Archives of Neurology 64(10): 1400–​1405. Fins, J. J., and Plum, F. 2004. “Neurological diagnosis is more than a state of mind: Diagnostic clarity and impaired consciousness.” Archives of Neurology 61(9): 1354–​1355. Fins, J. J., and Pohl, B. R. In press. Neuro-​palliative care and disorders of consciousness. In Oxford textbook of palliative medicine, 5th edition, edited by G. Hanks, N. I. Cherny, N. A. Christakis, M. Fallon et al. Oxford: Oxford University Press. Fins, J. J., and Schiff, N. D. 2006. “Shades of gray:  New insights from the vegetative state.” Hastings Center Report 36(6): 8. Fins, J. J., and Schiff, N. D. 2007. Hope for comatose patients. In The Dana Foundation’s Cerebrum 2007, edited by C. Read. New York: Dana Press, 185–​203.

168    Joseph J. Fins and Maria G. Master Fins, J. J., and Schiff, N. D. 2010. “In the blink of the mind’s eye.” Hastings Center Report 40(3): 21–​23. Giacino, J., Childs, N., Cranford, R., et al. 2002. “The minimally conscious state: Definition and diagnostic criteria.” Neurology 58: 349–​353. Giacino, J. T. 2005. “The minimally conscious state:  Defining borders of consciousness.” Progress in Brain Research 150: 381–​395. Giacino, J. T., Kalmar, K., and Whyte, J. 2004. “The JFK Coma Recovery Scale-​ Revised: Measurement characteristics and diagnostic utility.” Archives of Physical Medicine and Rehabilitation 85(12): 2020–​2029. Giacino, J. T., Schnakers, C., Rodriguez-​Moreno, D., et  al. 2009. Behavioral assessment in patients with disorders of consciousness: Gold standard or fool’s gold? In Progress in brain research, edited by S. Laureys et al. Philadelphia: Elsevier, 177: 33–​48. Giacino, J. T., Whyte, J., Bagiella, E., et al. 2012. “Placebo controlled trial of amantadine for severe traumatic brain injury.” New England Journal of Medicine 366: 819–​826. Hypothermia after Cardiac Arrest Study Group. 2002. “Mild therapeutic hypothermia to improve neurologic outcome after cardiac arrest.” New England Journal of Medicine 346: 549–​556. Jennett, B, Plum F, 1972; “Persistent Vegetative State after Brain Damage.” Lancet 299 (7753): 734–​737. Lammi, M. H., Smith, V. H., Tate, R. L., and Taylor, C. M. 2005. “The minimally conscious state and recovery potential: A follow-​up study 2–​5 years after traumatic brain injury.” Archives of Physical Medicine and Rehabilitation 86(4): 746–​754. Laureys, S., Faymonville, M. E., Peigneux, P., et al. 2002. “Cortical processing of noxious somatosensory stimuli in the persistent vegetative state.” Neuroimage 17(2): 732–​741. Master, M., and Fins, J. J. In progress. Family dynamics following severe disorders of consciousness. Monti, M. M., Vanhaudenhuyse, A., Coleman, M. R., et al. 2010. “Willful modulation of brain activity in disorders of consciousness.” New England Journal of Medicine 362(7): 579–​589. The Multi-​Society Task Force on PVS. 1994. “Medical aspects of the persistent vegetative state.” New England Journal of Medicine, 330, 1499–​1508; 1572–​1579. Owen, A. M., Coleman, M. R., Boly, M., et al. 2006. “Willful modulation of brain activity in disorders of consciousness. Detecting awareness in the vegetative state.” Science 313(5792): 1402. Phipps, E., DiPasquale, M., Blitz, C. L., and White, J. 1997. “Interpreting responsiveness in persons with severe traumatic brain injury: Beliefs in families and quantitative evaluations.” Journal of Head Trauma Rehabilitation 12: 52–​69. Phipps, E., and Whyte, J. 1999. “Medical decision-​making with persons who are minimally conscious.” American Journal of Physical Medicine and Rehabilitation 78: 77–​82. Posner, J. B., Saper, C. B., Schiff, N. D., and Plum, F. 2007. Plum and Posner’s diagnosis of stupor and coma. New York: Oxford University Press. Schiff, N. 2009. Coma and other pathological disorders of consciousness. In Encyclopedia of Neuroscience, edited by G. Adelman. Philadelphia: Elsevier, 1143–​1147. Schiff, N. D. 2010a. “Recovery of consciousness after brain injury: A mesocircuit hypothesis.” Trends Neuroscience 33(1): 1–​9. Schiff, N. D. 2010b. “Recovery of consciousness after severe brain injury: The role of arousal regulation mechanisms and some speculation on the heart-​brain interface.” Cleveland Clinic Journal of Medicine 77: S27–​S33.

Disorders of Consciousness and Neuro-Palliative Care    169 Schiff, N. D., and Fins, J. J. 2007a. “Deep brain stimulation and cognition: Moving from animal to patient.” Current Opinion in Neurology 20(6): 638–​642. Schiff, N. D., and Fins, J. J. 2007b. Hope for “comatose” patients. “A cerebrum classic” in The Dana Foundation’s Cerebrum 2007, edited by C. A. Read. New York: Dana Press, 185–​203. Schiff, N. D., Giacino, J., Kalmar, K., et al. 2006. Partial restoration of behavioral responsiveness and arousal regulation by electrical stimulation of the human intralaminar thalamic nuclei. Society for Neuroscience 36th Annual Meeting. October 2006. Abstract. Schiff, N. D., Giacino, J. T., Kalmar, K., et al. 2007. “Behavioral improvements with thalamic stimulation after severe traumatic brain injury.” Nature 448(7153): 600–​603. Schiff, N. D., Rodriguez-​Moreno, D., Kamal, A., et al. 2005. “fMRI reveals large-​scale network activation in the minimally conscious state.” Neurology 64(3): 514–​523. Schnakers, C., Vanhaudenhuyse, A., Giacino, J., et al. 2009. Diagnostic accuracy of the vegetative and minimally conscious state: Clinical consensus versus standardized neurobehavioral assessment. BMC Neurology 9(35): 1–​5. Scirica, B. M. 2013. “Therapeutic hypothermia after cardiac arrest.” Circulation 127: 244–​250. Shear, K., and Shair, H. 2005. “Attachment, loss and complicated grief.” Developmental Psychobiology 47: 253–​267. Stevens, R., and Sutter, R. 2013. “Prognosis in severe brain injury.” Critical Care Medicine 41: 1104–​1123. Suppes, A., and Fins, J. J. 2013. “Surrogate expectations in severe brain injury.” Brain Injury 27(10): 1141–​1147. Teasdale, G., and Jennette, B. 1974. “Assessment of coma and impaired consciousness: A practical scale.” Lancet 2: 81–​84. Wade, S. L., Taylor, H. G., Drotar, D., Stancin, T., and Yeates, K. O. 1998. “Family burden and adaptation during the initial year after traumatic brain injury in children.” Pediatrics 102 (1pt1): 110–​116. Wade, S. L., Taylor, G. H., Yeates, K. O., et al. 2006. “Long-​term parental and family adaptation following pediatric brain injury.” Journal of Pediatric Psychology 31(10): 1072–​1083. Whyte, J., and Myers, R. 2009. “Incidence of clinically significant responses to zolpidem among patients with disorders of consciousness: A preliminary placebo controlled trial.” American Journal of Physical and Medical Rehabilitation 88: 410–​418. Wilson, F. C., Harpur, J., Watson, T., et al. 2002. “Vegetative state and minimally responsive patients—​ Regional survey, long-​ term case outcomes and service recommendations.” NeuroRehabilitation 17: 231–​236.

Chapter 10

Ethical I s su e s i n Pro gnosi s a nd Pro gnost i c at i on Alexander K. Smith and Paul Glare

Introduction Making predictions is part of every domain of life, including health care, and when the prediction of health-​care outcomes influences the decisions that are made or the actions that are taken as a result of the prediction, these predictions take on ethical dimensions. In health care, the predicted outcome of a disease is referred to as its prognosis and the clinical act of predicting as prognostication. Any outcome of a disease can be prognosticated, including recovery, recurrence, progression, or death. Within the palliative care and hospice community, however, when people talk about “the prognosis” they are usually referring to the prediction of the time to death. Aside from its natural history, the other outcomes of an illness that may be predicted include the benefits, risks, and burdens of its treatment options; the cost of treatment and care; and the impact of the illness on others. Making a prognosis has long been held to be one of the three cardinal clinical skills of the competent physician, alongside diagnosis and treatment. Prognosis dwindled in importance in the twentieth century as the development of curative treatments for many previously untreatable conditions made the study of the natural history of diseases and the formulation of prognoses obsolete. Diagnosis, treatment, and prognosis played roughly equal parts early in the history of contemporary medicine. In a study describing the portion of medical textbook chapters devoted to prognosis, Nicholas Christakis found that early editions devoted roughly equal portions to these three areas (Christakis and Iwashyna, 1998). Over time, as we developed treatments for conditions that were previously fatal, such as pneumonia, the attention and space allocated to prognosis in medical textbooks waned.

Ethical Issues in Prognosis and Prognostication    171 However, in the latter decades of the twentieth century the burden of chronic incurable illnesses such as cancer, heart failure, and dementia led to some renewed interest in prognostication. This was largely stimulated by the creation of the hospice benefit, which required an expected survival of less than six months for eligibility (Sendor et al., 1999). Unfortunately, few clinicians receive education or training in prognostication, and consequently they are notoriously poor at estimating prognosis (Christakis, 1995). A majority of hospice patients are enrolled for only days to a few weeks (NHPCO, 2012). This brief time period may be insufficient to achieve the objectives of hospice. Amelioration of the physical, social, psychological, and spiritual domains of suffering takes time and involves developing relationships between patients, caregivers, and the hospice team. A few days in hospice is insufficient time for hospice to accomplish these objectives. Aside from hospice, prognosis is central to many ethical issues in end-​of-​life care. In contrast to hospice, palliative care is a specialty that does not have a prognostic requirement for eligibility. While palliative care is not tied to an explicit prognosis, prognosis is central to the practice of palliative care. The decision to initiate, withhold, or terminate life-​sustaining care is based primarily on prognosis. Although futility has fallen out of favor in contemporary bioethics, physicians still routinely invoke futility arguments for patients with a poor prognosis (Lamont and Christakis, 2001). And in the treatment of patients with serious life-​limiting illness, prognosis is essential for predicating the likely benefits and harms of treatment and the alternatives to treatment. The three components to the clinical act of prognostication are formulating the prognosis, communicating the prognosis, and using the prognosis to aid decision making. To date, prognosis research has focused on good formulation and communication; there have been few studies on the use of prognosis (Christakis, 1995). In this chapter we focus on ethical issues in prognosis from a clinical perspective. We begin with a brief historical overview, describing the waning importance of prognosis in medicine over time. This leads to a description of the importance of prognosis in palliative, hospice, and end-​of-​life care. We address ethical issues raised by the uncertainty inherent in prognosis, as well as the importance of estimating prognosis as accurately as possible despite this uncertainty. The chapter concludes with a discussion of how to communicate prognosis with patients.

Historical Perspective The three central tasks of medicine have been described as diagnosis or identification of disease, treatment or therapy to alter the clinical course of the disease, and prognosis or the expected outcomes of the disease. In fact, Hippocrates (2009) is on record as saying that it is a duty of physicians to prognosticate and that patients will be more trusting of the physician who is willing to do so. Should modern clinicians prognosticate? Prognosis in end-​of-​life care has seen a minor rise in prominence recently. Several factors drive this trend. First, modern

172    Alexander K. Smith and Paul Glare technological advances have increased our ability to extend patients’ lives by minutes, hours, and days. This has led to a discussion at the level of society and the bedside about the goals, costs, and alternatives to such treatments. For example, a clinician might say to the family of a patient on a ventilator, “What would happen if we disconnected the patient from the ventilator now? She would die immediately. And if we did not disconnect her, she may live for another day, but in what condition?” These discussions are heavily predicated on prognosis. Second, clinicians want to describe the prognosis as a method of controlling the future; anticipating death is a way of exerting control over death. Physicians may be concerned that their competence is called into question if a patient dies unexpectedly (Christakis, 2001). Third, modern advances in epidemiology and biostatistics have led to new models that are more accurate than previous models. These advances are driven by huge population-​based studies with detailed questions about risk factors, advances in statistical modeling techniques, and robust computer systems that can process complex algorithms on vast quantities of data instantaneously (Yourman et al., 2012).

Why It Is Important to Prognosticate in Palliative Care There are at least four major reasons why prognosis is important in palliative care. First, prognosis is necessary for compliance with the law, for example, in determining eligibility for services such as hospice. In the United States, Medicare—​the federal insurance program for seniors—​includes a hospice benefit that requires two physicians to certify that the patient has a prognosis of six months or less, should the disease take its usual course (Medicare. gov, 2014). In New York, the Palliative Care Information and Counseling Act amends the Public Health Law by adding section 2997-​c, which requires the “attending health care practitioner” (physicians and nurse practitioners) to offer to provide terminally ill patients information and counseling concerning palliative care and end-​of-​life options appropriate to the patient, including the prognosis, the range of options open to them, including palliative care or hospice, the risks and benefits of the various options, and the patient’s “legal rights to comprehensive pain and symptom management at the end of life”(Assembly, 2011).1 In 2008 California passed a similar law, Bill AB 2747, stating that if patients with less than a year to live ask about treatment options, health-​care providers must disclose information about all end-​of-​life options, including palliative care and hospice (Assembly, 2008). Outside the United States, there may not be legal mandate to prognosticate but there are examples of local authorities or clinicians using prognosis as a basis for access to palliative care services. For example, in Britain and Australia where inpatient hospice facilities are more widely used than in the United States, expected short-​term survival, such as that of less than one week, may be an admission criterion (Chapman and Bass, 2000).

Ethical Issues in Prognosis and Prognostication    173 Second, prognosis is critically important for many and perhaps most clinical decisions in palliative, hospice, and end-​of-​life care. The risks and benefits of many tests and treatments hinges critically on prognosis (see Table 10.1). These decisions range from those that are clearly end of life, such as choice of antidepressants, to decisions that are more upstream, such as the use of statin medications for treatment of hypercholesterolemia. Third, beyond obligations to estimate prognosis accurately for guidelines and clinical decisions, clinicians are often asked by patients and family members to estimate prognosis. In a study of what Americans define as a good death, knowing what to expect was rated as very important by 96% of respondents. Unfortunately only 15% achieved that objective (Steinhauser et al., 2000). The question “how long do I have, doctor?” is often the first question following the diagnosis of a terminal condition. While this question is sometimes asked explicitly, often it is not, and clinicians should watch for clues that the patient may be concerned about prognosis and offer to discuss prognosis with all patients with life-​limiting illness or a short life expectancy. Patients and caregivers do not just care about prognosis for life expectancy, however. They also care about prognosis for other factors that they view as critical to their quality of life. These factors may include, but are not limited to, the risk of: becoming disabled, relying on others to bathe, having difficulty walking without assistance, loss of

Table 10.1 Example Clinical Decisions that Hinge on Prognosis in End-​of-​Life Care Clinical Decision

Prognosis

Rationale

Choice of antidepressant

3 months, surgery has superior outcomes compared to radiation therapy

Discontinue statins for hypercholesterolemia

2 months) were highly significantly different at 7, 32, and 100 days, respectively. The PiPS model did perform as well as, or better than, doctors’ and nurses’ subjective prognostic judgments at the individual patient level, but the absolute agreement between actual survival and PiPS predictions was only 57%.

Ethical Issues in Prognosis and Prognostication    177

Imprecision/​Uncertainty of Prognoses The reason that the prognostic models perform well at the group level but not as well for individual patients is because of the large interindividual variation within groups. While prognosis is inherently uncertain, misclassification and uncertainty occurs throughout clinical medicine. Uncertainty in empirically derived estimates of prognosis are due to several factors. Empirically derived probabilistic prognostic models present the results as the risk of an event, in this case death, over a given time frame, such as five years. A typical result would be “a 68% risk of death over one year.” It is not clear, however, if the patient at hand would be one of the 68 out of 100 who die or one of the 32 out of 100 that would survive. Similar uncertainty is inherent in temporal prediction, such as the PiPS. Approximately 5% of patients allocated to the best “more than two months” group survived less than two weeks, while approximately 10% in the worst “less than two weeks” group were still alive after two months. Uncertainty is further derived from the sample size of the empirical study used to create the prognostic algorithm and strength of association between risk factors and the prognostic outcome. This uncertainty is commonly expressed as the 95% confidence interval. And because empirical prognostic models are often based on populations of patients, it may not be clear how the prognostic estimate for a population of patients with similar risk applies to the individual patient. Individual patients often have risk factors, both protective and harmful, that are not captured in the prognostic index. These sources of uncertainty and bias should not lead clinicians to throw up their hands and consider estimating prognosis a useless exercise. Several studies suggest that patients and caregivers make more informed decisions when they are given prognostic information. In a study of 917 seriously ill hospitalized patients with advanced lung or colon cancer, patients who thought they had a prognosis of greater than six months were more likely to favor and receive life-​extending treatments than patients who thought they had less than six months to live. Actual survival did not differ between groups (Weeks et al., 1998). In a separate study of 332 patients with advanced cancer, those who reported having discussing with their physicians preferences for end-​of-​life care were less likely to receive aggressive medical care (e.g., ventilation, resuscitation) in the final weeks of life (Wright et al., 2008). To assist clinicians to be as accurate as possible when estimating prognosis and associated uncertainty, we compiled a list of resources clinicians may use to assist them in formulating an estimate of the prognosis, both for patients with specific conditions such as cancer, as well as general nondisease specific prognostic indices (see Table 10.3). It bears emphasizing that clinicians should use these prognostic tools with caution. Prognostic indices do not account for individual variation in treatment or the response to treatment. Idiosyncrasies of a particular patient’s circumstance may lead the clinician to believe a prognosis that differs substantially from what a prognostic tool would suggest. This is good clinical medicine, and the combination of clinical intuition with a prognostic calculator may be more accurate than either alone.

178    Alexander K. Smith and Paul Glare Table 10.3 Models for Estimating Prognosis across Various Populations Resource

Population

Notes

ePrognosis

Elderly Nondisease specific

Access at www.eprognosis. org. Contains over 15 validated indexes. Builds on the results of a systematic review

Advanced Dementia Prognosis Tool

Nursing home residents with advanced dementia

Also available at www. eprognosis.org.

Seattle Heart Failure Model

Heart failure

Available at http://​depts. washington.edu/​shfm/​

Body-​Mass Index, Airflow Obstruction, Dyspnea, and Exercise Capacity Index

Chronic obstructive pulmonary disease

Can be used to estimate 4-​ year survival

The Palliative Prognostic Index)

Cancer patients enrolled in palliative care programs

Useful for predicting short term mortality (i.e., less than 6 weeks) when combined with clinician judgment

Foretelling: Should a Palliative Care Clinician Communicate the Prognosis, and How Should They Communicate It? Should the Prognosis Be Communicated? Surveys consistently find that most, but not all, cancer patients want prognostic information, whether it is good news or bad (Butow et al., 2002; Kutner et al., 1999; Hagerty et al., 2005; Hagerty et al., 2004; Parker et al., 2007; Innes and Payne, 2009). But talking to patients about prognosis is difficult, and clinicians are poor at this type of communication. It has been shown that there are often large discrepancies between patients’ and health-​care professionals’ perceptions about how much information was needed, how much information had been given, and what such information meant (Hancock et al., 2007). Clinicians tend to underestimate patients’ prognostic information needs and overestimate how much they had understood about their illness and its likely outcome (Beadle et al., 2004). The communication of prognosis near the end of life is fraught with ethical issues. These include the insufficiency of the informed-​consent framework for prognosis communication, issues that arise when patients do not want to discuss prognosis, racial and

Ethical Issues in Prognosis and Prognostication    179 ethnic differences in preferences for prognostic information, and issues around truth-​ telling. We address each of these issues in turn. In clinical practice, and on ethics consults, physicians often refer to informed consent as a justification for informing the patient of a poor prognosis. The doctrine of informed consent is a blunt instrument that is a poor framework for communication of prognosis (Manson and O’Neill, 2007). Informed consent was developed for research ethics, not for clinical care, and is largely framed as a one-​way transmittal of information: the physician informs the patients of the expected outcomes of treatment and the patient agrees or disagrees with the treatment. Communication about prognosis, however, should be a much more nuanced conversation, characterized by a two-​way interaction between a patient and a physician over time (Manson and O’Neill, 2007). The principles of informed and shared decision making serve as better models for prognosis conversations than the doctrine of informed consent. These conversations, like breaking bad news conversations, require training, skill, and practice to be done well. Prognosis is a sensitive topic, and dumping prognostic information on patients as the doctrine of informed consent mandates risks doing more harm to the patient than good. In communicating prognosis, physicians should first assess if patients are interested in discussing prognosis. They also should understand that patients who want prognostic information often both crave and dread it. They are caught between wanting to know what is going on and fearing the answers they might receive. Therefore, they want the prognosis to be given by someone whom they perceive to be an expert, and they find inconsistent information or evasiveness on the part of the professional to be distressing and unhelpful. Patients also want hopeful messages, even when they accept the terminal phase of the illness (Kirk et al., 2004; Kutner et al., 1999). Strategies clinicians may use to facilitate hope when discussing prognosis include • retaining professional honesty. • avoiding being blunt or giving more detailed information than desired by the patient. • pacing of information. • respecting patients’ need to follow alternative paths/​treatments. • exploring and facilitating realistic goals and wishes where appropriate. When employing these strategies, contemporary physicians should avoid the old “norms” of being vague, optimistic, and not extreme. A previous study found that not only did physicians overestimate survival but that they would add an optimistic premium to their prediction (Lamont and Christakis, 2001). Consequently, the patient was “twice removed” from the future really held for them. Many studies have stressed the importance of individualizing the content of prognostic discussions, but few patient characteristics have been identified to predict how much information patients want or how such information should be delivered (Kutner et al., 1999). Patients’ information needs and preferences can change during the course of their illness. While many want to discuss prognosis when they are first

180    Alexander K. Smith and Paul Glare diagnosed with metastatic disease, others want to negotiate with the clinician about when such issues are discussed. In one study, more than half the patients wanted the physicians to initiate discussions about prognosis, less than a quarter wanted the physician to tell them about survival only “if asked,” and approximately 10% of patients never wanted to discuss likely duration of survival (Hagerty et al., 2004). In general women want more information than men, and older patients request less information than younger patients (Fallowfield et al., 2002). Cultural differences may also be important (Parker et al., 2007). Likewise, patients tend to want less information as their underlying disease progresses and they approach the terminal phase of their illness. While many patients are interested in discussing prognosis, many others are not, and these preferences should be respected. In a study of sixty-​five elderly, disabled African American, white, Latino, and Chinese patients in San Francisco, two-​thirds of them preferred that a physician discuss a prognosis of less than five years with them, and one-​third did not (Ahalt et al., 2012). In-​depth interviews revealed the reasons for these preferences. Patients who wanted to discuss prognosis expressed that it would be important information to help them plan for the future. Those who did not want to discuss prognosis expressed concerns about increased anxiety, depression, or feeling that the act of a physician stating the prognosis would make it more likely to happen. The process of exploring the rationale for or against discussing prognosis is not only highly informative about the patient’s goals and values but may build a stronger relationship between the physician and patient (Back and Arnold, 2006a, 2006b). Interestingly, while the sample sizes in each ethnic group were small, there were no strong trends by ethnic group (Ahalt et al., 2012). Contrary to a popular stereotype, most Chinese elders wanted to know their prognosis and wanted the discussion to take place with them, rather than defer it to a family member such as an adult son. There were, however, some suggestions along ethnic lines about differences in preferences for how prognosis should be communicated. Latino patients often expressed a preference for a Latino physician to discuss prognosis in Spanish. Chinese patients expressed a great deference to physicians in the decision to discuss or not. White and African American patients were more likely to express a sentiment of entitlement to prognostic information. Rather than making assumptions about patients based on appearances, physicians should ask patients about their preferences and the reasons for their preferences for prognostic information. Patients who do not want to know their prognosis present an ethical dilemma for clinicians. When faced with a patient who is nearing the end of life, physicians feel it is important to talk with patients about their prognosis so that they can make informed decisions. For example, a patient who understands that the prognosis is poor might be less likely to opt for care that has a very small chance of prolonging life and a high chance of impairing quality of life (Weeks et al., 1998; Wright et al., 2008). On the one hand, physicians should never force patients to discuss prognosis. On the other hand, physicians can provide some important clues to patients that the prognosis is important to decision making. For example, the physician could say, “I understand that you

Ethical Issues in Prognosis and Prognostication    181 don’t want to discuss how long you have to live. I want to respect that. At the same time, I worry that you might make different decisions if you were fully informed of the situation. Could you reconsider talking to me about it? Or perhaps letting me talk to someone else in your family about it, and letting them help with these decisions?” A limitation of this approach is that it suggests that the prognosis is poor, thus revealing the prognosis inadvertently to patients who are not prepared or ambivalent.

How Should the Prognosis Be Communicated? Although most patients want prognostic information, the best way to communicate such information is not always clear. Guidelines stress the importance of communication occurring within the context of a caring, trusting relationship, consistency of information within the multiprofessional team, and the need to communicate prognostic information to other members of the family (Clayton et al., 2007, Back et al., 2009; Kiely et al., 2010). As highlighted previously, while most patients want to know “what to expect,” not all patients want to be provided with an estimation of their life expectancy. Hence it is very important to first clarify the person’s understanding of his or her medical situation and the information desired. Any information provided about prognosis should then be tailored to the individual needs of patients and their families. If patients want to be informed of their likely survival duration, it should ideally be delivered in a straightforward and clear manner. Given the uncertainty of the prognosis, as discussed in the previous section, this is not always easy to do. For patients who would like to be provided with a numerical estimation of their life expectancy, clinicians should be guided by studies that show that patients are strongly susceptible to the framing of risk information. For example, patients are more likely to accept a treatment if the outcomes are framed in terms of survival rather than in terms of death (McNeil et al., 1982; Verma et al., 2014). To minimize bias, physicians can present both the risks of survival and risks of death. It may be best, however, to give patients nearing the end of life a sense of their overall life expectancy (e.g., the “gist”) rather than the exact mortality risk over a given time. For example, rather than saying, “If there were 100 patients like you, 90 would die within six months and 10 would survive,” it may be better to say, “You probably have a few months left—​most likely months that can be counted on one hand.” It is noteworthy that the survival curve of patients with a variety of advanced cancers typically approximates an exponential function. Consequently, the median survival of the patient’s prognostic subgroup forms the starting point for framing this risk information (Kiely et al., 2010; Stockler et al., 2006). Whether the clinician has used a prognostic tool such as the PiPS model or relied on his or her clinical judgment to obtain this figure for the median survival of typical patients like this individual, the same approach can be used. To illustrate this process, let us assume the survival estimate is three months.

182    Alexander K. Smith and Paul Glare •​ Explain that a median survival of three months means that 50% will live longer than three months. •​ Use a simple multiple of the median to estimate and explain the typical, best-​case, and worst-​case scenarios: o Typical—​about half of similar patients would live for somewhere between six weeks and six months (half to double the predicted median). o Best case—​about 10% of patients could expect to live beyond one year (three to four times the predicted median). o Worst case—​about 10% of patients will experience more rapid decline and will die within one week (1/​6 of the predicted mean). Using multiples of the median to estimate and provide typical, best-​, and worst-​case scenarios as outlined here may offer a way of conveying more realism and hope than a single point estimate of the median survival. Finally, after providing patients with information about life expectancy, it is important to explore and acknowledge the patient’s and family’s emotional reaction to the news and to check their understanding about what has been discussed. As discussed, physicians have an obligation to disclose the uncertainty of prognostic information, gauge patient and caregiver responses to this information, and refocus on the present rather than an uncertain future (Smith et al., 2013). However, they should not be too caught up in the technical details of making sure the patient understands exactly how uncertain the information is and all potential various sources of uncertainty. Again, this risks information dumping and may detract from the overall message. Studies have shown that communicating uncertainty with precision may come at the expense of understanding of the “gist” of risk information (Fagerlin et al., 2011). Rather the focus of communication about uncertainty should attend to the affective response to prognosis and to the uncertainty associated with prognosis. Patients and caregivers are often caught up in trying to control the future. They may pay excessive attention to physical signs or laboratory values. They may feel trapped in an uncertain future that is, ultimately, unknowable. We advocate for a strategy of addressing the affective response to prognosis and associated uncertainty outlined in Table 10.4. Finally, physicians should strive to tell the truth about prognosis. In a study of physicians who cared for patients with cancer, Christakis and Iwashyna (2008) found that when physicians communicated prognosis to other physicians, they were far more pessimistic than they were with patients. Reports indicate that some physicians report that patients have less time to live than they actually believe, so that patients may qualify for services such as hospice (Christakis and Iwashyna 1998). A recent study suggested that patients with incurable cancer who believe their chemotherapy is curative are more satisfied with their physician (Weeks et al., 2012). Physicians should also avoid colluding with the patient not to discuss prognosis. An analysis of audiotaped oncology consultations found that the oncologist tended to focus

Ethical Issues in Prognosis and Prognostication    183 Table 10.4 Communication of Prognosis and Associated Uncertainty Communication Strategy Rationale

Words to say

Offer to discuss prognosis, don’t force

Many patients want to discuss prognosis and are waiting for the physician to raise the topic. Some do not.

“One of the things that may help us make decisions is talking about how much time you may have left. Is that something that we can discuss?”

Explore reasons for wanting to discuss prognosis, or not

Informative and may help build the relationship by showing caring and interest.

“Sometimes the reasons patients want to discuss how long they have to live are very different from the reasons doctors want to have these conversations. So I can help you better in the future, can you tell me why you [do/​do not] want to discuss this?”

Give the prognosis using Precise estimates give a false sense a range, normalize the of precision, are likely to be wrong, uncertainty and may threaten the doctor–​ patient relationship. A range acknowledges uncertainty.

“Like the weather, prognosis is not certain. I wish it could be more certain. Knowing everything I know about you, I think you may have weeks to a few months at the most left.”

Respond to the emotional reaction to the prognosis and uncertainty

Like hearing a terminal diagnosis, patients and caregivers first reactions are emotional. They are unlikely to process any other information right away. They may feel anxious and trapped by the uncertainty.

“I can see that you’re really worried about how much time you will have and if you will be able to do all the things you hoped. Can you tell me more about what that is like for you?”

Refocus on the present

Bringing patients and caregivers “Let’s refocus on what we can do back to the present rather than an now, today, to help you live each day uncertain future can be therapeutic as good as it can be.” in the short term, and may prevent regret for not being more “in the moment” after death.

on treatment options and the results of investigations rather than on questions of prognosis, often invoking “false optimism” about the prospects of recovery (The et al., 2000). This optimism may be fostered both by doctors’ reluctance to give clear information about prognosis and patients’ avoidance of asking direct questions. Telling the truth may have costs, and when physician reimbursement is tied to satisfaction, this cost may hit physicians in their pocketbooks. Not telling the truth is problematic, however, for several reasons. It undermines patient faith in the medical profession in general and the physician in particular. As described previously, studies suggest that patients make different decisions when they

184    Alexander K. Smith and Paul Glare understand their prognosis. While telling the truth may lead to lower satisfaction, it may also help the patient and family feel more prepared—​and sense of preparedness may be a better metric in end-​of-​life care than satisfaction.

Prognosis and Decision Making: How Should Prognostic Information Be Used? This topic has received little consideration in the palliative care literature to date, either in the form of empirical research or ethical discourse. Given the inaccuracy and imprecision of survival predictions at the present time and the challenges that arise when communicating them discussed in the previous sections, it seems difficult to recommend incorporating the prognosis into the clinical decision making process. The studies of prognosis and decision that have been undertaken have mainly evaluated the adverse consequences of inaccurate prognoses. For example, one element of the SUPPORT study evaluated the impact of patients’ evaluations of their prognosis and their preferences for aggressive care at the end of life. Patients who overestimate their survival are more likely to opt for aggressive, life-​sustaining treatment but without better outcomes (Weeks et al., 1998). Conversely, overly pessimistic prognoses may lead to premature withdrawal of life-​sustaining treatments and unnecessary deaths (Holloway et al., 2005). Recent research has begun to investigate how patients and families may use prognostic information to assist them with the complex decision making they face as death approaches. Several studies have shown that early end-​of-​life discussions are prospectively associated with less aggressive care and greater use of hospice at end of life. The landmark randomized trial of early palliative care in lung cancer showed that frequent meetings with palliative care staff to discuss prognosis and goals of care led to better a variety of better outcomes, including lower utilization of aggressive care in the final six weeks of life, and paradoxically a longer survival (Temel et al., 2010). The Cancer Care Outcomes Research and Surveillance Consortium, a large population-​and health system–​based prospective cohort study of the aggressiveness of end of life found that patients with stage IV lung or colorectal cancer who died found that patients who had end-​of-​life discussions with their physicians before the last thirty days of life were less likely to receive aggressive measures at end of life, including chemotherapy (p = .003), acute care (p < .001), or any aggressive care (p < .001). Such patients were also more likely to receive hospice care (p < .001) and to have hospice initiated earlier (p < .001; Mack et al., 2012). Cancer patients who viewed a goals-​of-​care video were more likely to prefer comfort care and avoid CPR. They were more certain of their end-​of-​life decision making than participants who only heard a verbal description (El-​Jawahri et al., 2010).

Ethical Issues in Prognosis and Prognostication    185 More research is needed on prognosis and decision making in relation to some of the issues raised here. If physicians are overly pessimistic when they are communicating with other physicians, does this lead to withholding treatment that may have been beneficial? Does formulating and openly discussing the prognosis create a self-​fulfilling prophecy that also leads to undertreatment when the outlook seems poor? Would prognoses that are more accurate, precise, and optimally communicated—​the current research agenda in palliative care—​actually change decision making and outcomes? While the clinical trial in lung cancer confirmed this at one academic center, the interventional arm of the SUPPORT study did not (“A Controlled Trial to Improve Care,” 1995; Temel et al. 2010). Medicine may be able to learn from other disciplines such as meteorology and business in this regard. Studies of decision making by authorities and the public in the face of storm surges or flooding risk have shown that the interaction between forecasts and decision making are complex. For residents, National Weather Service statements and evacuation orders are important considerations, but many of them also evaluate their own risk and make their own protective decisions. How these statements and orders are communicated is important, with warnings of “certain death” often producing negative attitudes in residents (Morss and Hayden, 2010). For authorities, the meteorological forecast is only one component of the decision on managing an extreme weather event. While they appreciate a quantification or analysis of the uncertainty, this would also have limited impact on their decisions, or even make them more complicated. Decision making on a binary outcome made under a tight deadline, such as whether or not to evacuate, is complex, nonlinear, and involves many considerations apart from science. The parallels of this situation with decision making in deteriorating patients at the end-​ of-​life are painfully clear. Paradoxically, when the stakes are high and the uncertainties large, the domination of “hard facts” over “soft values” can become inverted (Morss et al., 2005). Business theory can also shed light on how patients make their decisions, feel right about them, and commit to them. For example, Regulatory Focus theory predicts some patients respond better to prognostic information framed for avoidance of further losses, while others respond to framing that emphasizes positive aspects such as hopes, gains, or accomplishments (Myers, 2010). Similarly, Regulatory Mode theory indicates some patients are “assessors,” who would use prognosis as just one piece of information in their search for the optimal decision, whereas others are “locomotors,” who want a survival prediction they can act on (Avnet and Higgins, 2003).

Conclusion Prognostication remains a challenging topic in palliative care. Clinicians’ focus on this topic has been narrow, primarily addressing time until death, while patients’ interest is on what to expect, a much broader topic. Clinicians and researchers have been interested

186    Alexander K. Smith and Paul Glare in the difficult tasks of improving the accuracy of prognoses, reducing their uncertainty, and delivering this information to patients when it is both craved and dreaded. Much less attention has been given to how prognostic information influences clinical decision making, although there is some data that shows when it is done well it is beneficial and when it is done badly it causes harm. We believe that prognosis is important and that clinicians have a responsibility to develop, maintain, and improve their prognostic skills. Clinicians should also utilize prognostic information in their clinical decision making. Even if precise and accurate predictions of survival duration become available, this alone should never drive treatment plans. The prognosis will always be just one element of a complex calculus in which other considerations often trump hard science.

Note 1. The Palliative Care Information and Counseling Act has an accompanying piece of legislation, the Palliative Care Access Act, which is aimed at administrators of hospitals and nursing homes. It requires them to provide access to palliative care services for their patients/​residents so that the Palliative Care Information Act can be complied with. It is noteworthy, however, that (at the time of writing), the New York State Department of Health is not enforcing compliance is not being enforced with either act, but rather it is hoping that providers and administrators will be inspired to embrace the spirit of these laws.

References A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). 1995. JAMA 274 (20):1591–​1598. Ahalt, Cyrus, Louise C. Walter, Lindsey Yourman, Catherine Eng, Eliseo J. Pérez-​Stable, and Alexander K. Smith. 2012. “Knowing is better”: Preferences of diverse older adults for discussing prognosis. The Journal of General Internal Medicine 27(5):568–​575. Assembly, California State. 2008. California Bill 2747:  End of Life Care. http://​www.leginfo. ca.gov/​pub/​07-​08/​bill/​asm/​ab_​2701-​2750/​ab_​2747_​bill_​20080813_​amended_​sen_​v93.html. Assembly, New York State. 2011. Palliative Care Information Act, Section 2997-​c, February. http://​www.health.ny.gov/​professionals/​patients/​patient_​rights/​palliative_​care/​information_​act.htm. Avnet, Tamar, and E. Tory Higgins. 2003. Locomotion, assessment, and regulatory fit: Value transfer from “how” to “what.” The Journal of Experimental Social Psychology 39(5):525–​530. Back, A., R. M. Arnold, and J. Tulsky. 2009. Mastering Communication with Seriously Ill Patients—​ Balancing Honesty with Empathy and Hope. New  York:  Cambridge University Press. Back, A. L., and R. M. Arnold. 2006a. Discussing prognosis: “how much do you want to know?” talking to patients who are prepared for explicit information. Journal of Clinical Oncology 24(25):4209–​4213. doi: 10.1200/​JCO.2006.06.007.

Ethical Issues in Prognosis and Prognostication    187 Back, A. L., and R. M. Arnold. 2006b. Discussing prognosis:  “how much do you want to know?” talking to patients who do not want information or who are ambivalent. Journal of Clinical Oncology 24(25):4214–​4217. doi: 10.1200/​JCO.2006.06.008. Beadle, G. F., P. M. Yates, J. M. Najman, A. Clavarino, D. Thomson, G. Williams, L. Kenny, S. Roberts, B. Mason, and D. Schlect. 2004. Beliefs and practices of patients with advanced cancer: implications for communication. British Journal of Cancer 91(2):254–​257. Butow, P. N., S. Dowsett, R. Hagerty, and M. H. N. Tattersall. 2002. Communicating prognosis to patients with metastatic disease: what do they really want to know? Support Care Cancer. 10(2):161–​168. Byock, I. R. 1996. The nature of suffering and the nature of opportunity at the end of life. Clinics in Geriatric Medicine 12(2):237–​252. Chapman, K. Y., and L. Bass. 2000. A comparison of hospice in the UK and the US. The American Journal of Hospice & Palliative Medicine 17(3):173–​177. Christakis, N. A. 1995. Prognostication and Death in Medical Thought and Practice. Philadelphia: University of Pennsylvania. Christakis, N. A. 2001. Death Foretold:  Prophecy and Prognosis in Medical Care. Chicago: University of Chicago Press. Christakis, N. A., and T. J. Iwashyna. 1998. Attitude and self-​reported practice regarding prognostication in a national sample of internists. Archives of Internal Medicine 158(21):2389–​2395. Christakis, N. A., E. B. Lamont, J. L. Smith, and C. M. Parkes. 2000. Extent and determinants of error in doctors’ prognoses in terminally ill patients: prospective cohort study. British Medical Journal 320(7233):469–​473. Clayton, J. M., K. M. Hancock, P. N. Butow, M. H. Tattersall, D. C. Currow, J. Adler, S. Aranda, K. Auret, F. Boyle, A. Britton, R. Chye, K. Clark, P. Davidson, J. M. Davis, A. Girgis, S. Graham, J. Hardy, K. Introna, J. Kearsley, I. Kerridge, L. Kristjanson, P. Martin, A. McBride, A. Meller, G. Mitchell, A. Moore, B. Noble, I. Olver, S. Parker, M. Peters, P. Saul, C. Stewart, L. Swinburne, B. Tobin, K. Tuckwell, and P. Yate et al. 2007. Clinical practice guidelines for communicating prognosis and end-​of-​life issues with adults in the advanced stages of a life-​limiting illness, and their caregivers. The Medical Journal of Australia 186(12 Suppl.):S77, S79, S83–​108. doi: cla11246_​fm [pii]. Dawes. 1989. Clinical versus actuarial judgment. Science 243(4889):1668–​1674. doi:  10.1126/​ science.2648573. El-​Jawahri, A., L. M. Podgurski, A. F. Eichler, S. R. Plotkin, J. S. Temel, S. L. Mitchell, Y. Chang, M. J. Barry, and A. E. Volandes. 2010. Use of video to facilitate end-​of-​life discussions with patients with cancer: a randomized controlled trial. Journal of Clinical Oncology 28(2):305–​ 310. doi: 10.1200/​JCO.2009.24.7502. Fagerlin, Angela, Brian J. Zikmund-​ Fisher, and Peter A. Ubel. 2011. Helping patients decide: ten steps to better risk communication. The Journal of the National Cancer Institute 103(19):1436–​1443. doi: 10.1093/​jnci/​djr318. Fallowfield, L. J., V. A. Jenkins, and H. A. Beveridge. 2002. Truth may hurt but deceit hurts more: communication in palliative care. Palliative Medicine 16(4):297–​303. Glare, P., K. Virik, M. Jones, M. Hudson, S. Eychmuller, J. Simes, and N. A. Christakis. 2003. A systematic review of physicians’ survival predictions in terminally ill cancer patients. BMJ. 327(7408):195. doi: 10.1136/​bmj.327.7408.195. Gwilliam, Bridget, Vaughan Keeley, Chris Todd, Matthew Gittins, Chris Roberts, Laura Kelly, Stephen Barclay, and Patrick C. Stone. 2011. Development of prognosis in palliative care study (PiPS) predictor models to improve prognostication in advanced cancer: prospective cohort study. British Medical Journal 343:d4920.

188    Alexander K. Smith and Paul Glare Hagerty, R. G., P. N. Butow, P. M. Ellis, S. Dimitry, and M. H.  N. Tattersall. 2005. Communicating prognosis in cancer care: a systematic review of the literature. Annals of Oncology 16(7):1005–​1053. Hagerty, Rebecca G., Phyllis N. Butow, Peter A. Ellis, Elizabeth A. Lobb, Susan Pendlebury, Natasha Leighl, David Goldstein, Sing Kai Lo, and Martin H. N. Tattersall. 2004. Cancer patient preferences for communication of prognosis in the metastatic setting. Journal of Clinical Oncology 22(9):1721–​1730. Halpern, J., and R. M. Arnold. 2008. Affective forecasting:  an unrecognized challenge in making serious health decisions. Journal of General Internal Medicine 23(10):1708–​1712. doi: 10.1007/​s11606-​008-​0719-​5. Hancock, Karen, Josephine M. Clayton, Sharon M. Parker, Sharon Walder, Phyllis N. Butow, Sue Carrick, David Currow, Davina Ghersi, Paul Glare, Rebecca Hagerty, and Martin H. N. Tattersall. 2007. Discrepant perceptions about end-​of-​life communication:  a systematic review. The Journal of Pain and Symptom Management 34(2):190–​200. Hippocrates. 2009. The Book of Prognostics: London: Dodo Press. Holloway, R. G., C. G. Benesch, W. S. Burgin, and J. B. Zentner. 2005. Prognosis and decision making in severe stroke. JAMA 294(6):725–​733. doi: 10.1001/​jama.294.6.725. Inc. 2003–​2011. Adjuvant! Adjuvant! Online. https://​www.adjuvantonline.com/​index.jsp. Innes, S., and S. Payne. 2009. Advanced cancer patients’ prognostic information preferences: a review. Palliative Medicine 23(1):29–​39. doi: 10.1177/​0269216308098799. Kiely, B. E., M. H. Tattersall, and M. R. Stockler. 2010. Certain death in uncertain time: informing hope by quantifying a best case scenario. Journal of Clinical Oncology 28(16):2802–​2804. doi: 10.1200/​JCO.2009.27.3326. Kirk, P., I. Kirk, and L. J. Kristjanson. 2004. What do patients receiving palliative care for cancer and their families want to be told? A Canadian and Australian qualitative study. British Medical Journal 328(7452):1343. Kutner, J. S., J. F. Steiner, K. K. Corbett, D. W. Jahnigen, and P. L. Barton. 1999. Information needs in terminal illness. Social Science and Medicine 48(10):1341–​1352. Lamont, Elizabeth B., and Nicholas A. Christakis. 2001. Prognostic disclosure to patients with cancer near the end of life. Annals of Internal Medicine 134(12):1096–​1105. doi: 10.7326/​ 0003-​4819-​134-​12-​200106190-​00009. Mackillip, W. J., and C. F. Quirt. 1997. Measuring the accuracy of prognostics judgments in oncology. Journal of Clinical Epidemiology 50(1):21–​29. Manson, N. C., and O. O’Neill. 2007. Rethinking Informed Consent in Bioethics. New York: Cambridge University Press. McNeil, Barbara J., Stephen G. Pauker, Harold C. Sox, and Amos Tversky. 1982. On the elicitation of preferences for alternative therapies. The New England Journal of Medicine 306(21):1259–​1262. doi: doi:10.1056/​NEJM198205273062103. Medicare.gov. 2014. Hospice and Respite Care. Washington, DC:  Centers for Medicare & Medicaid Services. http://​www.medicare.gov/​coverage/​hospice-​and-​respite-​care.html. Morss, R. E., and M. H. Hayden. 2010. Storm surge and “certain death”: interviews with Texas coastal residents following Hurricane Ike. Weather, Climate, and Society 2(3):174–​189. Morss, Rebecca E, Olga V Wilhelmi, Mary W Downton, and Eve Gruntfest. 2005. Flood risk, uncertainty, and scientific information for decision making: lessons from an interdisciplinary project. The Bulletin of the American Meteorological Society 86(11):1593–​1601. Myers, Rachel E. 2010. Promoting healthy behaviors: How do we get the message across? The International Journal of Nursing Studies 47(4):500–​512.

Ethical Issues in Prognosis and Prognostication    189 NHPCO. 2012. NHPCO Facts and Figures:  Hospice Care in America. National Hospice and Palliative Care Organization. http://​www.nhpco.org/​sites/​default/​files/​public/​Statistics_​ Research/​2012_​Facts_​Figures.pdf. Parker, Sharon M., Josephine M. Clayton, Karen Hancock, Sharon Walder, Phyllis N. Butow, Sue Carrick, David Currow, Davina Ghersi, Paul Glare, Rebecca Hagerty, and Martin H. N. Tattersall. 2007. A systematic review of prognostic/​end-​of-​life communication with adults in the advanced stages of a life-​limiting illness: patient/​caregiver preferences for the content, style, and timing of information. The Journal of Pain and Symptom Management 34(1):81–​93. Sendor, V. F., P. M. O’ Connor, and W. A. Ury. 1999. Hospice and palliative care: questions and answers. Care Management Journal 1(4). Smith, A. K., D. B. White, and R. M. Arnold. 2013. Uncertainty—​the other side of prognosis. N Engl J Med. 368(26):2448–​2450. doi: 10.1056/​NEJMp1303295. Steinhauser, K. E., N. A. Christakis, E. C. Clipp, M. McNeilly, L. McIntyre, and J.A. Tulsky. 2000. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA 284(19):2476–​2482. Stephenson, Andrew J, Peter T Scardino, James A Eastham, Fernando J Bianco, Zohar A Dotan, Christopher J DiBlasio, Alwyn Reuther, Eric A Klein, and Michael W Kattan. 2005. Postoperative nomogram predicting the 10-​year probability of prostate cancer recurrence after radical prostatectomy. Journal of Clinical Oncology 23(28):7005–​7012. Stockler, M. R., M. H. Tattersall, M. J. Boyer, S. J. Clarke, P. J. Beale, and R. J. Simes. 2006. Disarming the guarded prognosis: predicting survival in newly referred patients with incurable cancer. The British Journal of Cancer 94(2):208–​212. doi: 10.1038/​sj.bjc.6602908. Temel, J. S., J. A. Greer, A. Muzikansky, E. R. Gallagher, S. Admane, V. A. Jackson, C. M. Dahlin, C. D. Blinderman, J. Jacobsen, W. F. Pirl, J. A. Billings, and T. J. Lynch. 2010. Early palliative care for patients with metastatic non-​small-​cell lung cancer. The New England Journal of Medicine 363(8):733–​742. doi: 10.1056/​NEJMoa1000678. The, A. M., T. Hak, G. Koeter, and G. van Der Wal. 2000. Collusion in doctor-​patient communication about imminent death:  an ethnographic study. British Medical Journal 321(7273):1376–​1381. Verma, A. A., F. Razak, and A. S. Detsky. 2014. Understanding choice: why physicians should learn prospect theory. JAMA 311(6):571–​572. doi: 10.1001/​jama.2013.285245. Weeks, J. C., E. F. Cook, S. J. O’Day, L. M. Peterson, N. Wenger, D. Reding, F. E. Harrell, P. Kussin, N. V. Dawson, A. F. Connors, Jr., J. Lynn, and R. S. Phillips. 1998. Relationship between cancer patients’ predictions of prognosis and their treatment preferences. JAMA 279(21):1709–​1714. Weeks, Jane C., Paul J. Catalano, Angel Cronin, Matthew D. Finkelman, Jennifer W. Mack, Nancy L. Keating, and Deborah Schrag. 2012. Patients’ expectations about effects of chemotherapy for advanced cancer. The New England Journal of Medicine 367(17):1616–​1625. doi: 10.1056/​NEJMoa1204410. Wright, A. A., B. Zhang, A. Ray, J. W. Mack, E. Trice, T. Balboni, S. L. Mitchell, V. A. Jackson, S. D. Block, P. K. Maciejewski, and H. G. Prigerson. 2008. Associations between end-​of-​life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment. JAMA 300(14):1665–​1673. doi: 10.1001/​jama.300.14.1665. Yourman, L. C., S. J. Lee, M. A. Schonberg, E. W. Widera, and A. K. Smith. 2012. Prognostic indices for older adults: a systematic review. JAMA 307(2):182–​192. doi: 10.1001/​jama.2011.1966.

Section II

T H E OR E T IC A L , C U LT U R A L ,  A N D P SYC HO S O C IA L I S SU E S

chapter 11

T he Smell of C h l ori ne Coming to Terms with Death Stuart J. Youngner

“Do you smell chlorine?” the dying woman asked me. I didn’t. When I first received the call from the oncologist to see Mrs. Smith, I had what had become a predictable reaction: I became uncomfortable. “What can I do that could help this poor woman?” I thought. “I have nothing to offer. I do not want to see her.” These thoughts persisted as I made my way to her hospital room. Mrs. Smith was alone and lying in bed. She appeared weary and worried. I took a seat next to her as she began to speak with little prompting. She told me she had metastatic breast cancer and had exhausted treatment possibilities. (This meeting took place 35 years ago.) She knew she was going to die soon and was very scared and upset. She could not sleep. She told me that her pain was under control. I asked if there was anything particular on her mind. “Yes,” she said, “It’s my husband.” She went on to tell me that she and her husband had purchased a gravesite near their home where they planned to be buried side by side. Her parents were buried together in a nearby town. She was scared to be buried alone, waiting for her husband to die. She wanted to be buried with her parents, but was unwilling to tell her husband this because she thought he would think she was betraying him. The conflict was torturing her. We talked a bit about her marriage, which seemed to be a happy one. I urged her to talk with her husband and told her that I would come back soon to see how things had gone. When I returned the next day, the patient seemed like a different person. She was smiling and more energetic. The worry seemed gone. Her first comment to me as I entered her room was the question, “Do you smell chlorine? There was a very strong smell of chlorine when you walked in.” She told me that the talk with her husband had gone well. He was very understanding of her choice and supported it fully. Her relief was palpable. “Now I can die without that worry,” she said. I asked her more about the smell of chlorine. “You know, I loved my Daddy so much,” she said. “He was so warm and caring. Every day I waited at the door for him to come home. I would rush into his arms and he would pick me up and hug me.” She went on to say that her father had worked in

194   Stuart J. Youngner a plant where he was exposed to chlorine all day and his clothes were saturated with it. “The smell in this room is just like that,” she said with a smile. Over the four to five years I worked with dying patients, I began to recognize a pattern. When the consultation request came in, I had a sense of helplessness and a wish to avoid the entire situation. When I finally saw the patients and their families, however, I almost always came away with a sense of great accomplishment and even gratitude. It may seem strange that, despite repeated experiences to the contrary, each request to talk with a dying patient triggered a sense of helplessness and a wish to avoid the situation. I later realized why. My first personal experience of talking with those who are dying was with my mother, who died of Hodgkin’s disease when I was 19. She had struggled with it all my life; now there is a cure. She was in and out of the hospital and often suffered from the effects of nitrogen mustard and other nasty drugs that were used to temporarily push back the cancer. I have a sister who is 7 years younger than I. The coping style in my family was denial and avoidance. We never, individually or as a family, discussed what was happening, what the future held, and what we were all feeling. There was little or no opportunity to help each other. We talked with my mother as she was dying, but we never talked with her or amongst ourselves about the fact that she was dying and what that meant to us emotionally. Lisa Humphrey, a pediatric palliative care expert, discusses in her chapter later in this book the importance of “tagging your baggage” when you work with dying patients (Humphrey, this volume). “Baggage” refers to the life experiences that can and do impinge on our ability to work with and be helpful to dying patients and their families. Cleary, I had personal baggage. I also had cultural and professional baggage. When I entered medical school in 1966, the United States was a death-​denying society. The word “cancer” almost never appeared in the newspaper or was mentioned in polite company. People died of “lingering” illnesses. In my second year of medical school, a year crammed with dense lectures about medical science, we had three or four clinical correlation conferences. These were much anticipated and loved events at which senior clinicians would discuss and interview actual patients. It was a chance to venture into the “real world” of medicine, a place for which most of us yearned after two years of memorizing facts. The Chairman of the Department of Medicine, Dr. A, led a conference titled, “How to Handle Patients with Terminal Illness.” The professor, who was a well-​respected scholar and hematologist-​ oncologist, told us to outright lie to patients. To demonstrate the success of his tactic, he interviewed three or four of his patients with malignancies ranging from leukemia to lymphoma. When he asked them what was wrong, they answered that they had a “problem” with their blood. All of them, like their doctor, were cheerful and seemingly carefree. The professor went on to tell us that keeping the truth from patients was the only way to protect them from the devastatingly bad news. He urged us to follow his example and go so far as to write fake diagnoses in the hospital charts for patients who, through physician family members, might gain access to the medical record. Many years later, one of the Chairman’s colleagues told me that he had lunch with one of Dr. A’s patients during the period in which I had heard Dr. A’s lecture. The patient revealed

The Smell of Chlorine    195 to him that he had cancer. “Did Dr. A tell you?” the colleague had asked in surprise. “No,” said the patient. “We never talked about it. He would have been too upset.” No one in the class of eighty students challenged Dr. A’s clinical approach to the dying patient, and we had no other formal education about the subject during my four years of medical school. My informal “education” was consistent with the cultural climate. When we did walking rounds and encountered patients who were dying, we found the doors to their rooms closed. We rarely opened them. One time the attending physician did open the door. The room was dark and the shades were pulled down. “She’s dying,” he said, and closed the door as if that were all we needed to know. I did my internship in pediatrics. At that time, the treatment of childhood malignancy was in its early stages. Dr. B, a well-​known pediatric oncologist, took care of a great many referrals to our tertiary pediatric center. He was known for not talking with his patients. He did speak with parents, but avoided almost any interaction with their dying teenagers. One of my fellow interns became so incensed that he made an appointment with Dr. B and literally yelled at him. Dr. B calmly listened and gave no response. His behavior did not change. Attention to the problems of the dying changed dramatically in 1969, when Kubler-​ Ross’s book, On Death and Dying, was published. Elizabeth Kubler-​Ross was a physician working in a Chicago hospital who was interested in speaking with dying patients. She was told there were none. Due to her persistence, she was able to talk with many dying patients, and her revelations immediately exploded in the public media. It was as if American society was hungry for a different approach. Her findings were a bit formulaic—​that is, five stages people went through when coping with a diagnosis of impending death. Kubler-​Ross blurted out truths that society had chosen to ignore, just as the little boy did in the Hans Christian Andersen story, the “Emperor’s New Clothes.” For example, she reported that dying people were generally not only willing, but actually pleased to speak about their worries and concerns. All that the physician had to do was demonstrate that he or she was willing to listen. She also pointed out that dying patients were well aware that they were dying. Their bodies told them the truth, as did the demeanor of their loved ones and their physician. Avoidance created a wall of silence that left patients (and often families) isolated and alone. Of course, I had lived through these unfortunate circumstances both as a medical student and a son. Kubler-​Ross’s work helped inspire a general interest among clinicians and researchers. It gave credibility to both previous and future efforts to illuminate the deep shadows surrounding death and dying. As a beginning resident in psychiatry in the early 1970s, I was excited by these developments. I was highly motivated to engage in this new work and apply it to the patients and families I encountered in my rotation on the consultation-​liaison (CL) service (where we saw patients in the medical and surgical wards) and in my early career as CL psychiatrist. Fortunately, I had mentors who both supported and encouraged my work. Two of the authors who spoke most helpfully to me were Avery Weisman and E. Mansell Pattison, both psychiatrists. Pattison made the helpful observation that death is not the primary problem (Pattison, 1967). No one really understands what death is. Fear of death is the problem,

196   Stuart J. Youngner and it is a common part of the process of dying. Pattison also said that the question of whether to tell a patient that he or she is dying is false. “The care of the dying does not revolve around telling or not telling but rather the whole panorama of human interactions that surround the dying person (p. 45).” Pattison wrote that fear of death can be broken down into various components. Once the physician identifies which components are most troublesome to an individual patient, a way can be found to alleviate the suffering they cause. Some of the components identified by Pattison are: Fear of the Unknown: For example, what life experiences will I not have? What is my fate in the hereafter? What will happen to my body, my family, my friends, my life experiences? Pattison notes that some of the questions “can be answered rather well immediately, some answers will be found over time, and some can not be answered (p. 49).” Fear of Loneliness: The physical and emotional isolation patients experienced during my training only exacerbated their feelings of loneliness. The patient I described earlier is an example. Her fear of death was a dread of being buried alone and of alienating her husband. I was able to hear this fear and suggest a way out of it. When she resolved the problem, her suffering diminished considerably. Fear of Loss: “The process of dying,” writes Pattison, “confronts the person with the reality of losing one’s family and friends through one’s death, just as much as if they were dying (p. 51).” Other losses loom as well: loss of body or body image, loss of job, loss of social identity (e.g., as provider, nurturer). These losses need to be mourned; if they are, great solace will follow. Fear of Pain: With the involvement of palliative care and hospice, pain can almost always be reduced to tolerable levels. This knowledge may bring great comfort. This brief summary of Pattison’s work leaves out many helpful observations and suggestions that remain extremely useful for persons who work with the dying. Avery Weisman was another writer who was clearly immersed in talking with the dying. A half-​century ago, he wrote that, “Professional training prepares the physician and psychiatrist inadequately for the confrontation with the plight of dying patients (Weisman, 1970).” “Few clinicians readily acknowledge the depth of their misgivings and apprehensions about death (p. 71),” he noted. Weisman said that the psychological management of dying patients makes it “evident that the personality of the physician is an essential ingredient of his profession (p. 74).” Confronting death “arouses some of the most pervasive fears of all men—​extinction, victimization, passivity, abandonment, disfigurement and above all loss of self-​esteem (p. 80).” He warned that, “Professional impersonality which denies vulnerability is a defense that is highly vulnerable (p. 75).” To help dying patients, A doctor must be accessible as a person not only available as a physician. He can then maintain respect for the patient, because his primary antipathy will be directed only toward the cause of suffering, not toward the person who suffers, whether that person is the patient or the doctor himself.

The Smell of Chlorine    197 Management based on unexamined misgivings and dogmatic misconceptions tends to encourage denial, deception and dissimulation. Such strategies are not only harmful, but implausible and impractical. We may be grossly shocked by the cruel realities of death, but when we shun human confrontation we cast ourselves into the position of cultivating a phobia for an event that cannot be avoided. The physician then finds himself face-​to-​face with unyielding despair. (Weisman, 1970, p. 75)

Weisman also points out that the experience of caring for the dying can be both rewarding and enlightening. One of my supervisors told me that helping dying patients work through their anxieties can be more gratifying than working with non-​dying patients. Work with dying patients may demand greater energy and attention to one’s own baggage, but it often suggests clear and decisive interventions with nearly immediate positive results. My patient’s quick psychic healing and almost magical identification with me as a loving father was an experience I will always remember and cherish. There are many such experiences to be had. Although my work with dying patients was both rewarding to me and respected by others, it became wearing. I never got over the anxiety and wish for avoidance when I received a referral. One day, I was asked to see a man who had just received news of an inoperable cancer. “He won’t talk to anyone,” the referring doctor told me. “He just lies in bed and trembles.” The patient had no known prior psychiatric history. When I went to see the patient, I found him lying on his back staring at the ceiling. A trembling shook his entire body. I asked him a few questions, and he did not respond. Rather than sitting with him for a while or finding another psychiatrist to talk with him, I left and suggested that the referring physician contact behavioral therapy, another service available in the hospital. By way of context, earlier that week, I had had a suspicious mole removed. The dermatologist told me he was “ninety percent sure it was not a malignant melanoma.” A 10% chance was very frightening to me, and I was in no shape to try to help this challenging patient. Of course, I should have, as Dr. Humphrey advises, “checked my baggage” and found somebody who was in a better position to help the patient. Unlike the story about the smell of chlorine, this is one I would like to forget. I learned that talking to dying patients was extremely rewarding if done correctly, but it took too much out of me emotionally. It drained energy that my young children needed. I had a pretty good understanding of my psychic baggage, but knowing was not enough. I had to stop working directly with dying patients. This realization made me have an even greater respect for physicians and other health professionals who are able to look death in the face day after day as they work effectively with dying patients and their families. In Chapter 20, Dr. Humphrey describes how she manages to do it. Early in my training I, like the fellow pediatric resident I described earlier, was very critical of physicians who were uncomfortable with and avoidant of dying patients. With time, however, I became more understanding of health professionals who, like me, found that close involvement with dying patients was too much to handle. To expand on this thought further, I will tell another story about Dr. B, the pediatric oncologist who avoided

198   Stuart J. Youngner his dying patients. One day, I  received a call from Dr.  B asking me to see one of his patients, a 16-​year-​old boy who was dying of a sarcoma. I said I would be happy to do so. Then, Dr. B said something that surprised me. “If you take this case on,” he said, “you have to stay to the end. You can’t pull out.” Somewhat puzzled, I reassured him I would not. The patient was a wonderful young man named John who died in the hospital over a three-​month period. I saw him almost every day and developed a close relationship with him and his family. They were all quite open about discussing the impending death and their fears and expectations about it while at the same time trying to be positive. They all commented but did not complain that Dr. B never entered the patient’s room. He only stuck his head in the door every day to ask the patient perfunctorily, “How are you doing?” He did, however, speak with the parents separately from the patient. One day, I got an urgent call from Dr. B. “John blew up at me today. He wants to fire me,” he said. “Tell him I would be willing to find him another doctor.” When I came to the patient’s room, he was furious. He railed about Dr. B and his failure to communicate directly with him. When I told him that Dr. B would be willing to switch his care to another physician, the patient became very agitated. “Oh no,” he said. “That would be terrible.” The patient’s family felt the same way. I called Dr. B and told him everyone wanted him to remain the physician in charge. The next day, when I saw John, he told me the following story. That morning, Dr. B had altered his routine in one way. When he briefly stuck his head in the door to say hello to the patient, he was wearing a clown mask. He never explained why and never apologized to the patient or changed his routine; he never wore the mask again. Things returned to their pre-​blowup status until the patient died three weeks later. I was there. There was much crying and hugging among family, staff, and health professionals. I looked around for Dr. B. The families on this cancer ward knew each other well, and John’s death was a major event. Parents were standing anxiously in the doorways of their children’s rooms watching the grieving that had spilled out into the hallway. Dr. B was there, too. He was going from room to room reassuring the worried parents. What a job he had, day in and day out. I guess he was able to fulfill an important part of his responsibilities—​treat patients and talk with families—​but he called me in to talk directly with the patient. Was his admonishment to stay to the end his way of saying that he could not handle the job alone? As a footnote, I should mention that after making his rounds of other parents immediately after John’s death, and after I had finished with John’s family, Dr. B and I sat down for a cup of coffee. He spontaneously revealed to me the awful experience of his own mother’s death.

Death and Dying: The Next Phase It is interesting if not entirely coincidental that my inadequate response to the case of the trembling man occurred about the same time I turned my attention to the problem of when to withhold or withdraw life-​sustaining treatment. In the mid 1970s and early ‘80s,

The Smell of Chlorine    199 at the same time as I was seeing dying patients, high-​tech medicine and the creation of intensive care units appeared on the scene. While it might be difficult to imagine a time when medical students were taught to lie to their patients about fatal prognoses, it is equally difficult to imagine acute care hospitals without the intensive care units that now make up nearly 20% of their beds. For me, this area of inquiry was very exciting and free of conflict. I could talk about death in a way that was much less personally threatening. Does my experience mirror a similar phenomenon that occurred among health professionals and American society at the same time? In retrospect, the conversation stimulated by Kubler-​Ross seemed to fade into the background as health professionals, the courts, the public, and the news media became preoccupied with the right to refuse life-​sustaining treatment and the possibility of death as a least-​worst alternative. Suddenly, patients were hooked up to machines and monitors that saved and prolonged their lives. They were placed in special units with staff who were trained to deal with life-​threatening illnesses and to operate the machines that fought them off. Very soon, however, the new technology created clinical situations in which patients were prisoners of the machines and specialty units that were intended to restore their health. Some people began to question the application of this new technology. Must we always keep the patient alive? Can we turn off a mechanical ventilator or stop dialysis? Can death sometimes be allowed to come as quickly as possible? There were no ready answers to these questions in the law, hospital policies, professional guidelines, or the bioethics literature. It was all so new—​which made it exciting to me and to many others. The talk no longer seemed to be about helping patients face their impending deaths. Rather, it had shifted to discussions about patient and family rights, harm–​benefit ratios, active versus passive measures, advance directives, and surrogate decision-​making, prognostication, quality of life, double effect, and terminal sedation. These discussions became more trenchant when talk turned to physician-​assisted suicide and active euthanasia. Overall, attention seemed focused on controlling the timing of death, rather than on looking death in the face. Are these issues important? Of course they are, especially in the unavoidable context of high-​tech medicine. Most of the chapters in this volume deal with these issues. Yet one must conclude that patients, families, physicians, and nurses still must look death in the face—​but perhaps not quite as directly. If “control” means escaping death entirely, we will never achieve it. If, on the other hand, we can control the timing, the issues of who, how, and when stay in the forefront. The existential issue can remain in the shadows, where it is less overwhelming. It seems that modern acute care hospitals have turned the task of talking with dying patients over to hospice and palliative care. Physicians such as Dr.  Humphrey are trained and hired to do this difficult work. Yet, I wonder whether modern hospice, and especially palliative care (practiced in tertiary medical centers), gives adequate attention to patients and families who cannot escape the existential issues. In their discussions of hospice and palliative care, Clark and Cassell hint that there may be reason for apprehension. Clark, for example, notes that, early on in the hospice movement, there

200   Stuart J. Youngner was concern that psychosocial and spiritual concerns were being shortchanged by an emphasis on physical symptoms (Clark, this volume). Even if spiritual and psychosocial care are emphasized in hospice, most dying patients only spend their last two or three days in this environment. Clark raises similar concerns about palliative care when he talks about a “consumerist orientation” in which the model of care must follow the demands and orientations of the patients and users. It argues that delivering palliative care to achieve better symptom control, improved communication and greater alignment with patient and family wishes are primary goals and that only once these are achieved is it appropriate to engage with questions of meaning, mortality and the reality of death (Clark, this volume).

In Cassell’s chapter about suffering, he notes the limitations of a fact-​based, objective emphasis in medical science: In fact, our knowledge of persons in the 21st century is not as great as our knowledge of them was in the 1950s or 1960s. At that time, there was intense interest in the psychological makeup of persons and the psychodynamic characteristics common to all persons… . That has all disappeared from sight as the scientific worldview has become a dominant social force. Most of these facts about persons are subjective, and subjectivity is greatly disvalued at this time (Cassell, this volume).

Most ironic, perhaps, are the conclusions of Tracy and Morrison in their chapter on “Marketing Palliative Care.” Successful targeted marketing of palliative care, they say, should avoid the subjects of death and dying.

Conclusion By sharing some of my personal biography and the cultural context of my own professional development, I have tried to illuminate some of the difficulties inherent in the efforts of health professionals to help dying patients and their families. Modern medicine and the modern approach to the care of dying patients seems to focus most comfortably on choice and control. One hears less about the psychological and spiritual ramifications of helplessness in the face of impending death. Should we be confronting these issues more? Perhaps our society, with its emphasis on “scientific” knowledge, self-​ realization, and personal freedom, may have projected a new symbol upon which to focus its inescapable fear of death: the possibility of prolonged and isolated dying in the hands of strangers and attached to machines amid the uncomfortable mix of medical uncertainty and vanishing hope.

The Smell of Chlorine    201

References Kübler-​Ross, E. 1969. On death and dying. New York: Routledge. Pattison, E. M. 1967. “The experience of dying.” American Journal of Psychotherapy 21(January): 32–​43. Weisman, A. D. 1970. “Misgivings and misconceptions in the psychiatric care of terminal patients. Psychiatry 33(1): 67–​81

chapter 12

Talking and Worki ng with Dy ing Pat i e nts  True Grief and Loss Lisa Humphrey

Like all physicians, I am impacted by the death of a patient. I have patients I remember with a wistful smile as I remember their antics or their laughter. I also have cases that haunt me. I see the patient; I remember the family’s reactions when I spoke to them about their dying child. I ruminate about what I could have done differently, replaying events and decisions. I have moments of self-​doubt followed by reassurance that I did the best I could with the data I had at the time. I also have a few cases where that reassurance has yet to come. Unlike most of my physician colleagues, however, I am a pediatric palliative care and hospice physician. As such, I am expected to be “okay” with death and be an expert in managing grief. Yet when I did my fellowship, there was no evidenced-​based practice to teach me how to become comfortable with these expectations. I was given no guidelines, let alone evidenced-​based research, during my fellowship on the topic of physicians and grief; rather, I just learned through experience. Certainly, mentorship and apprenticeship are a time-​honored tradition in the art of medicine, and I was a recipient of great teaching. However, there is a growing sense in medicine that we can do better. The goal of this article is to impart how I view, engage, and incorporate grief into my work as a physician. Unlike the other chapters in this book, this one is not driven by evidenced-​based research or best practice models; instead, it stems from experiences that I have had and lessons learned from my mentors, patients, and families. My hope is that readers will be challenged toward self-​reflection and perhaps provided with some skills to allow them to analyze and, if necessary, adjust their relationship with grief so they can practice both the best medicine and self-​care possible.

Talking and Working with Dying Patients    203

Identifying Grief As a second-​year resident, I remember a night in the emergency department. A girl came in with profound emesis and profound lethargy. The differential was daunting, but ultimately we determined that she was in diabetic ketoacidosis and diagnosed her with diabetes mellitus. I eagerly entered the room to discuss what I considered to be relatively “good news” as it was a manageable disease process. Clearly I was naïve, for as I relayed the information in my most empathic manner, I invoked a wailing from her mother that lasted forty-​five minutes: enter my understanding that death is not a mandate for grief and that one can grieve the promise of a type of life as much or more than the loss of a life. Since that embarrassing educational moment, I have strived to better understand the types of loss one can experience. I have subsequently learned about the loss of “normalcy” that comes with the diagnosis of a chronic condition. I have also been educated on anticipatory loss, which is the knowledge that loss is coming and the grief one feels anticipating this loss. There is also the grief of an imminent death both for patients and their loved ones who sit vigil. Finally, there is the grief that I experience as a health-​care provider. A grieving health-​care provider comes in many forms. I have provided empathic listening to colleagues as they cognitively express their grief by reviewing labs and images and their frustration that their responses to this data did not “save the patient.” I have also been there as colleagues cry and remember what a patient’s life meant to them in a very personal way. Finally, I have had to acknowledge my own grief, which has been the hardest for me to both recognize and accept. My grief is that no matter how good I am as a palliative care and hospice physician, I cannot protect those crying in front of me from the waves of grief that are engulfing them. My grief is one of being so humbled at the idea that I personally have never yet felt the loss that my patients and families are experiencing but knowing that I will, and that anticipatory grief takes my breath away. I package this grief away on the vulnerable days and, in doing so, I risk being a less effective physician. Evidence supports this risk. Finlay and Dallimore (1991) asked parents who provided better disclosure regarding a patient’s death after a trauma: the police officer at the scene or the physician. Participants found police officers to be superior, noting that they exhibited greater compassion by showing their distress and concern and demonstrated less professional detachment. I first read this study when I was a fellow, and I strongly questioned the findings. However, as my fellowship continued, I reflected again on the study and began to recognize its truth. In general, I think many health-​care providers seek restraint over emoting when participating in end-​of-​life care and view this as a conscious act of professionalism. However, I think this also serves to create boundaries that are often a primary pillar of self-​care in medicine. If one does not emote, others do not emote (as much), and thus one can protect oneself from a wave of emotions. We can avoid facing our own vulnerability.

204   Lisa Humphrey Vulnerability is a scary word for a physician. Many of us, I suspect, are drawn to the profession because we gain comfort in identifying the bucket that a patient’s symptoms fall into, applying the labeled diagnostic sticker on the bucket and then following the evidenced-​based algorithm to assist in making a patient’s life better. We feel in control, which minimizes our sense of vulnerability. However, emotions do not often follow algorithms, despite medicine’s attempt to force them into one through communication acronyms, cue cards, and controlled simulation cases with mock patients. And while these educational interventions have improved our ability to communicate, I  would argue that these actions have not improved our ability to self-​assess and understand our own emotional relationship with death and grief. Yet for me to have the courage to be vulnerable, to be more demonstrative of my grief with patients and families, and to improve end-​of-​life care, I needed to gain understanding through introspection in two particular areas: my own experiences in death and grief and my expectations regarding how an emotional event such as a death should occur.

“Tagging Your Baggage” I have coined the term “tagging your baggage” to represent the necessary work of recognizing and labeling prior loss experiences. All of us have experienced loss, and we retain the impact of these losses such that we experience future loss through the lens created by prior losses. The accrual of loss is the baggage we carry with us. The task at hand is to be aware of our baggage and tag it so we know what is shaping our lens. For me, tagging my baggage is an active step of checking in with myself before entering a room. Am I  anxious about the pending conversation? If so, why? Am I  overwhelmed with a sense of protectiveness for the patient and/​or family? If so, why? Am I emotionally exhausted and pondering the feasibility of having a resident or nurse give the bad news because I just realized I have an email that has become more paramount than the conversation at hand? If so, why? These are just examples, and I cannot guess what each reader’s baggage might be. The point is that we all have a piece or two of baggage, and if we are not actively identifying and tagging that baggage, we unknowingly bring it into the room with us when discussing difficult news, which in turn may negatively affect our effectiveness as physicians. Two examples follow.

Case 1 I was consulted on an infant who had just been diagnosed with a lethal peroxisomal disorder. As I approached the room to meet the family, a doctor from a different consulting service emerged from the room. She became visibly upset on realizing that palliative care had been consulted. She then quipped that the “murder for hire service” had just been

Talking and Working with Dying Patients    205 consulted. Having become used to not being anyone’s favorite consulting service line but surprised by her vehemence, I asked if she questioned the high risk of mortality associated with the diagnosis. She acknowledged the known life expectancy of less than one year but voiced concern that the family was not ready to be devastated and felt that palliative care was inappropriate to insert. Over the next several days, the physician’s discomfort with my participation in the patient’s care plan escalated, and she stated the parents’ discomfort in my presence warranted the discontinuation of palliative care. No one else had detected parental discomfort in my presence, and this was relayed to the physician. Still unsatisfied, the physician asked that I accompany her while she asked the mother if she was comfortable with palliative care’s continued participation in her child’s care. Initially, the mother had difficulty understanding the question or why it was being asked. This awkwardly led the consultant to explain her concerns that my presence made the mother uncomfortable and her desire to protect her from this discomfort. The mother then replied, “My daughter is going to die. Of course I want the person who specializes in this to be in the room.” After this encounter, I gave the physician full credit in her willingness to collaborate with me. I had the sense that the tension had broken and her fears had dissipated. The patient lived for six more months, and, per the family’s wishes, she died at home in her mom’s arms. Since that encounter I have had the opportunity to develop a very good working relationship with this physician. A year after our initial encounter and over an innocent cup of coffee, the physician relayed that in her childhood, she once had a sibling. At approximately six months of age, her sibling fell ill and went to the hospital but never returned home and was never spoken of again. It was not until adulthood that this physician insisted on knowing what had happened to her sibling. She admitted that pediatric death was and would continue to be traumatizing to her. This baggage had colored her interpretation of this mother’s actions and intentions.

Case 2 When I was completing my palliative and hospice medicine fellowship, my grandmother’s overall health status rapidly declined. My aunt summoned me to her bedside, and there I realized that my grandmother was actively dying. My aunt inherently must have understood a part of my grandmother that I did not, for she comfortably decided that it was time for her to go home. I decided that I would stay and give the gift of presence as my grandmother died. I called my father, an internal medicine physician, and discussed my grandmother’s current status. He lived approximately eight hours away and, based on my assessment, would not arrive in time to be with her. I thought my presence would bring him comfort, so I emphasized this over the phone. To my surprise, he announced that I was incorrect and that my grandmother was not dying. To this day, I am uncertain how I so easily abandoned my professional self and became just a daughter. Yet I did, and as such, I immediately agreed with him and followed his directions to go home. That night, my grandmother died, and no one was with her.

206   Lisa Humphrey To this day, my guilt regarding this is a piece of my baggage. I catch myself needing to be superhuman and trying to be present and available to all of my patients and families as a child lies dying. However I have also come to realize that this baggage can impede my ability to correctly read a situation. The reality is that I am not the perfect fit for every family, and my presence does not bring everyone solace. Some people need to complete this journey by themselves, and my baggage creates a risk of me not detecting this nuance and inserting myself where it is not desired. Additionally, I do not have the emotional capacity to be that present for every patient, so I run a real risk of this baggage inducing burnout. Now I check in with myself on each and every case. I label my baggage and determine whether or not it goes into the room with me. I hope that it allows me to be the best fit I can be for each patient while simultaneously allowing me to remain balanced and fulfilled in my interactions with my patients.

Expectations Like our own baggage, we need to also be aware of our expectations when working with a patient and/​or family facing end-​of-​life issues. A pervasive concept in palliative and hospice medicine is known as the “good death,” and it is a noble and worthwhile concept. However, I worry that we unknowingly assign a unidimensionality to the concept, which then undermines our intent and our ability to assist a patient with attaining a good death. Two cases to illustrate follow.

Case 1 I had the opportunity to work with a reticent seventeen-​year-​old male with terminal sarcoma. Due to refractive pain, he was admitted for inpatient end-​of-​life care. We spoke with him about his goals for end-​of-​life care, but he retained his quiet ways and would discuss only his desire for better pain control. We also had several conversations with his mother regarding his imminent death, but she voiced an inability to come to the hospital and no other family was available to visit. The patient ultimately died with a familiar hospital chaplain at his bedside and nursing support that expertly managed his pain. Afterward, I overheard a distraught oncology nurse voicing her anger that he had “died alone” and labeled it a “horrible death.”

Case 2 I also had the opportunity to work with a hospitalized six-​week-​old identical twin with a glycogen storage disorder who was actively dying. His twin was also affected but not as

Talking and Working with Dying Patients    207 severely and currently lived at home with his mother. We had a series of conversations with the mother regarding his declining status and his imminent death. She visited but ultimately stated that she could not bear witness to his death, only to go home to his twin who would also one day die. The nurses were stunned and scared by her absence, voicing their lack of experience with pediatric death and a worry that he would die alone. They came up with a schedule whereby nurses volunteered their time to hold him. This schedule lasted for five days until he died in the arms of one of the nurses. The nurses described his death as “beautiful” and the event as “an honor.” I was struck by these two cases, for they involved patients dying without the presence of their families yet the staff ’s interpretation of their absence was received so differently. I believe that the different expectations of the two nursing staffs impacted their perceptions. Death is an expected part of an oncology nursing job description and, as such, they frequently experience death. To manage this difficult part of their job, they have an expectation that, should their patient die, they will have the opportunity to provide the patient with a good death. In speaking with the oncology nurses in this case, a good death meant having family present. Conversely, the other group of nurses had no experience with death and thus no voiced expectations of what would constitute a good death. They were able to accept the situation and define a good death for this individual based on this individual situation. As such, they experienced a good death and felt that their patient did as well. While I too strive for a good death, I often quip that its acquisition is the holy grail of hospice medicine. There is no one set definition of a good death, and the reality is we do not get to define a good death for others. The definition of a good death is intimate, unique to the individual, and based on past experiences, cultural norms, and ongoing hopes. As much as it upsets me to see a cystic fibrosis patient with profound dyspnea refusing opioids based on a long history of fear of opioids or a young adult telling me that he will pursue inpatient palliative chemotherapy solely for the sake of his mother, I remind myself that a good death is not mine to own or define. I try to modify my expectations to permit these nuances and refocus myself on what I can impact: for example, the best pain management I can offer in light of a patient’s wishes, the attainment of a legacy piece, or helping a physician forgive herself for not curing the patient.

Communication Style To manage all one’s baggage and expectations while simultaneously utilizing exceptional communication and supportive skills to dying patients and their families is a daunting task. I wish I could simply provide a hand-​out with the algorithm that would ensure, based on evidenced-​based medicine, success in all of these arenas. I have even gone so far as to create the algorithm in the figure. However, I present it with the disclaimer that it is not foolproof.

208   Lisa Humphrey Grief work identified

Tag your baggage

Manage expectations

Patient’s grief work

Identify audience’s communication style

Identify your audience’s needs

Your grief work

Communication toolbox

Debrief

Self care techniques

Medical schools have done admirable work increasing medical education on end-​of-​life care. Through simulated patients and didactics, physicians are gaining the foundational tools to help patients and families through decision-​making and symptom management. However, the task of assisting patients and families with their grief still remains daunting to many physicians. Some of that is grounded in the fear of opening Pandora’s box: If one starts a conversation, there is no telling where it will go, and for people who love algorithms that uncertainty can be terrifying. Some may also fear the emotions inherent to loss and would prefer to neither witness nor experience them. Finally, we fear that we could fail at being supportive and cause more harm than good. I have no simple answers to any of these concerns, but I do believe that conscious efforts to learn a patient’s communication style and needs in addition to develop one’s own communication toolbox can allay fears and maximize one’s impact.

Identify One’s Communication Style I completed my fellowship in palliative and hospice medicine under the guidance of seven physicians, a nurse practitioner, an art therapist, a social worker, and a chaplain. In other words, I trained with eleven people with eleven styles and confirmed the old adage that there is more than one way to skin a cat. I also learned that the singular theme that allowed each of them to be superb communicators was their willingness to be authentic. Patients and families perceive this authenticity as humanity and, through

Talking and Working with Dying Patients    209 this, partnerships are created. These partnerships then create trust so that hard conversations can occur. I soon realized that I needed to incorporate authenticity into my communication style if I was going to succeed. Over time, harnessing my authenticity has unintentionally led to a communication style that is antithetical to any professionalism ever taught to me. I speak to families and patients as I write: casually, with an assumption that I am somewhat amusing, and absent any assumptions of intellect. I assert myself as such because that is my comfort zone, and it has allowed me to best partner with my patients and families. I acknowledge to patients that I was a feisty kid who landed herself in the principal’s office for fighting, so I get what it means to fight the good fight. I announce that I was an average medical student, so I am incapable of effectively using fancy doctor words but pretty good at translating what my smarter physician colleagues say to patients and families. I confess to reading trashy gossip magazines instead of medical journals at home, because I too need distraction at times of stress. I weave these tidbits of myself into our conversation regarding goals of care and how to help a child die with dignity and I hope that parents will see my humanity and open up. It also reveals my fallibility, which helps me at my job because I can focus on being present in the moment, since perfection is no longer an option.

Identify the Audience’s Communication Needs It would be incorrect to say my preferred communication style is the only style I use. Patients bring their own communication needs, and to be most effective in assisting them with their grief process, I need to identify their style and then modify my own to maximize the experience for both of us. I have learned to pay particular attention to the following: 1.  What are the patient’s goals? This question appears self-​evident. Most answer “a miracle,” and then they give the look of “Why would you entertain a miracle?” I ask it because I prefer hearing it out loud than trying to avoid it. Asking the question allows me to provide empathic listening while they discuss this hope and acknowledge it as a commonality of the human spirit to desire a miracle in times of need. I also ask it so I can then ask what other goals they have—​for most people have other goals. Armed with this list, I can partner with them to mitigate their suffering through the acquisition of goals such as improving pain control or obtaining forgiveness from others. When I know this list, I can modify my communication style accordingly. 2.  What cultural norms do I need to know about? I ask these both in regard to patient’s self-​identified communities (i.e., Islamic, Japanese, lesbian, etc.) but also in regard to a patient’s family microculture. The family may identify with a culture in which women defer to their male partner’s opinion, but this same family may have a very influential granddaughter who holds a subtle but important position when decisions are made. Similarly, just as

210   Lisa Humphrey individuals have baggage and expectations, so too does a family unit, and these will similarly alter the family’s filter for incoming information. To best partner with a family and have a message received well, it behooves one to assess these dynamics and adjust communication accordingly. 3.  How does the listener like to receive information? Some patients love data and others appear overwhelmed by it. While this likely represents a learning style, it also can reveal how much information a listener can emotionally absorb at one time. Both of these issues, learning style and emotional capacity, are paramount when gauging communication style. Sometimes one can easily detect their styles. When I cannot ascertain it, I just ask the patient or family. “How do you like to receive information? For example, some families prefer lots of data while others prefer their doctors to deliver a summary statement. “We have a lot of complicated and difficult stuff to discuss. Which way would you prefer us to speak about your health-​care issues?” I have been impressed by how many times patients and families have a clear understanding of their style and a strong sense of how they wish to receive information. I have also encountered families in which individual members have differing styles. I once had a patient whose father was an extrovert who needed to know every lab value, every medical consideration, and everyone’s opinion. It was also clear that his wife preferred not to ask or be asked questions unless absolutely necessary. I had an open conversation regarding my observations to determine if I was correct. My first communication step was to make some general rules of communication, which included having separate conversations with the father in addition to those I had with the mom. Despite these different communication styles, they were able engage in collaborative decision-​making and, as a result, made excellent decisions for their daughter. However, to have a maximal relationship with them, I needed to adopt different communication styles for each of them. 4.  Is the patient ready to hear the news? Patients’ and families’ readiness to receive difficult news as a requisite for successful, collaborative communication at the end of life cannot be overstated. Often mislabeled as “denial,” lack of readiness is subtly different. Denial is often a word we invoke when a patient “does not get it.” I am always amused, in fact, when I have a consult as a palliative care physician to essentially eradicate the patient’s denial. I then need to reveal to my colleagues that I do not own a magic wand that will dispel denial, nor am I necessarily in agreement that it is the right thing to seek. Denial is a healthy coping mechanism that permits our patients to withstand the burdens of medicalization in an attempt to cure. It allows patients to persevere when we wonder where they get their resolve to do so. Denial also allows every health-​care worker to come to work every day. How else do I look at a seven-​year-​old with cancer and not panic that my own daughter is in jeopardy? So I do not view denial as my enemy in palliative and hospice medicine. Rather, I see it as my job to assess readiness: readiness to receive information, readiness to grapple with the notion that God did not answer prayers, readiness to tell one’s

Talking and Working with Dying Patients    211 wife that she will be a widow, or readiness to face the unnatural status of being a parent to a child that has died. Sometimes health-​care providers do not have the luxury of time to await readiness because clinical needs are urgent. Often, however, we have some time. I spend that time looking for the clues and teach people that the patient’s or family’s readiness journey mirrors that of a smoker pondering cessation. They are precontemplative for a long time and utilize denial during that time as they prepare themselves for the realization of what they must do and endure. Our job as we wait is to have faith that the listener will come around and to modify our communication to acknowledge gently those touch-​points of reality, to not pressure the listener to move faster than he or she is able to, and to listen for the cues that the tide has changed and that we may communicate in a much more forthright manner.

Create a Communication Toolbox It is often around this time in my didactics that a look of fear with a hint of discomfort emerges on readers’ or listeners’ faces. Comprehension of the task at hand occurs: I expect them to present with authenticity but to simultaneously be communicating chameleons that effortlessly adapt to the needs of their listeners. They are officially terrified at the idea of opening Pandora’s box lest they lose all control of the situation and the feared complication of making a bad situation worse materializes. This is when I introduce the notion of creating an individualized communication toolbox. A communication toolbox is a self-​identified collection of phrases we feel are consistent with our individual communication styles, have shown effectiveness in discussing end-​of life issues, and that we believe will help us manage the feared complications. Such a toolbox is indirectly referenced when medical students are asked to review cases where bad news was delivered effectively and to remember these phrases when it is their time to give bad news. While I agree with this exercise, I would insert that not all phrases are meant to be uttered by every individual. I learned this all too well during my fellowship. I was the very first fellow at my hospital; therefore, my teachers were learning how to teach palliative care through me. I had an esteemed attending who eagerly wanted to impart all of his knowledge to me, and I desperately wanted to learn from such a skilled physician. We shared many cases together, and, after some time, I reflected on my observation that men responded better to him than me. He doubted it, but we decided to dissect his approach with the mutual assurance that I would add a tool to my toolbox as a result. We came up with a communication plan and I applied it to the next grieving father that we met. I walked into the room with a decidedly solemn look on my face. I acknowledged both the mom and the dad in the room but went to the dad’s side. The mom was tearful; the dad appeared stoic. I remained, as planned, pensive, and then after approximately ten seconds, I placed my hand on his shoulder and paused there. I then let out

212   Lisa Humphrey what I thought to be an empathic exhalation just as my mentor often did, utilizing his words, “You have been a good father.” Silence ensued, and not the type of silence we palliative care physicians like. Rather it was an awkward, “please, just let this end” moment as the father purposefully shrugged my hand off of his shoulder, thereby requiring my attending to intervene and repair the damage I had just caused. My attending and I walked out of the room soon after and just stared at one another. Neither of us had an immediate answer as to why the very tactic when taken by my attending often created tearful, emotive, collaborative fathers but had, when utilized by me, caused a near palliative care disaster. However, I was able to immediately process how I had felt when approaching the dad and especially when I had placed my hand on his shoulder. I felt awkward and imposing. I do not like physical contact with my parents even with a gesture as innocent as the one I had performed. Upon further reflection, I am certain that the dad felt my hesitation and assumed I was uncomfortable with the material at hand, not with the physical contact, and he assumed insincerity in me to the point that I could not make the connection my attending could. Then I began to realize that a toolbox is an individualized collection based on some degree of trial and error. Some tools I could wield with greater ease than others, and I was going to have to risk failure in order to ascertain the comfort and effectiveness of each tool I came across. These days I tell people that they need a communication toolbox, but it has to be filled with phrases that they have tried and that they feel comfortable utilizing. I encourage them to listen to others but to be prepared for epic successes and failures when trialing out new tools. I also encourage them to reassess their toolbox continually, for the phrases that worked for them in one part of their career will likely grow stale or become ineffectual for them over time. My communication toolbox includes the following (note that my mentor is in full agreement that the dramatic sigh has no place in my toolbox!). 1.  I am a fighter and will fight alongside you, but we need to make sure we are fighting for the right thing. Let’s talk about what equals the good fight these days. 2.  I am worried … 3.  I have been thinking about your child a lot and I am fearful that we have some hard choices to make together today. 4.  Well, this stinks. 5.  I am so frustrated for you. 6.  I cannot promise you a crystal ball, but I promise you my honesty.

Grief Work None of this work comes easy. It is emotionally draining to have difficult conversations, even with a communication toolbox at hand. We as physicians also experience grief

Talking and Working with Dying Patients    213 when a patient dies, and so we have our own grief work to attend to. Grief work is the process one must complete in order to resolve grief. Created as a concept by psychiatrist Erich Lindemann, it suggests that one must actively work toward resolving one’s grief rather than passively experience it. While there many books written on the topic that are outside the scope of this article, I do encourage readers to create a grief plan that incorporates, at a minimum, debriefing after an event and creating an identified self-​care plan.

Debriefing At each hospital I have worked, I have been asked to participate in the creation of a formalized debriefing program for employees. The discussion typically reveals that prior attempts have been made but have faltered secondary to poor participation. This is not unusual in my experience. There is also ongoing debate regarding the effectiveness of formalized debriefings and even the suggestion that they may cause more harm than good, for they may disrupt the natural grieving process that is required to maximize recovery (Bledsoe, 2003). Because of this, I encourage institutions to await the creation of a systemwide debriefing program until more is known. For individuals, however, I do encourage finding some way to debrief, especially if one recognizes signs of acute stress disorder. This may involve seeking out a mental health professional or having a cup of coffee with a close colleague to discuss the events. It may be a simple acknowledgement of “Whoa, that was a rough one,” with the person who accompanied and then no further acknowledgement. Note that I encourage people to always have a fellow health-​care provider accompany them when giving difficult news. They then have someone who can review the meeting with them and assess if it went as well or as poor as the physician suspected. It also means companionship. Even if it is never verbally acknowledged, having someone in the foxhole with us means we will always have a sense of unity when we encounter them in the future. Through this, the loneliness that can accompany this difficult work may dissipate.

Self-​Care I present this article as a lecture to second-​year medical students every year. When I get to this point in the discussion, I see the immediate rift in the audience: some welcome conversations on self-​care, and others are leery of it, for they have visions of having to overshare or emote, and neither resonates with their coping style. Readers may have the same response, but hear me out! Self-​care is a very idiosyncratic, malleable technique. Do not assume that it must be meditation or incantations and then disregard it if that is not appealing to you. Rather, each person needs to find out what refills his or her gas tank at the end of an emotionally draining day. Next, identify it and distinguish it as part of a self-​care plan and utilize the

214   Lisa Humphrey activity only when actively seeking self-​care. I find this to be of the greatest importance; otherwise, its impact gets lost in the fast-​paced lives we all lead. Also, people must complete a self-​care plan each and every time they identify having had a taxing, emotional day. Some examples of self-​care elements I have encountered are

1.  Playing a specific song only when having an identified need for self-​care 2.  Drinking a cup of tea out of a particular cup reserved for self-​care events 3.  Going to the gym and completing a particular workout reserved for self-​care days 4.  Going to a place of worship during off hours with the sole intention of self-​care, not to complete a duty to attend 5.  Playing a particular game with one’s children 6.  Meeting girlfriends for dinner with a publicly announced need for self-​care Not all of these will resonate with all readers. Each person needs to find his or her equivalent. The most important part is to establish the tradition and be true to it. When I was a fellow, I met a beautiful, sixteen-​year-​old girl with an equally beautiful personality. She did not walk down the hallway in the oncology outpatient clinic so much as she appeared to float down it as she giggled, smiled, and engaged with everyone she met along the way. When she was in the hallway, multiple exam room doors would fly open by excited staff and fellow patients who wanted to be acknowledged by her. I was equally infatuated by her. She was an extraordinary girl with an extraordinary gift, but she was also a patient with a terminal diagnosis, which made knowing her a gut-​wrenching experience. She was also very intelligent and courageous, so she asked questions like “What is it like to die? What will my younger brother witness as I lay dying?” Most stressful to me, she relayed that she had befriended another patient of mine, also an adolescent girl with terminal cancer. She told me they would stay up for hours at night discussing their impending death over the phone, but she never told me the details of those discussions. To this day I cannot determine if that omission is a blessing or a curse, for I have never been able to shake the image of the two of them sitting up in their beds, in the dark, on the phone looking terrified. I do not even know if they were terrified, but that is the image I retained for a long time, and it negatively impacted my grieving for I felt that I should have been able to better help this vivacious, intelligent, brave girl. My team of mentors encouraged me to attend her funeral as part of my grief work. It seemed like sound advice. However, I was ill prepared for the throngs of sobbing teenagers that barricaded the entrance to the church and even less prepared for the panicked parents of these teenagers who had no idea how to comfort their children. I felt powerless and disoriented. In the hospital, we have crowd control by virtue of the size of the room. We also have rules and policies. And it is true that I have a sense of control in my domain that I woefully lack once I leave a hospital. All these realizations and emotions came crashing onto me as I sat trapped in a pew for an hour and a half long service. After this encounter, I took time to process all of my feelings from that day. In the end, I realized that my self-​care plan needed to include not attending funerals. Now I go to

Talking and Working with Dying Patients    215 viewing hours only, and, even then, I attend just as it starts because fewer people are typically present. For me, this has been an effective self-​care strategy because it allows me to acknowledge the family’s loss but also gives me a small space to say my own good-​ byes and begin my grief work.

Final Thoughts My intentions in this article were to impart whatever pearls about grief and dying I have acquired through my training and career as a palliative care and hospice physician. I restate that I do not assume to have a perfect or singular solution that helps all. However, I do hope that readers now have something to consider: either about themselves, the grieving process, or how to engage with a terminal patient. I also hope that they strongly consider retaining my greatest pearl, which is authenticity. Authenticity allows us to partner with patients, better see our own blind spots, and identify our baggage, expectations, and grief work to best assist our patients.

References Bledsoe, B. E. 2003. “Critical incident stress management: benefit or risk for emergency services?” Prehospital Emergency Care 2: 272–​279. Finlay, I., and D. Dallimore. 1991. “Your child is dead.” British Medical Journal 302: 1524–​1525.

Chapter 13

The Nature of Su ffe ri ng Eric J. Cassell

I want to discuss the problem of suffering and establishing therapeutic goals for patients at the end of life. Before I do that, it is important to review what it is that makes suffering the unique form of distress that it is. I want also to look at the difficulty clinicians have dealing with suffering as suffering because of the persistence of habits of mind in which all Western clinicians were trained—​particularly the habit of focusing primarily on the physical problems of very sick patients, almost to the exclusion of the person of the patient. In the West, idea of the separation of body and person or body and mind is characteristic of medical science and persists, as does the inexplicable way that reductionism always seems to win. All these factors are evident in the persistent reversion in palliative medicine to “pain and symptom control” rather than to “total person care.” One hears discussions of different kinds of suffering. For example, people talk about existential suffering, a category in which the person is suffering by virtue of the fact or state of his or her existence. Often included in this category is so-​called “psychological” suffering. These kinds of suffering, it is said, are distinct from “physical suffering.” As we continue, I think you will see that there is no such thing as different kinds of suffering—​ there is only suffering. Suffering is most commonly associated with pain or other physical afflictions. It is now generally accepted, however, that pain and suffering are distinct and different forms of distress. Despite this, you will hear people again and again speak as though the physical stimulus is the suffering. For example, “he is suffering from terrible pain,” rather than “his terrible pain started his suffering.” The distinction between pain and suffering is exemplified by the fact that the magnitude of pain is only one factor in the distress it causes. People will tolerate very severe pain if they know its significance (what it is) and if they know that it will end. Pain of lesser degree may be poorly tolerated if it appears to be endless or is considered to have dire meaning (such as cancer). Other facts that distinguish pain and suffering are that suffering may stop even if pain continues, and patients may continue to suffer from pain even when it is not present—​for example, when the patient fears that the pain means a malignant recurrence. Virtually everything one says about the relation of pain to suffering is true of other symptoms as well. People

The Nature of Suffering    217 who have no symptoms may suffer, for example, because of the pain of a loved one, especially when the person is helpless in the face of that pain. Helplessness itself may be a source of suffering, as may hopelessness. There are two constant features of suffering. The place of the future—​for example, “if the pain continues I won’t be able to take it,” even though the person is tolerating the pain at that time. Suffering may be started by the fear of malignancy recurrence even when no such event has happened. Similarly, meaning always plays a part in the onset and even the continuation of suffering. It is what the pain or other symptoms mean that initiates the suffering, and it is what the continuation of the symptom means that brings suffering on the scene when previously there was only the symptom. These facts lead to the most crucial understanding that suffering has to offer. Bodies do not have a sense of the future, and bodies do not assign meaning. Only persons have a sense of the future, and only persons assign meaning. Bodies do not suffer; only persons suffer. This is the crucial truth of suffering. Suffering is the specific distress that occurs when persons feel their intactness and integrity as persons are threatened or disintegrating, and it continues until the threat is gone or intactness or integrity are restored. The key term is “person.” What is meant by persons? Persons are not selves, although persons probably have many selves. The person is the complete and total individual. A person is what we mean when we say “I am a person,” “Derek is a person,” “each of you is a person.” What we are referring to is not what Socrates or Alcibiades meant when they said “I’m a person in fifth-​century Athens.” We are not even referring to what your great-​grandfather meant when he said, to whomever, “I am a person” (except that he probably would not have said it). To say that bodies do not suffer and only persons suffer makes the person an object of medicine, and the person as a person was never the object of medicine before. In fact, this person, as in “bodies do not—​only—​,” is an invention of the twentieth century in the sense that, in our eyes, the person is more complete than ever before. Think of all the things that persons have now that were not there before. For example, subjectivity which, believe it or not, received its first official notice in 1830 when Kierkegaard wrote that a Christian was defined by the feeling of guilt for sin. And subjectivity did not make it into novels before the end of the nineteenth or beginning of the twentieth centuries when actual feelings were talked about—​feelings, emotions, private thoughts—​and the patient’s symptoms, which nowadays we want to hear. The patient’s needs, desires, concerns, and wants represent a new concern that many people believe should be the object of person-​centered medicine. It is also novel to speak of seeking out those goals and purposes that the sick or dying patient cannot achieve because of impairments of function—​from the molecular to the spiritual. The issue is not that such things did not exist before; they existed, but they were not an object of medicine. That is because the goal “I want to make things right with my brother before I die” is actual and objective for us, but is subjective to the patient. It is amazing how little we know about persons, even persons close to us who we think we know well. It is surprising that the focus on “person-​centered” medicine has not prompted a closer examination of the idea or concept of person. And it is equally amazing that curricula in medical schools that pride themselves on being

218   Eric J. Cassell person-​centered do not, in any school of which I’m aware, contain a course on the nature of persons. We would never teach about an organ or tissue without an in-​depth examination of those things. Persons are so complex, jointly and severally, that there is not one description that would satisfy the notion of person. The ancient saying that “the person is unknowable” continues to be true. However, although understanding of persons has increased exponentially from the middle of the nineteenth century to the present, there is still no one knowledge of persons, similar to that of the liver or heart: there are many. It is as though persons were a text and there are many interpretations. In fact, our knowledge of persons in the twenty-​first century is not as great as our knowledge of them in the 1950s or 1960s. At that time, there was intense interest in the psychological makeup of persons and the psychodynamic characteristics common to all persons. For example, a great deal of discussion focused on the “unconscious” and other psychodynamic features that were introduced starting with Freud at the beginning of the twentieth century and greatly elaborated upon in the ensuing decades. That has all disappeared from sight as the scientific worldview has become a dominant social force. Most of these facts about persons are subjective, and subjectivity is greatly disvalued at this time. The problem of objectivity versus subjectivity is made more difficult because of the prominent place in the social structure of our culture that science has come to play. So we are divided. We have the legacy of the mid-​twentieth century, when many were beginning to discover about personal subjectivity in depth. And we have the drive for complete objectivity that represents medical science and puts in doubt the importance and utility of knowledge of the subjective life of persons. It is necessary to understand the power that the embrace of objectivity and the object have on our lives and especially on medicine. It is one of the reasons it is so difficult to overcome the power of technology, whose allure is the provision of objective information. The evidence-​based medicine movement rests on the idea that all decisions should be based on medical evidence that is derived from objective research. Because of that, we now have many sources for the information necessary to decision-​making. The price, however, has been to diminish and mark as inferior the subjective information that good clinicians require in addition to objective evidence in order to understand and make decisions about their patients. Most have enough knowledge from the everyday observation of themselves and others to understand and treat suffering. Furthermore, we can all begin to learn about persons by observing our patients carefully over time. To do that, of course, requires actively paying attention and—​even better—​learning how to write descriptions of patients. Demanding of yourself that you write a description of a patient (it doesn’t have to be every patient) will teach more about persons than you would imagine. Persons are of a piece. Whatever happens to one part of a person happens to the whole. Whatever happens to the whole happens to every single part. Even after the passage of time since René Descartes in the sixteenth century, we still act as though there is a separation between mind and body. Now, this separation is between the physical and biological of medical science and the “psychological” mind or the “existential” self. Based on this historical

The Nature of Suffering    219 background, it is not easy to understand that “all persons are of a piece, whatever happens to one part happens to the whole.” Whether it is easy or difficult to understand, we are of one piece. A brief aside. As I noted, suffering is something that happens to persons not to bodies. This is true without reservation. But you could say the same thing about sickness and the same about pain. Nociception is a physical phenomenon, but if you really want to know about patients in pain you must know about persons. Sickness is a physical phenomenon, but it is also personal and cannot be understood as human sickness without knowing about persons. It is an axiom that all persons hear what they want to hear and do not hear what they do not want to hear; the same is true of clinicians. And the same is true of seeing and feeling. Therefore, good clinicians train themselves and constantly exercise the discipline to hear, feel, see, and smell everything. They try not to foreclose observation by premature interpretation until they have assured themselves that they know what is present. This is easy to write but difficult to do, and most difficult when clinicians are trying to know whether the patient is suffering because the information in itself is painful. It is impossible to carefully and fully hear the patient’s symptoms, what is burdening him, and his experiences and yet not hear who the patient is—​that is, that particular person. We are not interested in the usual things people come to know about people: whether they are nice, make good conversation, are attractive or unattractive, are smart or dumb, or cooperative—​all the social things. Instead, clinicians are interested in who the patient is in terms of his sickness, pain, symptoms, medications, and behavior. This is a different dimension of personhood. It is very useful knowledge and is enough for everyday medical care. It is not enough, however, for suffering and dying patients. One clue as to why that knowledge is inadequate for suffering patients is that a universal aspect of suffering is its loneliness. As you come to know someone, he knows you are knowing him. If you know him well enough, and he is aware of that, you break through the loneliness to some degree. Suffering can occur in relationship to any part of a person, but it is always because the stimulus to suffering threatens the integrity or intactness of the person. I will show you examples in a moment, but first we must address a crucial issue that is necessary to understanding. When pain is the stimulus, the suffering is suffering and not pain. When existential issues are the stimulus to suffering, it is the suffering and not the unbearable existence. When emotional issues are the stimulus, it is the suffering and not the terrible fear. When patients suffer, it is suffering and not pain. Suffering is not fear and suffering is not loneliness; it is suffering itself. Suffering is suffering, it is suffering, it is suffering—​ it is what it is and not another thing. What does it mean to say that suffering occurs when the intactness or integrity of the patient is lost or threatened? I have put together four brief scenarios that illustrate the concept. I believe you will recognize these as authentic episodes, but also recognize the patients—​even though the character descriptions of the last three are only a sentence long.

220   Eric J. Cassell What if? What if you were in remission from cancer of the breast for more than several years? And then you developed pain in the hip and went to the emergency department. What if you have been tolerating the pain for hours and hours and days and it never stops? Now you are on a gurney and waiting for a bed. You are exhausted because you haven’t been able to sleep. You try so hard not to cry and call out to every passing nurse or doctor. They don’t stop anyway because they’re too busy for you. And it goes on and on and on. And finally you have no stamina left any more … no control… . And you just start to whimper and you don’t even hear yourself because there is no “yourself ” any more, just pain. What if you have always been proud of your appearance, really proud, and a careful dresser maybe for 100 years? And then lately you started having diarrhea—​all the time—​for days on end. They will not give you anything to stop it because—​you’re not sure why. And when you finally come to the hospital and the nurse undresses you standing up, there is excrement running down your legs and on your clothes and everywhere. And everybody just goes in and out of your room. What if you always been a good patient and tried hard to do everything they asked you to do—​because you’ve always been good and did the right thing even as a kid? And everyone always said you were special. You came into the emergency room this time, and you have been lying on a gurney for 10 hours waiting for a bed. And you’re not sure what’s happening. And you’re so thirsty and cold, for hours, but nobody helps. When people do stop, they tell you how busy they are and to be patient, and when you try to find out what’s happening nobody listens. And some are just mean and short. And you’re so cold … What if you’ve always been really social and have a million friends because you love that stuff? And what if you just found out last week that you had cancer in your colon and they couldn’t operate because it’s in your liver too? And you’re sure you’re going to die soon. And you’re afraid to tell anyone because they’ll stop talking to you like they did when your mother was dying and when everybody ran away from her … and you … and left you to take care of her alone. That was awful. And you don’t even tell your girls because—​you know—​they have their own lives. And you’re scared, really terrified, about dying.

Even though these descriptions are abbreviated, they are recognizable people and the possible source of the disruptions or threat to the integrity or intactness of these persons should be recognizable. In each of these cases, except perhaps the first, it is not the physical symptoms per se that are the stimulus to suffering. In the second case, the woman with unremitting diarrhea, the injury to her vanity that has been a lifelong characteristic in an apparently beautiful woman is destructive to her as a person. In other words, it is not the diarrhea, as awful as that might be; it is the impact of being seen soiled and foul-​smelling. The next patient, somebody who has always been a

The Nature of Suffering    221 good girl, is injured because not only is she waiting on the gurney hour after hour, but nobody is paying any attention to her. The patient takes the disregard as abuse directed specifically at her. After all, she’s always been such a good girl and nobody abuses such good persons. Daneault and his group have shown that the suffering of hospitalized patients often started with their perception that they were being abused by hospital personnel (Daneault and Dion 2004; Daneault, Lussier et al. 2004). In the last case, the woman with metastatic cancer to the liver from the colon is terrified of dying. The abandonment of her mother (and her as caregiver) provides the script for what she fears will happen to her. It is important to understand that the fear of dying is often a fear of being alone, isolated, and separated from the group. The injury to her arises from what she fears will be isolation. Again, although she has physical disease, the injury that destroys her intactness is not physical, but social—​psychological. (It is important to keep in mind that fear of dying is often fear of isolation and abandonment, not of death itself.) These cases illustrate what it means to say that suffering is personal. Another person might simply hate the idea of being exposed while she is being undressed and stained by feces; as bad as that is, it isn’t the same as the injury that occurs to the vain woman in the same circumstances. Even the woman in terrible pain from what is probably metastatic disease to the hip suffers because of who she is and what she thinks it means, not just because of the pain. In her memory of that episode (which will never leave her mind), she may feel herself as having been abused—​as though it had been done to her purposefully. Or she may accuse herself of having been inordinately weak. The sound of whimpering when it occurs in those circumstances is very difficult for listeners to tolerate. It sounds like “uh uh uh uh uh,” and it is always the sound of a suffering patient. I have said that meaning plays a part in suffering and that is also illustrated in these cases. In each of the scenarios, it is the meaning of the circumstances that precipitates the suffering—​not only the physical insult. All suffering is unique and individual. That means that it does not matter what the stimulus is to the suffering; because of the nature of the person and the meaning of the stimulus, it will apply to that person uniquely. Even when the source of suffering is pain, why each person suffers—​not why each has pain—​will be individual and related to them as individual persons. Suppose two men are suffering from the pain of dissecting aortic aneurysms—​one of the most severe pains possible; what it means to each of them will be unique and individual and, therefore, so will the suffering. As an example, a patient had multiple pathologic fractures from widespread metastatic cancer of the lung, and these fractures were extremely painful. Even though it was possible to get the pain under control through radiation therapy, another fracture would occur and the pain would be once again severe. When asked whether she was suffering, she said that she was. When asked why, she said, “It’s food.” The answer seemed unbelievable because it was so obvious that it would be related to her pain. She continued, “You know, I’ve been married many years and every morning when my husband and I wake up the first thing we talk about is what we’re going to have for dinner. Then we begin to plan the day around dinnertime and the food. Now food is the last thing I want to

222   Eric J. Cassell hear about or think about and it has destroyed our closeness.” Suffering is unique and individual. Because of this, it isn’t persons in general who have to be known—​it is this particular person. I have been writing about suffering as though it were one thing. This is not true. Suffering can be present in varying intensities. It can be mild or severe. It can last a short time or go on for months and months. It is, however, a distress that is never forgotten because it involves the person to his or her innermost self. Does this variation mean that suffering is psychological, in the sense of affecting or pertaining only to the mental and emotional as opposed to the physical state of a person? No, because just as reading, writing a bicycle, or working as a physician may involve any aspect of a person—​physical, emotional, and social—​so, too, does suffering. In fact, it could not be otherwise because we are all of a piece; whatever happens to one part of us happens to the whole, and whatever happens to the whole happens to every part. Just as suffering reveals its ultimate truth that it is not bodies that suffer but persons, it also makes clear that dividing persons up into parts—​which may make sense in a reductionist view, leads away from the whole. Suffering always involves self-​conflict. This may seem strange at first because we identify the source of the suffering as outside or distinct from the sufferer, even when it is a disease that is inside the person; it is inside, but not of the person. The same is true when the source of suffering is existential or fate itself. The clue to the apparent contradiction is that meaning always enters into suffering, and we are not always of one mind about what something means. Sickness changes the course of life, but sickness may not change goals or purposes, particularly when it first starts. I want so much to continue my career, but I cannot because of what’s happened to me. I want to be a loving child or partner, but my suffering prevents my expressing the love that I want to feel and that I am not feeling. Sometimes suffering occurs in situations that put to the lie something extremely important to the person. The classic example is Job, who is always used as an exemplar of suffering. No one would ask whether Job believed in God; there is no Job where that would even be a question. God is as much a part of Job’s life as air itself. That the God of Job is just is not an issue for question. Job’s “friends” keep insisting that it must be the evil he has committed in the eyes of God that has brought down on him the terrible events and suffering. As Job answers with a description of a life of goodness and wisdom, he says at one point, “Yet does not God himself see my ways and count my every step?” (Job 31:4). Later in the same chapter, he says that he would like to defend himself directly to God: “I would plead the whole record of my life and present that in court as my defense” (Job 31:37; New English Bible 1976). Here is true suffering and true self-​conflict. If God, who sees his every step, is just, why does he have to present his case in court before God? Sometimes the event that precipitates suffering puts in doubt the way life has been lived. (In some versions of the Book of Job, Job wishes to plead his case so that the punishing hand be lifted from him so that he can speak. This is a point to remember when talking to the suffering patient.) Self-​conflict is also present in the dying and in acute illness. There would not be suffering in the face of terrible pain if one simply did not resist the pain, but let it simply wash over one. People resist pain and fight against it; it takes supreme willpower to allow

The Nature of Suffering    223 any symptom to just be without fighting back. One wants to live, but there is self-​conflict when the only life offered is terrible—​for example, wanting to be alive but not wanting to be a burden. At a certain point in suffering or in severe illness, it isn’t life that matters. What matters is what matters, and that is always personal. Thus, suffering is always personal, individual, and marked by self-​conflict. All suffering involves a loss of or profound change in central purpose. In Viktor Frankl’s well-​known book, Man’s Search for Meaning, the word “meaning,” as in finding meaning in life, is equivalent to the word “purpose” (Frankl 1959)—​so that it can be accurately paraphrased as man’s search for purpose. When I use the word “meaning,” I use it in the more traditional sense. The meaning of something is what is intended, its sense, significance, importance, or interpretation. For “purpose,” I mean intentions, goals, the objects for which things are done. Purpose pervades all. Every moment, every movement, every thought, and every action are all expressions of purpose. A hierarchy of purposes joins together to allow us to achieve our ends. The culmination of all purpose is central purpose—​the purpose of being “me.” A cartoon by Larson shows an island packed with penguins; penguins are everywhere. Out of this mass of penguins, a balloon arises from one that says, “I wanna be me!” Purpose is fundamentally not only individual but social. It draws from within the individual, but it requires the world for fulfillment. What I do, want, act on, and make is ultimately both social and individual and lives in time, in rhythm with the world; my purposes create me and create my world. Loss of purpose (usually not perceived as such) feels like the breakdown of the self. The focus of the person becomes entirely directed toward the sickness and the sick part. All purpose becomes directed at the relief of pain, sickness, and suffering. The more total and compelling the injury, the more complete the redirection of purpose. The loss of purpose, related to the loss of hope, is a profound loss. In the suffering person, purpose disengages from the social world, from common spirit, from “we-​ness.” All suffering is lonely. The suffering person is lonely for three reasons: the individuality of the suffering, the withdrawal of purpose from social engagement, and the fact that the person suffering is not understood (just as pain is often not understood). Thus, suffering is always personal, individual, marked by self-​conflict, and lonely. For all these reasons, suffering is also a state of social deprivation and isolation—​even in the midst of others. Suffering is a spiritual lesion. As with all treatment, it is important to choose therapeutic goals carefully. In suffering patients, these things are present: there is a stimulus, a cause, a reason; there is a person who is susceptible to suffering because of the person he or she is; there is something about this person that makes him or her liable to suffer given a source; there is self-​conflict, which, given an appropriate source, may precipitates suffering; there is loss of purpose; there is loneliness and social and spiritual isolation. (Understand “spiritual” in this context as meaning those things that transcend individual persons, the connections between persons, not necessarily religious). First and foremost and above all, symptoms must be relieved. That is not the end of treatment; it is only the beginning because symptoms can be completely controlled and suffering may continue. It is also true that suffering can be relieved even if its source and stimulus continue. The

224   Eric J. Cassell fundamental lesion of suffering that can be relieved is the loneliness and the social and spiritual deprivation. This is accomplished by the clinician and his or her connection to the patient. Clinicians are able to accomplish the relief of suffering because they are trained and accepted by patients as a source of their care and the relief of their symptoms and suffering. The social role of the clinician is the basis of that power and authority. “Power” is a word that makes medical people uncomfortable; it seems to lead to social incorrectness. Professionals often confuse the power and authority of their profession with their personal power, belonging to them as “the mighty person they are” instead of deriving from their role, training, and responsibility. I stress this point because when clinicians understand these things and accept them, they increase their opportunity to do everything that can actually be done for patients through them as persons. Let us go back to the patients I presented previously so that you can see the magnitude of the problem represented by their suffering and you can also see the opportunities for therapeutic intervention. The patient in pain has been reduced to a pitiable whimpering object. Understand that the memory of that event will never go away by itself unless its meaning is changed. If the clinician has the opportunity to be at the side of the gurney while what is described in the vignette is happening, the extreme nature of the pain has to be acknowledged—​as must the fear of obliteration and the fear of whatever disease is believed to be causing the pain. Even after the event, the same issues can be addressed. Instead of being weak and pitiable, she must be changed to strong because she was able to withstand the pain as long as she did. She can’t know that. You will have to say it to her and that you know it because you have seen many awful episodes like hers and few patients would have done as well. Remember, if you cannot change the event, change its meaning. Her fear of the disease that must cause such terrible pain must be addressed. The probability is metastatic disease to the bone, but she must know that if that is true, it can be relieved and effectively treated. I use this example first in part to suggest how important it is to change the meaning of what happened to that woman, even though you may be stretching the truth somewhat. In each of these episodes, medical personnel may not have behaved as well as one would wish. Do not defend their actions. Take care of the patient first and above all. The good and special person on the gurney feels herself humiliated and abused. Once again, the memory of the event will never go away unless its meaning is changed. In truth, she was abused in terms of who she is, even though the behavior of the staff may not have been particularly directed at her. She should be acknowledged as being special, and, without question, she will provide an opportunity to make that statement. She is special (and so is every other patient). Many who work in hospitals, especially in emergency departments, have become so accustomed to seeing patients on gurneys in the hallway hour after hour after hour waiting for a bed that we no longer want to think of it as abuse. The fact that we can do nothing about it and understand why it happens does not change the event for this particular person. Again, no excuses should be made; only the person’s injury need be acknowledged.

The Nature of Suffering    225 The vain and fastidious patient with feces running down her leg is a scene that is awful to even think about. Do you think she will ever forget or not be humiliated by that memory unless you directly address what it means to her? Simple acknowledgment of an injury begins its repair. In this case, there will be a greater need for further conversation before you really understand what it signified for her and before you know what must be done to relieve her suffering. In this instance, alcoholism played a part in her illness and that increases the difficulty posed by her suffering. It goes without saying that her diarrhea should be stopped. The danger of megacolon is probably considerably less than the danger of the suffering. The ultimate requirement in all these cases is understanding that suffering is an injury as real as a laceration—​but it is a profound injury to the person, not to a part of the person. Finally, the woman with the metastatic carcinoma of the colon should be helped to understand that her feeling that she is abandoned and will die alone is not necessary. It comes not from a defect in her, but from her loving desire to protect others. In fact, with her physician present, she was able to tell her daughters and to discover that, contrary to her fears, she and her daughters and grandchildren all became closer and more loving. She said before she died that her last year was the best year of her life. Most people, including doctors, find such statements unbelievable, but those who care for the dying not infrequently hear the same words from patients whose medical care has allowed them to repair important relationships. My first article about suffering was published in 1982 in the New England Journal of Medicine (Cassel 1982). My book, The Nature of Suffering, was published in 1991(Cassell 1991). By now, suffering is written about and discussed quite widely, primarily but not exclusively in the world of hospice and palliative medicine. Yet, in a recent issue of the New England Journal of Medicine, a Perspective piece entitled, “The Word That Shall Not Be Spoken,” where the word is suffering, was published (Lee 2013). This would be even stranger than it may seem except that the idea of suffering as a defined medical concept is not known throughout medicine as would, I believe, an important new finding about, for instance, cancer. I think that the reason for this is the fact that it is the person that suffers, and medicine has difficulty coming to terms with the idea of the person as a subject of medicine. This is the same problem that besets the concept of “person-​centered medicine.” We end up with definitions of that term as either fundamental politeness—​doctors should attend to the needs, concerns, desires, and wants of the patient; or political—​ doctors should make sure that the patient is treated as a person and as an autonomous being, and so forth. In fact, the nature of the person has an impact on the occurrence, onset, diagnosis, treatment, and outcome of the his or her sickness or disease. In light of this, patient-​centered medicine should focus on the impact of the patient on the disease as well as the impact of the disease on the patient. Medicine changes slowly. It will be decades, I believe, before the concepts that are so important in the care of the dying and the very sick (and probably every other patient) are widely accepted and acted upon. Meanwhile, physicians and other clinicians have the opportunity and responsibility to have a profound impact on the suffering of sick persons.

226   Eric J. Cassell

References Cassel, E. J. 1982. “The nature of suffering and the goals of medicine.” New England Journal of Medicine 306(11): 639–​645. Cassell, E. 1991. The nature of suffering. New York, Oxford University Press. Daneault, S., and D. Dion. 2004. “Suffering of gravely ill patients. An important area of intervention for family physicians.” Canadian Family Physician 50: 1348–​1350. Daneault, S., V. Lussier, et al. 2004. “The nature of suffering and its relief in the terminally ill: a qualitative study.” Journal of Palliative Care 20(1): 7–​11. Frankl, V. 1959. Man’s search for meaning. New York, Pocket Book. Lee, T. 2013. “The word that shall not be spoken.” New England Journal of Medicine 369(19): 1777–​1779. New English Bible. 1976. Edited by S. Sandmel. Oxford: Oxford University Press.

Chapter 14

On Our Diff i c u lt i e s Spe aking To a nd A b ou t the Dy i ng David Barnard

Some new, alien, extraordinary thing seemed to have made an appearance, a secret that they all read in each other’s eyes; the thought of death had been admitted into the house and now held silent sway in the spacious rooms.       (Thomas Mann, Buddenbrooks)

A Paradox of Modernity In the summer of 1570, Michel de Montaigne travelled to Paris from his home in the Bordeaux region of France. His purpose was to arrange for the publication of the literary works of his closest friend, who had died a few years before. Among these works was his friend’s translation of Plutarch’s “Letter of Consolation to His Wife,” which Plutarch wrote near the end of the first century ad when he learned of the death of his two-​year-​old daughter while he was away from home. The letter follows traditional literary forms from the ancient and classical world, incorporating themes of sympathy, consolation, and advice for carrying on with life. While Montaigne was in Paris, word reached him that his own first-​born child, a two-​month-​old daughter named Thoinette, had died. He responded to this news by writing a letter to his wife in which he urged her to read Plutarch’s letter, acknowledging how sad it was that fortune had made the ancient text so appropriate to her current situation. He then added, “But I leave to Plutarch the charge of consoling you and advising you of your duty in this, begging you to believe him for my sake. For he will reveal to you my views, and what can be said in this matter, much better than I could myself ” (Montaigne, 2003, pp. 1300–​1301).

228   David Barnard Many modern readers, coming upon Montaigne’s words four and half centuries later, are apt to consider them surprisingly artificial and detached. From the vantage point of the modern era—​with its commitment to sincerity, authenticity, and individuality—​for Montaigne to rely on an ancient author to convey his feelings of sadness and consolation to his wife almost seems like emotional laziness, or even an evasion of personal responsibility. If Montaigne genuinely cared about his wife’s feelings, surely he could, and should, have said something to her in his own words.1 Do we moderns really fare any better when we are on our own, however? Crossing the centuries since Montaigne’s letter, we come upon the experience of C.S. Lewis after the death of his wife, chronicled in Lewis’s memoir, A Grief Observed, originally published in 1963. “I cannot talk to the children about her,” Lewis (1976) wrote. “The moment I try, there appears on their faces neither grief, nor love, nor fear, nor pity, but the most fatal of all non-​conductors, embarrassment. They look as if I were committing an indecency. They are longing for me to stop” (pp. 8–​9). “It isn’t only the boys either,” Lewis continues. “An odd byproduct of my loss is that I’m aware of being an embarrassment to everyone I meet. At work, at the club, in the street, I see people, as they approach me, trying to make up their minds whether they’ll ‘say something about it or not.’ I hate it if they do, and if they don’t… . I like best the well-​ brought-​up young men, almost boys, who walk up to me as if I were a dentist, turn very red, get it over, and then edge away to the bar as quickly as they decently can. Perhaps the bereaved ought to be isolated in special settlements like lepers” (pp. 10–​11). What Lewis describes as the profound and painful embarrassment that his bereavement engendered in the people around him (painful to him as well as to them) points to a paradox of modernity. Celebrating our individuality, personal sincerity, and authenticity and dismissing many of the traditional forms of meaning making and expression as relics of an oppressive, unenlightened past, many people in modern society find themselves feeling helpless, awkward, and tongue-​tied when actually called upon to express feelings of sadness, sympathy, or comfort to the dying or the bereaved (Mellor and Shilling, 1993). Despite approximately 50 years of effort by the modern hospice and palliative care movements (dating the movements loosely from the establishment by Dame Cicely Saunders of St. Christopher’s Hospice in London in 1967 and the publication of On Death and Dying by Elisabeth Kübler-​Ross in 1969) and the claims of palliative care’s proponents that, in the optimistic formulation of the Open Society Institute’s Project on Death in America (2004), we are “transforming the culture of dying,” there is ample evidence that the dynamic of awkwardness, embarrassment, and withdrawal that Lewis experienced continues to affect relationships and communication with dying people. Almost exactly 50 years after Lewis wrote A Grief Observed, his British countryman Julian Barnes (2013) published a memoir of the death of his wife that includes this striking echo of Lewis: I remember a dinner conversation in a restaurant with three married friends… . Each had known [my wife] for many years… . I mentioned her name; no one picked it up. I did it again, and again nothing. Perhaps the third time I was deliberately trying

Speaking to and About the Dying    229 to provoke… . Afraid to touch her name, they denied her thrice, and I thought the worse of them for it. “Your grief [Barnes imagines the friends thinking] is an embarrassment. We’re just waiting for it to pass.” (pp. 79–​80, 84)

That awkwardness, embarrassment, and withdrawal in the presence of the dying and bereaved remain widespread among the general public is demonstrated in part by the proliferation of advice books and how-​to manuals designed to help people through these difficult social encounters. One recent example, How to Be a Friend to a Friend Who’s Sick, by Letty Cottin Pogrebin, captures precisely the paradox of modernity wherein we feel we ought to be able to say something personal and original but end up getting stuck: “You don’t want to sound like a Hallmark card,” Pogrebin (2013) writes, “but anxiety thickens your tongue. You want to say something real, but the words that come to mind add insult to your friend’s injuries” (p. 16). Robert Buckman, an oncologist and communications skills expert who contributed greatly to the professional literature on communication with the dying, also wrote an advice book for the general public that had been reprinted 16 times within 12 years of its original publication in 1988. Its title neatly captures the problem: I Don’t Know What to Say (Buckman, 1988). Advice books and how-​to manuals have proliferated for health professionals as well, whether they are general surveys with chapters devoted to communication in end-​of-​life care (Smith, 2002; Platt and Gordon, 2004) or more specialized treatments (Buckman, 1992; Faulkner and Maguire, 1995; Back, Arnold, and Tulsky, 2009). Evidence that they are needed includes data showing patients’ and families’ dissatisfaction with physicians’ communication of information about serious illness and their attention to emotions (Fine, Reid, Shengelia, and Adelman, 2010; Pardon, Deschepper, Stichele, et al., 2011; Schenker, Crowley-​Matoka, Dohan, et al., 2012); physicians’ reported reluctance to have serious end-​of-​life or advance care planning conversations with their patients (Cherlin, Fried, Prigerson, et al., 2005; Keating, Landrum, Rogers, et al., 2010; Mack and Smith, 2012); and evidence of widespread failure to discuss or document the care preferences of hospitalized older adults even when they had clear opinions on the subject and wished to discuss them (Heyland, Barwich, Pichora, et al., 2013). These deficiencies in communication show up in patients’ experiences of intensity of care near the end of life and in their pathways through the health-​care system as they die. Teno, Gozalo, Bynum, and colleagues (2013) report, for example, that between 2000 and 2009, despite an increase in hospice utilization and patient deaths at home, the rate of utilization of the intensive care unit in the last months of life, the number of decedents with three or more hospitalizations in the last 90 days of life, and the number of transitions in care settings in the last three days of life also increased. Even among hospice admissions, 28.4% were for stays of three days or less, strongly suggesting that for many people “hospice is an “add-​on” to a growing pattern of more utilization of intensive services at the end of life” (p. 475). A similar pattern of increasing intensity of care in very advanced illness appears in cancer treatment, with a continuing increase in the proportion of patients receiving chemotherapy in the last 14 days of life, with no increase in the effectiveness of the treatments being used (Earle, Landrum, Souza, et al, 2008).

230   David Barnard As both Teno and colleagues (2013) and Earle and colleagues (2008) acknowledge, it is impossible to determine from their data alone how much aggressive treatment in what turns out to be the last days or weeks of life may reflect patients’ preferences. And even when choices for aggressive treatment that patients or their families come to regret are the result of inadequate communication, patients often collude with the doctor in avoidance of hard discussions of the future, “seem[ing] to accept gratefully every opportunity offered by doctors to ‘forget’ the future and to focus on the present, which is full of action (treatments, tests, etc.). The recovery story is the dominant social discourse, and, in general, it is difficult for patients to deviate from it” (The, Hak, Koёter, and van der Wal, 2000, p. 1380). Nevertheless, as Earle and colleagues (2008) point out in the case of oncologists, physicians have broader experience and perspective on the likely benefit of aggressive treatments in late-​stage illness than their patients can possibly have and the responsibility to use their experience to counsel patients and families that it might be time to stop. The problems brought on by our difficulties speaking to and about the dying are thus two-​fold. One relates to utilization of services and the possibility that large numbers of people receive more aggressive, less person-​and family-​centered care than they would have preferred. The other relates to the quality of people’s social interactions, including with health professionals, during the final phase of their illness when our discomfort and reticence can cut people off from the healing potential of kindness and intimate companionship. We return, then, to the paradox of modernity, captured poignantly by the German sociologist Norbert Elias (1985) in his aptly titled book, The Loneliness of the Dying: At present those close to the dying often lack the ability to give them support and comfort by proof of their affection and tenderness. They find it difficult to press dying people’s hands or to caress them, to give them a feeling of undiminished protection and belonging. Civilization’s overgrown taboo on the expression of strong, spontaneous feelings ties their tongues and hands. And living people may half unconsciously feel death to be something contagious and threatening; they involuntarily draw back from the dying. But, as with every parting of people who are intimate, a gesture of undiminished affection is, for the one taking final leave, perhaps the greatest help, apart from the relief of physical pain, that those left behind can give. (pp. 28–​29)

Explaining our Difficulties Hypotheses abound purporting to explain the silence or withdrawal with which so many people react in the presence of the dying and bereaved. Elias (1985) hinted at one in the passage quoted above: the primitive fear that death may be contagious—​best not to get too near! While some explanations refer to difficulties encountered by professionals as well as by people in the general population, a few focus on health professionals in

Speaking to and About the Dying    231 particular who, besides the inhibitions they share with the public at large, experience others related to their professional roles. For example, the dying patient may appear to the physician, especially, as a painful symbol of the latter’s helplessness or failure, or the physician may feel guilty for letting the patient down. To avoid these feelings, the hypothesis goes, the physician avoids the patient. Another common explanation reaches back to the tradition that the physician should be, in Thomas Percival’s phrase, the “minister of hope and comfort to the sick” (Percival, 2013, p. 391). At least since Hippocratic times, physicians have been exhorted to avoid “gloomy prognostications” (Percival, 2013, p. 391) so as to avoid depressing their patients and thereby possibly hastening their death. Despite considerable empirical evidence that evasions and false encouragement are more destructive of patients’ feelings of hopefulness than open communication and sharing of information at the level desired by the patient, an “ideology of hopefulness” (Delvecchio-​Good, Good, Schaffer, and Lind, 1990) still shapes, and restricts, many professional–​patient interactions near the end of life. And here we should speak not only of professionals. Many people seem to agree with the sentiment expressed by Simone de Beauvoir (1965), after describing her own sometimes cruel efforts to prevent her dying mother from expressing her fears of death to her daughters in the last weeks of her mother’s life: My unfair harshness wrung my heart. At the time the truth was crushing her and when she needed to escape from it by talking, we were condemning her to silence; we forced her to say nothing about her anxieties and to suppress her doubts: as it had so often happened in her life, she felt both guilty and misunderstood. But we had no choice: hope was her most urgent need. (pp. 65–​66)

If many health professionals feel constrained by role-​specific concerns or feelings when speaking to or about the dying, they may simultaneously be influenced by psychological processes that affect people in general. One is “survivor guilt,” a concept described by Robert Jay Lifton (1967) in his study of hibakusha—​civilian survivors of the atomic bombings of Hiroshima and Nagasaki—​but applied widely to other contexts such as natural disasters, warfare, or mass atrocities. In each case the survivor is left to wonder, “Why, when so many others have died, did I remain alive?” In many cases, the survivor may feel that he or she lived precisely because the others died. Long after the event, the survivor may be haunted by feelings of inauthenticity, illegitimacy, or fear of ultimate retribution for having lived while others died in his or her place. Such feelings may be lurking in the background of many encounters with the dying or bereaved, as the living give silent thanks that (this time and for now) it is the other who has fallen into the grip of death (Barnard, 1985). Being close to someone who will soon die may also rekindle childhood fears of abandonment, or provoke anger, distancing, or withdrawal as part of anticipatory mourning (Bowlby, 1975; Rando, 2000). It may also, according to the terror management theory of Sheldon Solomon, Jeff Greenberg, and Tom Pyszczynski (1991), provoke strong, nonconscious defenses against death anxiety (the fear of personal annihilation

232   David Barnard or extinction) that take the form either of suppression, distraction, or denial, or strenuous efforts to bolster one’s own self-​esteem and intensification of one’s identification with the dominant values of one’s own culture. In the context of health-​care professionals’ encounter with dying patients, Solomon and Lawlor (2011) suggest that the latter defense against death anxiety takes the form of committing all the more strongly to the medical culture’s predominant value of cure and prolongation of life. Finally, withdrawal and avoidance of the dying or bereaved may be natural protections against overwhelming emotions. Nancy Eisenberg’s research on “empathy-​related emotional responses” describes two divergent patterns of response to the perception of another person’s distress (Eisenberg, Wentzel, and Harris, 1998; Eisenberg, 2001). One is sympathy and the desire to take some action to give help or comfort. The other is personal distress, what Eisenberg (2001) describes as “a self-​focused, aversive emotional reaction to another’s emotion or condition [that is] experienced as discomfort, anxiety, or concern about one’s own welfare” (p. 135). Among the factors determining which response will predominate, Eisenberg’s research suggests, are neurophysiological mechanisms for regulation of emotional reactivity and the ability to reframe one’s perception of another’s distress cognitively (e.g., shifts of attention to positive aspects of the situation or reminders of the separation between self and other) in ways that moderate the effect of the negative events or cognitions. The inadequacy of these regulatory mechanisms can lead, Eisenberg and colleagues (1998) argue, to a state of “empathic overarousal” in which self-​focused actions to mitigate personal distress overshadow impulses of concern for the other person. Cynda Rushton, Alfred Kaszniak, and Joan Halifax (2013), who have applied Eisenberg’s work to the context of palliative care, characterize these self-​protective responses as “(1) avoidance and/​or abandonment of the patient and family physically, emotionally, or spiritually (flight); or (2) angry or contentious responses … aimed at mitigating the caregiver’s own suffering (fight); or (3) ‘freeze’ or ‘numbing,’ a generalized shutting down emotionally and being immobilized to respond or take appropriate action to address the situation” (p. 1076)—​very much the picture of our difficulties speaking to and about the dying.

An Alternative, Cultural Perspective: Death, Shame, and Embarrassment The explanations summarized so far construe our silence and withdrawal in the presence of the dying and bereaved primarily as a problem of personal psychology. Personal psychology and individual behavior, however, are embedded in a wider social, historical, and cultural context. Individual behavior, and individual emotions, are profoundly shaped by social and cultural norms that define what is acceptable and praiseworthy to

Speaking to and About the Dying    233 do, think, or feel. These norms are enforced through socialization processes that rely heavily on messages of approval, love and affection, and incorporation into the social group and by their opposites—​disapproval, threats to withhold love and affection, and ostracization. Among the most powerful means for enforcing social norms is the experience of shame, defined by Thomas Scheff (2003) as “the class name for a large family of emotions and feelings that arise through seeing the self negatively, even if only slightly negatively, through the eyes of others, or anticipating such a reaction. [This family of emotions] includes many cognates and variants, most notably embarrassment, guilt, humiliation, and related feelings such as shyness that originate in threats to the social bond” (pp. 254–​ 255). Elaborating on the power of shame as a means of social control, Scheff observes: First, shame is a key component of conscience, the moral sense:  it signals moral transgression even without thoughts or words. Shame is our moral gyroscope. Second, shame arises in elemental situations of threat to a bond; it signals trouble in a relationship. Since an infant’s life is completely dependent on the bond with the caregivers, shame is as primitive and intense as fear. This idea subsumes the more usual one that shame arises when one feels one has failed to live up to one’s standards, since these standards are, for the most part, held in common with the significant others in one’s life. The sense that one has failed to live up to one’s standards would usually also signal a threat to one’s social relationships. Third, shame plays a central role in regulating the expression, and indeed the awareness of all our other emotions. We are not likely to express anger, fear, grief, and love, for example, outwardly to the degree that we will be ashamed of them. One can be so ashamed of one’s emotions that they can be repressed completely. (p. 254)

In other words, what in a particular social, historical, and cultural context is defined as outside the boundaries of acceptable sights, thoughts, feelings, or actions will induce strong feelings of shame or embarrassment, or—​what leads to the same result—​the anticipation of such feelings and the desire to avoid them will lead us to avoid unacceptable sights, thoughts, feelings, or actions or try to prevent their enactment in a particular social situation. The more the sight, thought, feeling, or action threatens to cross a socially constructed boundary, the more intense or extreme the feeling of embarrassment or shame. Erving Goffman (1967) provides a behavioral description of the experience of extreme embarrassment that bears striking resemblance to the painfully awkward, tongue-​tied reaction of many people in the presence of the dying and bereaved that is the focus of this discussion: An individual may recognize extreme embarrassment in others and even in himself by the objective signs of emotional disturbance: blushing, fumbling, stuttering, an unusually low-​or high-​pitched voice, quavering speech or breaking of the voice, sweating, blanching, tremor of the hand, hesitating or vacillating movement, absentmindedness, and malapropisms. As Mark Baldwin remarked about shyness, there

234   David Barnard may be “a lowering of the eyes, bowing of the head, putting of hands behind the back, nervous fingering of the clothing or twisting of the fingers together, stammering, with some incoherence of idea as expressed in speech.” There are also symptoms of a subjective kind: constriction of the diaphragm, a feeling of wobbliness, consciousness of strained or unnatural gestures, a dazed sensation, dryness of the mouth, and tenseness of the muscles. In cases of mild discomfiture, these visible and invisible flusterings occur but in less perceptible form. (p. 97)

The association of shame with death and dying centers on what Annette Street and David Kissane (2001) refer to as the “the shameful body discourse”: “The idea of the shameful body,” they write, “relates to the person’s response to the breakdown of bodily integrity and the subsequent feelings of humiliation and embarrassment that arise from an incapacity to manage the body in a socially acceptable manner… . The unwanted identity accompanying deterioration in physical functioning and appearance in terminal illness may engender feelings of shame and humiliation, stigma, disgust, and loathing. The psycho-​visceral response to the dying body-​self can be disgust that it can no longer be managed as a social body; it has become a shameful, abject body that prefigures the post-​mortem body” (pp. 166–​167). Julia Lawton (1998) argues that feelings of embarrassment and even disgust at the sights, smells, and emissions of the dying and disintegrating body are among the significant motivations that relatives of dying patients have for urging admission to a hospice. Hospice workers and other professionals, despite their training and aspirations, are not themselves immune from the need to constantly repress reactions of revulsion and disgust when intimately engaged with “the unbounded body” (Mansfield, 2000), whose boundaries “are constantly broken by flows of urine, tears, faeces [sic], vomit, blood, sweat, and semen… . We defend against this pollution and rupturing of subjective boundaries by shame and loathing” (Street and Kissane, 2001, p. 167). The paradox of modernity—​the tendency of many people in modern society to feel awkward, tongue-​tied, and tempted to flee the presence of the dying and bereaved despite feeling compelled to say something personal, authentic, and individualized—​ now comes more sharply into focus. For the social and cultural construction of the physical body as embarrassing, and shameful is part of the historical “civilizing process” (Elias, 1994) that has produced modernity itself. Included by extension within this shameful orbit, I would argue, are dying and bereaved people whose presence reminds us that we are all “ensconced in a fleshly carcass that eats and excretes and breeds and bleeds and breaks” (Solomon and Lawlor, 2011, p. 101). Elias traces a process, originating with the upper classes in the late Middle Ages and Renaissance, whereby the display of the naked body, and a whole range of behaviors associated with the body—​belching, farting, urinating and defecating, sexual activity, even eating with the fingers or picking snot from one’s nose—​came to be defined as outside the bounds of what is acceptable in “polite society.” As a rising middle class aspired to the respect and status associated with the nobility, the upper classes’ definition of acceptable public behavior took hold more widely. A seminal text in this “history

Speaking to and About the Dying    235 of manners” was written in 1513 by Erasmus of Rotterdam. On Civility in Children was reprinted 30 times within six years of publication and, according to Elias, went through more than 130 editions by the eighteenth century. Here is a representative passage: A peasant wipes his nose on his cap and coat, a sausage maker on his arm and elbow. It does not show much propriety to use one’s hand and then wipe it on one’s clothing. It is more decent to take up the snot in a cloth, preferably while turning away. If when blowing the nose with two fingers something falls to the ground, it must be immediately trodden away with the foot. The same applies to spittle. (Nathanson, 1992, p. 439)

What Elias, a sociologist, describes as “the civilizing process,” anthropologist Mary Douglas (1973) describes in terms of a correlation between increasing demands for social conformity in society with control of bodily function, or, in her terms, “the rule of distance from physiological origin,” according to which “the more the social situation exerts pressure on persons involved in it, the more the social demand for conformity tends to be expressed by a demand for physical control. Bodily processes are more ignored and more firmly set outside the social discourse, the more the latter is important. A natural way of investing a social occasion with dignity is to hide organic processes” (p. 12). These developments culminate, according to Douglas, in a clear division between two bodies—​the social body and the physical body. The social body is the arena in which members of society interact with each other according to rules. The physical body and its requirements are, according to the rule of distance from physiological origins and what Douglas calls “the purity rule,” contrasted with and subordinated to the social body and its requirements. Here Douglas’s and Elias’s analyses converge, in that within increasingly complex and socially differentiated societies, social status and acceptable interpersonal behavior depend on a progressive disembodiment of the actors, as a complex social system devises ways for itself of behaving that suggest that human intercourse is disembodied compared with that of animal creation. It uses different degrees of disembodiment to express the social hierarchy. The more refinement, the less smacking of lips when eating, the less mastication, the less the sound of breathing and walking, the more carefully modulated the laughter, the more controlled the signs of anger, the clearer the priestly-​aristocratic image. (Douglas, 1973, pp. 101–​102)

As noted above, and crucially for this discussion, the primary enforcer of cultural and social norms around the body—​its raw physicality, its insistent demands, its weakness, and its ultimate capacity to undermine and terminate our noblest aspirations—​is shame, whose very root and etymology “is intimately linked to the human need to cover that which is exposed” (Nathanson, 1987, p. 199, italics in original).2 Shame results from looking at what must not be seen (Sedgwick and Frank, 1995, pp. 144–​148), and in this

236   David Barnard context what must not be seen are direct evidence or strong reminders of ourselves as mortal bodies—​“desiring, hungry, diseased, in pain… . In both death and life, the body’s shame is its naked vulnerability, its materiality. Like the grass, we wither and die; but the soul lives on, we say, defying death. The shame of death is in the gap between our immortal aspirations and our mortal imperfections” (Clark, 2010, p. 152). The averted gaze, flushing, and constricted throat, the urge to hide, to be anywhere but in the immediate presence of the dying or bereaved person in front of us, are—​however helpfully they may be understood in terms of personal psychology—​shame responses that reflect centuries of social and cultural history.

Conclusion Some people who shunned C.S. Lewis in his bereavement viewed him, Lewis (1976) wrote, as “a death’s head” (p. 11). Lewis experienced the power of the very idea of death to induce fumbling, evasive, awkward silence in anyone who came in contact with him. What is required to diminish this power, and to spread more widely among us in modern society the healing potential of kindness and intimate companionship, those “gesture[s]‌of undiminished affection [that], for the one taking final leave, [are] perhaps the greatest help, apart from the relief of physical pain, that those left behind can give” (Elias, 1985, p. 29)? From the vantage point of those perspectives that interpret awkward silence or withdrawal before the dying and bereaved as problems of personal psychology and individual behavior, the remedies follow logically the directions they have in fact followed, particularly in the training of health professionals: exercises in self-​awareness (Novack, Suchman, Clark, et al., 1997; Kearney, Weininger, Vachon, et al., 2009); recognition of one’s emotional responses to dying patients and “working through” one’s emotions in support groups or consultation with colleagues (Meier, Back, and Morrison, 2001); mastery of communications skills techniques with mnemonics such as SPIKES (Baile, Buckman, Lenzi, et al., 2000) or NURSE (Arnold, Nelson, Prendergast, et al., 2010); and recommendations for specific words and phrases for clinicians to use that, in the words introducing a new series in the journal Annals of Internal Medicine, “have proven particularly powerful as tools for understanding patients and helping them manage their illnesses effectively” (Platt, Gaspar, Coulehan, et al., 2001). Personal psychology is certainly part of the picture, and the strategies proposed to assist people with their feelings and words related to the dying and bereaved have had some beneficial results—​though, as suggested in the introductory section above, not to the extent hoped and sometimes claimed for them by their proponents. From the perspective of social and cultural history, however, these techniques and behaviors resemble wind-​driven ripples on the surface of the ocean, attractive but superficial movements with little effect on the depths below. Change at this deeper level would be cultural change, by which the thought of death, and its reminders of embodiment, finitude, and our inevitable physical decay, become acceptable subjects for personal reflection and

Speaking to and About the Dying    237 social engagement. When we speak of this type of cultural change, we are speaking of changes in the most basic and powerful symbols and narratives by which people order and value their perceptions and experiences of the world. The role of these symbols and narratives in shaping and supporting worldviews, though the symbols and narratives need not be explicitly religious in themselves, is well captured by the definition of “religion as a cultural system” proposed by the anthropologist Clifford Geertz (1993), according to which a religion is “(1) a system of symbols which acts to (2) establish powerful, pervasive, and long-​lasting moods and motivations in men by (3) formulating conceptions of a general order of existence and (4) clothing these conceptions with such an aura of factuality that (5) the moods and motivations seem uniquely realistic” (p. 90). Cultural change of this nature can only be the product of social movements much broader and more profound than the adoption of new communications skills protocols by health professionals or advice manuals for people seeking to comfort their bereaved friends and relatives. There are at present the glimmerings of social movements that aspire to bring death and loss into the mainstream of socially acceptable contemplation and conversation. Examples include local initiatives such as Death Café (www.deathcafe.com) where “people, often strangers, gather to eat cake, drink tea and discuss death”; Death Over Dinner (www.deathoverdinner.org) that “invites you to gather friends and family and fill a table” to discuss how you want to die; and The Conversation Project (www.theconversationproject.org) that encourages people to express their wishes for end-​of-​life care and believes “the place for this to begin is at the kitchen table.” There are also recent Internet and wireless-​enabled phenomena such as the very public tweeting by National Public Radio’s Scott Simon about his mother’s dying days (Pearce, 2013) and daily blogs by dying people themselves, such as Katie Scarborough in England (www.katiescarbrough.co.uk), whose daily reports about her experience of dying from bowel cancer—​which she intended primarily as a record for her children—​attracted 72,000 followers. The mass media are also beginning to play a role, whether through moving stories of how famous people like Jacqueline Kennedy Onassis or film critic Roger Ebert faced death (Schneider, Reed, Levitt, and Gleick, 1994; Chi and Shira, 2013), or television series such as Showtime’s “Time of Death,” which aims to answer, for a mass audience, the question “What are the final weeks, days, and very moments of life really like?” (www.sho.com/​sho/​time-​of-​death/​home). These projects and happenings might at first appear to be counterexamples that challenge the hypothesis of this chapter that culturally pervasive shame and embarrassment in the face of death are at the root of people’s difficulties in speaking to or about the dying. To be sure, they are part of something that might augur a new openness in the culture to public, social exchange about death and dying. But are they counterexamples to the hypothesis? That is much less clear. For one thing, most of these initiatives are extremely young. Death Café, Death Over Dinner, and The Conversation Project, for example, all came into being between 2010 and 2012. It is also hard to know how far the Internet-​based productions have moved beyond a relatively narrow, media-​savvy, and wired segment of society, or—​more important—​whether people’s interest or even fascination with the details of the dying processes of distant others translates into greater

238   David Barnard comfort with the dying of people close to them. Time will tell whether these and other movements are helping to establish “powerful, pervasive, and long-​lasting moods and motivations” that provide broad cultural support for open, welcoming relationships, and communications with the dying and bereaved. Just such a cultural shift will be required for widespread movement beyond tongue-​ tied awkwardness. This cultural shift might well entail a recovery of the attitude toward physical embodiment extolled by Michel de Montaigne (2003), for whom “the most barbarous of our maladies is to despise our being” (p. 1039). That being, Montaigne insisted, is thoroughly embodied, frail, and transient, and his acceptance of our embodied state—​ his comfort within his own skin and with all that his skin contained—​is among the most salient leitmotivs of the Essays. As he wrote in his final pages, I, who operate close to the ground, hate that inhuman wisdom that would make us disdainful enemies of the cultivation of the body. I, who boast of embracing the pleasures of life so assiduously and so particularly, find in them, when I look at them thus minutely, virtually nothing but wind. But what of it? We are all wind. And even the wind, more wisely than we, loves to make a noise and move about, and is content with its own functions, without wishing for stability and solidity, qualities that do not belong to it… . It is an absolute perfection and virtually divine to know how to enjoy our being rightfully. We seek other conditions because we do not understand the use of our own, and go outside of ourselves because we do not know what it is like inside. Yet there is no use our mounting on stilts, for on stilts we must still walk on our own legs. And on the loftiest throne in the world we are still sitting on our own rump. (pp. 1035, 1044)

Notes 1. Montaigne’s reliance on the words of others to express his thoughts was habitual, as he explains in the essay “On Books”: “I make others say what I cannot say so well, now through the weakness of my language, now through the weakness of my understanding” (2003, p. 359). 2. “The common element is apparent in the Indo-​European root *(s)kem-​; *(s)kam-​, meaning, ‘to cover,’ which gives us not only our English word shame, but the Italian word camera (a little room and therefore something hidden) which becomes the English camera (a little room which reveals); the French chemise, and the German Hemd, all involving covering or protection” (Nathanson, 1987, p. 109).

References Arnold, R.M., Nelson, J., Prendergast, T., et al. (2010). Educational Modules for the Critical Care Communication (C3) Course—​A Communication Skills Training Program for Intensive Care Fellows. Available at: http://​www.capc.org/​palliative-​care-​professional-​development/​ Training/​c3-​module-​ipal-​icu.pdf. Accessed February 17, 2014.

Speaking to and About the Dying    239 Back, A., Arnold R., and Tulsky, J. (2009). Mastering Communication with Seriously Ill Patients:  Balancing Honesty with Empathy and Hope (New  York:  Cambridge University Press). Baile, W.F., Buckman, R, Lenzi, R., et al. (2000). “SPIKES—​A Six-​Step Protocol for Delivering Bad News: Application to the Patient with Cancer.” Oncologist 5(4): 302–​311. Barnard, D. (1985). “Survivorship and Medical Ethics.” Death Studies 9: 115–​131. Barnes, J. (2013). Levels of Life. (New York: Knopf). Bowlby, J. (1975). “Attachment Theory, Separation Anxiety, and Mourning.” In American Handbook of Psychiatry, edited by S. Arieti. 2d ed., Vol. 6, 292–​309 (New York: Basic Books). Buckman, R. (1988). I Don’t Know What to Say (Toronto: Key Porter Books). Buckman, R. (1992). How to Break Bad News:  A  Guide for Health Care Professionals (Baltimore: Johns Hopkins University Press). Cherlin, E., Fried, T., Prigerson, H.G., et al. (2005). “Communication Between Physicians and Family Caregivers about Care at the End of Life: When Do Discussions Occur and What is Said?” Journal of Palliative Medicine 8(6): 1176–​1185. Chi, P., and Shira, D. (2013). “Chaz Ebert: Roger Smiled Right Before He Died.” People, April 4.  Available at http://​www.people.com/​people/​article/​0,,20688468,00.html. Accessed February 17, 2014. Clark, H. (2010). “Mask of Shame, Mask of Death:  Some Speculations on the Shame of Death.” In The Shame of Death, Grief, and Trauma, edited by J. Kauffman, 141–​153. (New York: Routledge). de Beauvoir, S. (1965). A Very Easy Death. Translated by P. O’Brian (New York: Pantheon). Delvecchio-​Good, M.G., Good, B.J., Schaffer, C., and Lind, S.E. (1990). “American Oncology and the Discourse on Hope.” Culture, Medicine, and Psychiatry 14: 59–​79. de Montaigne, M. (2003). The Complete Works. Translated by D.M. Frame (New York: Knopf). Douglas, M. (1973). Natural Symbols: Explorations in Cosmology (New York: Vintage). Earle, C.C., Landrum, M.B., Souza, J.M., et  al. (2008). “Aggressiveness of Cancer Care Near the End of Life:  Is It a Quality of Care Issue?” Journal of Clinical Oncology 26(23): 3860–​3866. Eisenberg, N. (2001). “Empathy-​ Related Emotional Responses, Altruism, and Their Socialization.” In Visions of Compassion: Western Scientists and Tibetan Buddhists Examine Human Nature, edited by R.J. Davidson and Anne Harrington, 131–​164 (New York: Oxford University Press). Eisenberg, N., Wentzel, N.M., and Harris, J.D. (1998). “The Role of Emotionality and Regulation in Empathy-​Related Responding.” The School Psychology Review 27(4): 506–​521. Elias, N. (1985). The Loneliness of the Dying (Oxford: Basil Blackwell). Elias, N. (1994). The Civilizing Process (Oxford: Blackwell). Faulkner, A., and Maguire, P. (1995). Talking to Cancer Patients and Their Relatives (New York: Oxford University Press). Fine, E., Reid, M.C., Shengelia, R., and Adelman, R.D. (2010). “Directly Observed Patient–​ Physician Discussions in Palliative and End-​of-​Life Care:  A  Systematic Review of the Literature.” Journal of Palliative Medicine 13(5): 595–​603. Geertz, C. (1993). “Religion as a Cultural System.” In The Interpretation of Cultures: Selected Essays, 87–​125 (Waukegan, IL: Fontana Press). Goffman, E. (1967). Interaction Ritual (New York: Anchor). Heyland, D., Barwich, D., Pichora D., et al. (2013). “Failure to Engage Hospitalized Elderly Patients and Their Families in Advance Care Planning.” JAMA Internal Medicine 173(9): 778–​787.

240   David Barnard Kearney, M., Weininger, R., Vachon, M.L.S., et al. (2009). “Self-​Care of Physicians Caring for Patients at the End of Life.” JAMA 301(11): 1155–​1164. Keating, N.L., Landrum, M.B., Rogers, S.O., et al. (2010). “Physician Factors Associated with Discussions About End-​of-​Life Care.” Cancer 116(4): 998–​1006. Lawton, J. (1998). “Contemporary Hospice Care: The Sequestration of the Unbounded Body and ‘Dirty Dying.’” Sociology of Heath & Illness 20(2): 121–​143. Lewis, C.S. (1976). A Grief Observed (New York: Bantam). Lifton, R.J. (1967). Death in Life: Survivors of Hiroshima (New York: Random House). Mack, J., and Smith, TJ. (2012). “Reasons Why Physicians Do Not Have Discussions About Poor Prognosis, Why It Matters, and What Can Be Improved.” Journal of Clinical Oncology 30(22): 2715–​2717. Mansfield, N. (2000). Subjectivity:  Theories of the Self from Freud to Haraway (New  York:  New York University Press). Meier, D.E., Back, A.L., and Morrison, R.S. (2001). “The Inner Life of Physicians and the Care of the Seriously Ill.” JAMA 286(23): 3007–​3014. Mellor, P.A., and Shilling, C. (1993). “Modernity, Self-​Identity, and the Sequestration of Death.” Sociology 27(3): 411–​431. Nathanson, D.L. (1987). The Many Faces of Shame (New York: Guilford Press). Nathanson, D.L. (1992). Shame and Pride: Affect, Sex, and the Birth of the Self (New York: W.W. Norton). Novack, D.H., Suchman, A.L., Clark, W., et  al. (1997). “Calibrating the Physician:  Personal Awareness and Effective Patient Care.” JAMA 278(6): 502–​509. Pardon, K., Descheper, R., Stichele, R.V., et al. (2011). “Are Patients’ Preferences for Information in Medical Decision-​Making Being Met? Interview Study with Lung Cancer Patients.” Palliative Medicine 25(1): 62–​70. Pearce, M. (2013). “NPR Host Scott Simon Tweets His Mother’s Dying Days.” Los Angeles Times, July 30. Available at http://​www.latimes.com/​news/​nationworld/​nation/​la-​na-​0730-​ scott-​simon-​mom-​20130730-​dto,0,5644002.htmlstory#ixzz2tcSOhHSF. Accessed February 14, 2014. Percival, T. (2013). Medical Ethics. In The Works, Literary, Moral, and Medical, of Thomas Percival, M.D. To Which Are Prefixed, Memoirs of his Life and Writings, and a Selection from his Literary Correspondence. Vol. 2. (Cambridge, UK: Cambridge University Press). (Originally published 1803) Platt, F.W., Gaspar, D.L., Coulehan, J.L., et al. (2001). “‘Tell Me about Yourself ’: The Patient-​ Centered Interview.” Annals of Internal Medicine 134(11): 1079–​1085. Platt, F.W., and Gordon, G.H. (2004). Field Guide to the Difficult Patient Interview. 2d ed. (Philadelphia: Lippincott Williams & Wilkins). Pogrebin, L.C. (2013). How to Be a Friend to a Friend Who’s Sick (New York: Public Affairs). Project on Death in America. (2004). Transforming the Culture of Dying: The Project on Death in America October 1994 to December 2003 (New York: Open Society Institute). Available at http://​www.opensocietyfoundations.org/​sites/​default/​files/​a_​transforming.pdf. Accessed February 17, 2014. Rando, T., ed. (2000). Clinical Dimensions of Anticipatory Mourning: Theory and Practice in Working with the Dying, their Loved Ones, and their Caregivers (Champaign, IL: Research Press). Rushton, C., Kaszniak, A.W., and Halifax, J. (2013). “A Framework for Understanding Moral Distress among Palliative Care Clinicians.” Journal of Palliative Medicine 16(9): 1074–​1079.

Speaking to and About the Dying    241 Scheff, T. (2003). “Shame in Self and Society.” Symbolic Interaction 26(2): 239–​262. Schenker, Y., Crowley-​Matoka, M., Dohan, D., et al. (2012). “I Don’t Want to Be the One Saying ‘We Should Just Let Him Die’: Intrapersonal Tensions Experienced by Surrogate Decision Makers in the ICU.” Journal of General Internal Medicine 27(12): 1657–​1665. Schneider, K., Reed, J.D., Levitt, S., and Gleick, J. (1994). “Hour of Farewell.” People 41(21 June 6). Available at http://​www.people.com/​people/​archive/​article/​0,,20108270,00.html. Accessed February 17, 2014. Sedgwick, E.K., and Frank, A., eds. (1995). Shame and Its Sisters: A Silvan Tompkins Reader. (Durham, NC: Duke University Press). Smith, R.C. (2002). Patient-​ Centered Interviewing:  An Evidence-​ Based Method. 2d ed. (Philadelphia: Lippincott Williams & Wilkins). Solomon, S., Greenberg, J., and Pyszczynski, T. (1991). “A Terror Management Theory of Social Behavior:  The Psychological Functions of Self-​Esteem and Cultural Worldviews.” In Advances in Experimental Social Psychology, edited by M. Zanna. Vol. 24, 92–​159. (Orlando: Academic Press). Solomon, S., and Lawlor, K. (2011). “Death Anxiety: The Challenge and the Promise of Whole Person Care.” In Whole Person Care: A New Paradigm for the 21st Century, edited by T.A. Hutchinson, 97–​107 (New York: Springer Science+Business Media). Street, A.F., and Kissane, D.W. (2001). “Discourses of the Body in Euthanasia: Symptomatic, Dependent, Shameful, and Temporal.” Nursing Inquiry 8(3): 162–​172. Teno, J.M., Gozalo, P.L., Bynum, J.P.W., et al. (2013). “Change in End-​of-​Life Care for Medicare Beneficiaries: Site of Death, Place of Care, and Health Care Transitions in 2000, 2005, and 2009.” JAMA 309(5): 470–​477. The, A-​M., Hak, T., Koёter, G., and van der Wal, G. (2000). “Collusion in Doctor–​Patient Communication about Imminent Death: An Ethnographic Study.” BMJ 321(December 2): 1376–​1381.

Chapter 15

The C ost of Dyi ng A mong the Elderly i n t h e United Stat e s  Ethical Issues Susannah L. Rose and Janelle Highland

Introduction More than 80% of patients say that they wish to avoid hospitalization and intensive care during the terminal phase of illness, but those wishes are often overridden by other factors. If more intense intervention does not improve life expectancy, and if most patients prefer less care when more intensive care is likely to be futile, the fundamental question is whether the quality of care in regions with fewer resources and more conservative practice styles is better than in regions where more aggressive treatment is the norm (“End of Life Care,” 2014). In addition to its effects on patients’ quality of life, unnecessarily aggressive care carries a high financial cost. About one fourth of all Medicare spending goes to pay for the care of beneficiaries in their last year of life, and much of the growth in Medicare spending in recent decades is the result of the high cost of treating chronic disease. (Goodman et al., 2011)

End-​of-​life care in the United States is expensive and may not serve the intended goals of medicine—​to prolong life and to reduce suffering—​and, in addition to not meeting these goals, such intensive care may not be what people want. Furthermore, what do we do about providing end-​of-​life care to the elderly? Claims are laid by The Dartmouth Atlas of Health Care researchers and others that the elderly disproportionately use

The Cost of Dying Among the Elderly    243 medical resources at the end of life. This problem may be compounded given the growing portion of the elderly in our society as baby boomers retire (Callahan, 2012). Some claim that the United States is reaching a crisis (Callahan, 2012), and we need to find ways to stabilize our health-​care economy while keeping in mind an accurate portrayal of the empirical research and how these findings may impact normative claims on the elderly’s use of medical resources. In this chapter, we evaluate the evidence regarding the costs of end-​of-​life care among the elderly and the evidence regarding the claims regarding quality of end-​of-​life and patient preferences. Although many of these claims are accurate, a recurring theme we emphasize is that the picture is much more complex than most acknowledge. We provide support for the claim that the rise in health-​care costs in the United States is not solely driven by end-​of-​life care. Second, as stated previously, most people do not want intensive end-​of-​life interventions and would rather prefer to die at home. We argue that (a) this empirical claim may not be as pervasive as people believe it to be and (b) if there is significant variation in preferences, then ethical claims that we should provide less intensive care need a different basis. We propose that the money spent on end-​of-​ life care is generally well spent, but we claim that changing what we are spending this money on, including improving access to palliative care, will improve the quality of care and likely reduce costs for society and individuals in the long term. However, as part of our proposal we accept the implication that providing a good death is of primary ethical importance, and achieving this goal may actually cost more money or require reprioritization of money already spent. Last, we suggest policy changes based on conceptions of autonomy proposed in behavioral economics.

Cost of End-​of-​L ife Health Care in the United States According to the Centers for Medicare & Medicaid Services, in 2012 the United States spent approximately $8,915 per person on health care, which amounts to around $2.8 trillion or about 17.2% of the gross domestic product (GDP) (Centers for Medicare & Medicaid Services, 2013). Importantly, health-​care spending is growing at an alarming rate, despite a small slowdown recently, and there is significant concern regarding the United States’ ability to sustain the amount of resources on health care—​particularly given that those health-​care dollars are not delivering the highest quality care (Yong and Olsen, 2010). A primary purpose of the Patient Protection and Affordable Care Act (2009) is to reduce costs and improve quality; however, it is too early to determine the impact this legislation will have on health care in the United States, particularly related to cost savings. However, economists disagree about the impact of health care’s high percentage of the GDP: some argue that this aspect of the economy has positive effects, while others argue that it may create problems in the labor market and slow overall economic growth (“Effects of Health Care Spending on the U.S. Economy,” 2005).

244    Susannah L. Rose and Janelle Highland Medicare is federally run insurance, primarily for elderly Americans, and is the largest payer of end-​of-​life care in the United States (Medicaid is the second largest). Approximately 10% of Medicare beneficiaries accounted for 55% of Medicare spending (Medicare Payment Advisory Commission, 2011). However, who are these high-​ spending patients, and what is the money being used for? It has been argued that the high cost of health care in the United States has more to do with increases in chronic diseases and our inability to effectively manage “complex” patients who have multiple chronic diseases using current practices than increases in end-​of-​life care spending itself (Crippen and Barnato, 2011). In addition, as these authors point out, the likelihood of being diagnosed with a chronic illness, and even multiple chronic illnesses, increases with age. According to The Dartmouth Atlas of Health Care, “Patients with chronic illness in their last two years of life account for about 32% of total Medicare spending, much of it going toward physician and hospital fees associated with repeated hospitalizations” (“Care of Chronic Illness in the Last Two Years of Life,” 2014). These dire statistics are often presented in arguments claiming that the United States is spending too much on acute care, particularly hospital admission and expensive intensive care units. Given that increased age may reduce the overall success of intensive interventions, it has been suggested that the elderly may receive too much invasive, unnecessary care and that we need to spend less on this group of people (Callahan, 2012). Ethicists and policy experts have explored different justifications for arguments to curtail health-​care spending for the elderly. A central tenant of many of these arguments is nicely summated by Fleck, who points out that it is often implicitly or explicitly stated that we have, “limited resources to meet virtually unlimited health care needs ” (2011: 156). This concern about the limited resources that we can dedicate to health-​care spending leads to concerns about equity and justice. Crippen and Barnato (2011), while not taking a particular moral position, argue that spending too much on health care for the elderly may place younger generations at a financial and social disadvantage. Other ethical arguments focus on the idea of using health-​care funds to maximize either the quality or the quantity of life. The “fair innings” and “natural lifespan” arguments propose that health-​care funds should be spent on those who have not had the opportunity to live a full life or who are not yet approaching the end of their lifespan and that life-​extending treatments should not be used on those at the end of their lifespan at the expense of others (Gruenewald, 2012; Andre and Velasquez, 1990). We do not examine these arguments in this chapter, but rather highlight that there are ethical arguments to support care limits for the elderly, although none are conclusive. Despite the aforementioned arguments and statistics, the use of resources is more complicated than what is typically quoted in the media or even in many academic discussions on health-​care spending among the elderly and in end-​of-​life care. Skilled nursing facility care, prescription drug coverage (under Medicare Part D), and home care health care account for significant portions of Medicare expenditures and even larger portions of Medicaid spending. In fact, inpatient hospital admission spending was substantially reduced as a portion of Medicare spending between 2000 (39%) and 2010 (27 %; Medicare Payment Advisory Commission, 2011). It has also been argued that the way

The Cost of Dying Among the Elderly    245 in which analyses on end-​of-​life spending are conducted—​examining Medicare spending retrospectively over a designated time period opposed to calculating how money is spent on patients with life-​limiting illnesses prospectively—​encourages the view that spending on these patients is unwise and that end-​of-​life spending is ripe for efforts at cost containment. This method treats all expenditures for these patients as if they were clearly terminal when in reality it may be difficult, if not impossible, to determine prospectively when they reach that stage (Scitovsky, 1984/​2005; Bayer et al., 1983). The picture becomes even more complicated when we investigate spending by the elderly in particular. Many studies have demonstrated that Medicare expenditures actually decrease with age, particularly for people age eighty-​five or older (Levinsky et al., 2001). It also appears that the eldest elderly receive less aggressive care than their younger elderly counterparts (ages 65 to 74), as evidenced by experiencing less frequent hospital admissions and fewer expensive medical interventions (Levinsky et al., 2001; Shugarman, Decker, and Bercovitz, 2009). These findings may also demonstrate that is not merely age that is driving health-​care expenditures at the end of life but rather life expectancy and age (and likely many other factors)—​people who are not expected to live long given that they are very old are getting fewer and less intensive services, including chemotherapy, which may cost less on average. Therefore, it is important to look at the variation of utilization among the diverse group considered elderly and to critically evaluate these empirical claims. Given that the elderly, as a whole group, may not be responsible for high health-​care costs, perhaps targeting the elderly for cost reductions is not the best way to approach policy changes. This raises the question of what the right targets for policy changes related to end-​of-​life spending may be. We propose more appropriate policy changes intended to curtail end-​of-​life medical care spending and in doing so examine the use of the principal of autonomy in the discourse related to end-​of-​life care spending. Autonomy-​related arguments tend to generally focus on evidence demonstrating that people with serious life-​limiting illnesses do not want invasive (expensive) medical care and would prefer to die a “good death” at home. Therefore, we should honor such preferences and provide less intensive health care (which would also reduce costs). Given that this argument rests on an empirical claim—​that people do not (generally) want invasive treatment—​we next explore the empirical evidence related to this claim, demonstrating that conceptions of a good death and end-​of-​life care preferences vary greatly, which raises concerns about the foundations of arguments based on the belief that people do not want invasive end-​of-​life care.

A Good Death and End-​of-​L ife Care Preferences A good death is defined in many ways in the literature, and its conception is important to discuss, given that achieving a good death is an often-​stated goal for end-​of-​life

246    Susannah L. Rose and Janelle Highland interventions (Granda-​Cameron and Houldin, 2012; Kehl, 2006; Emanuel and Emanuel, 1998; Steinhauser, Clipp, et al., 2000; Steinhauser, Christakis, et al., 2000). Emanuel and Emanuel highlight the importance of conceptualizing a good death in order to know how to provide it. They cite the Institute of Medicine’s definition: “[A]‌decent or good death is one that is: free from avoidable distress and suffering for patients, families, and caregivers; in general accord with patients’ and families’ wishes; and reasonably consistent with clinical, cultural, and ethical standards” (Field and Cassel, 1997: 4). However, there are problems with this conception. Most important, a good death, defined in this way, is frequently not a realistic possibility, regardless of care setting but particularly in the home environment. For example, symptom management is often too complicated to adequately achieve in the home, due to the complexity of the medical and symptomatic problems (Townsend et al., 1990) and to a general lack of support in our society to help informal caregivers to provide such a dignified and peaceful death in the home. Although Medicare and other insurers cover home hospice services, often this includes only very limited hands-​on care, leaving families to provide the care themselves or to hire privately. The burden and costs of providing such care can be staggering (Gardiner et al., 2014; Rabow et al., 2004). Therefore, providing a good death at home, even if it is preferred by the patient, is often not possible in our current health-​care system. We discuss ways to improve the quality of care that may remove some of the practical barriers to a desired death at home in the next section. We agree with Emanuel and Emanuel (1998) that the Institute of Medicine’s definition is too vague to be practically helpful in guiding care decisions. Instead, the authors provide an interactive, multiple systems–​ level approach to achieving a good death, one that reflects not only the individual patient’s medical and psychosocial characteristics but also includes community, spirituality, and pragmatism. As we discuss later, palliative care, both as a philosophy of care and as a subspecialty, can achieve these goals because its primary aim is to attain “the best-​possible quality of life for patients and their family caregivers, based on patient and family needs and goals and independent of prognosis” (Meier, 2011: 344) and because of the diversity of settings in which it is practiced (National Consensus Project for Quality Palliative Care, 2009). If we want to construct a health-​care system that can achieve a good death for most people, we must consider people’s preferences regarding end-​of-​life care—​not only for individual care decisions but also to help policymakers decide how to direct resources. We need to ask: What do people really want when they die? Studies have revealed that many people wish to die at home and not have invasive intervention at the end of life (Connors et al., 1995; Barnato et al., 2007; Higginson and Sen Gupta, 2000; Pritchard et al., 1998; Gomes et al., 2012). In my (SR) clinical experience as a clinical oncology social worker and bioethicist for almost two decades, these results are generally consistent with what patient say. However, we do have concerns that many of these studies base such claims on overall percentages, or averages in people’s end-​of-​life care preferences, and these methods of describing research findings may not accurately reflect the true variation and complexity of people’s end-​of-​life care preferences. These variations are ethically relevant: if we want to help people exercise their autonomy, we need to allow

The Cost of Dying Among the Elderly    247 for a wide variation of end-​of-​life care options and opportunities to carefully deliberate these options and not merely force a certain perspective of a good death, which constrains such options. While supporting that many would prefer to die at home, work by Gomes et al. (2012) and Khan, Gomes, and Higginson (2014) on preferences for place of death also shows that a considerable minority would prefer to die in hospice or palliative care units. Variations in end-​of-​life care preferences often change depending on when people are asked. For example, people may say that they want to trade quality of life for length of life early in their disease trajectory, but as they get sicker, they alter these opinions. They may experience what is labeled “adaptive preferences”—​when people grow accustomed to their condition and alter their preferences to accommodate this shift (Townsend et al., 1990; Gomes et al., 2013; Colburn, 2013). For example, people with cancer may decide that the pain and nausea is not so bad after all and may wish to receive invasive treatments to extend life longer, even though they previously did not think that they would prefer such a trade-​off. Empirical research on end-​of-​life care preferences has shown that some people will change their minds about their end-​of-​life medical treatment wishes (Wittink et al., 2008; Fried et al., 2006). Ditto and colleagues state, “Preferences for life-​sustaining treatment are dependent on the context in which they are made, and thus individuals may express different treatment preferences when they are healthy than when they are ill” (2006: 313). In one study, among a sample of 226 elderly people diagnosed with serious illnesses (cancer, heart failure, or chronic pulmonary disease) living in the community, those who experienced pain or a decline in their functional abilities were more likely to accept pain and disability as a side effect of treatment compared to those participants who did not suffer from pain or physical decline (Fried et al., 2006). In addition to intrapersonal variations in preferences, there may be systematic interpersonal variations in end-​of-​life care preferences. For example, there is growing evidence that racial differences exist in preferences for end-​of-​life care in the United States. A 2013 Pew Research Center found that 61% of blacks and 55% of Hispanics surveyed woud want their doctors to do everything that they could to sustain their lives if they faced an incurable disease and were experiencing considerable pain, compared to only 26% of whites who stated the same preference (Pew Research Center, 2013). Other empirical studies have had similar findings; blacks and Hispanics are more likely than whites to prefer more aggressive life-​sustaining treatment, and these minority groups are less likely to prefer hospice care (Barnato et al., 2009; Borum, Lynn, and Zhong, 2000; Crawley et  al., 2000; Steinhauser, Clipp, et  al., 2000; Steinhauser, Christakis, et al., 2000; Phipps et al., 2003; Hopp and Duffy, 2000; Garrett et al., 1993; Degenholtz, Thomas, and Miller, 2003; Blackhall et al., 1999). Barnato and colleagues conducted a large, national survey of 2,847 Medicare beneficiaries to assess their preferences regarding treatment at the end of life. Confirming the findings of other studies, they found that a majority of the sample wanted to die at home if they had a terminal illness and to avoid burdensome treatments with little benefit. After controlling for key confounding variables, however, they found once again that Hispanics and blacks more frequently wanted more intensive medical care than whites, including a preference for dying in the

248    Susannah L. Rose and Janelle Highland hospital, having more extensive drug treatment despite negative side effects, having ventilator support for short-​term life extension (of a month or a week), and not receiving palliative treatments if reduced life expectancy was a consequence.

Palliative Care as a Method of Improving Quality and Costs There is strong consensus that the medical system needs to provide better care for patients with advanced illnesses, including providing better symptom control, reducing patient and family distress, and improving quality of life (Hui et al., 2010; Ferrell, Paice, and Koczywas, 2008; Gawande, 2010). One way to effectively manage patients with complicated and advanced illnesses is to introduce hospice, palliative care, and other support measures earlier into the disease trajectory. Palliative care focuses on treating patients’ symptoms and facilitates decision-​making so that patients experience the highest quality of life for the longest time possible (Kelley and Meier, 2010; Walsh, 2001; Temel et al., 2010). Importantly, palliative care promotes a model of care that helps patients, family members, and treatment teams regularly evaluate the goals of care, which can minimize the use of ineffective, burdensome treatments (Temel et al., 2010). The use of multidisciplinary teams, advance directives, and increased family support has been shown to enhance the quality of care and to reduce costs associated with aggressive end-​of-​life care (Rummans et al., 2006; Gries et al., 2008). By providing patients with a means to assess their goals of care, palliative care consults can improve the quality of care provided to sick patients. The opportunity to assess or establish goals of care can also have the effect of reducing the cost of care. In practice, many palliative care services do not produce a profit. However, they offer a means of “cost avoidance”; payers and hospitals may experience cost savings when patients are transferred to palliative care (Smith and Cassel, 2009; Meier, 2011). Several trials have been published that aimed to quantify the cost savings associated with palliative care. A study published in Health Affairs found that palliative care consultations among Medicaid beneficiaries cut hospitals costs by $6,900, on average, compared to a group of matched controls who received usual care. The authors stated, “We estimate that the reductions in Medicaid hospital spending in New York State could eventually range from $84 million to $252 million annually,… if every hospital with 150 or more beds had a fully operational palliative care consultation team” (Morrison et al., 2011: 454). Another study (Smith et al., 2003) showed that patients receiving specialized palliative care treatment had lower daily costs of care, as well as direct and indirect costs compared to the control group. Gade et al. (2008) conducted a three-​site prospective trial comparing an inpatient palliative care service with usual care and found a reduction in the mean cost of care for patients in the inpatient palliative care service. Meier, a national expert in health policy and palliative care, states,

The Cost of Dying Among the Elderly    249 …palliative care programs could be serving approximately 6 percent of all hospital discharges. [Using 2009 data, PC programs affect] … approximately 1.5 percent of all discharges. PC is estimated to save $1.2 billion per year under the current penetration of services… . This figure would increase to approximately $4 billion per year if capacity were to expand to meet the needs of 6 percent of hospital discharges in 90 percent of all U.S. hospitals with more than fifty beds. (2011: 350)

In addition to improving the cost effectiveness of care, palliative care improves the quality of care in several ways. Aspinal et al. described quality as multidimensional, consisting of “effectiveness, acceptability, efficiency, access, equity and relevance” (2003: 325). If, as we propose, the aim of palliative care is ensure that patients experience the highest quality of life for the longest time possible by facilitating decision-​making and treating patients’ symptoms, it may be useful to think about quality of palliative care in these terms. Several studies have shown that palliative care interventions are effective in improving the quality of life of seriously ill patients and their families, as well as reducing symptoms and distress associated with complicated diseases (Finlay et al., 2002; Casarett et al., 2008). The evidence is growing regarding the benefits of palliative in outpatient care settings (Temel et al., 2010; Jacobsen et al., 2011; Temel et al., 2007; Rabow et al., 2013). Clinical trials of palliative care interventions have demonstrated a variety of improvements in patients’ physical, emotional, and spiritual well-​being (Meier, 2011; Temel et al., 2010; Bakitas et al., 2009; Pantilat et al., 2010; Brumley et al., 2007; Follwell et al., 2009; Rummans et al., 2006). Studies also show that, overall, patients and family members are satisfied with the care that they receive from palliative care services (Gade et al., 2008; O’Mahony et al., 2005). This evidence has convinced the American Society of Clinical Oncology to issue a consensus statement recommending palliative care in conjunction with standard care for advanced-​stage cancer patients (Smith et al., 2012). Strong support also comes from palliative care advocates and professional associations for the integration of palliative care into end-​of-​life care for patients with other illnesses, including heart disease, stroke, dementia, amyotrophic lateral sclerosis, and other serious illnesses (Meier, 2011; Hunt el al., 2005). Based on this evidence, perhaps we need to focus resources on providing these helpful interventions. However, developing a national system of standard palliative care may be expensive, and there needs to be a culture shift in the US health-​ care system that values palliative care. Professional guidelines and empirical research with patients show that common elements of good quality end-​of-​life care include pain and symptom control; avoiding inappropriate prolongation of dying; helping to establish a sense of control; relieving burden for patients and families; preserving autonomy; and addressing practical, existential, or spiritual concerns (Singer, Martin, and Kelner, 1999). As the studies cited previously demonstrate, a care plan that includes palliative care can provide patients with quality end-​of-​life care. Palliative care has been promoted by the National Priorities Partnership (2008) as one of six priorities of improving quality in health care, advocating that palliative care can reduce harm, disease burden,

250    Susannah L. Rose and Janelle Highland disparities, and waste in care. It may increase the quality of care because it aims to provide patients’ family members and clinicians with the tools needed integrate care across multiple systems and levels, address chronic illnesses, and encourage the elicitation and documentation of preferences (National Consensus Project for Quality Palliative Care, 2009). For example, palliative care guidelines encourage advanced care planning, including the completion of advanced directives, and there is evidence that advanced directives improve the quality of communication between physicians, patients, and family members as well as accuracy of expectations about outcomes (Teno et al., 2007; Temel et al., 2007). Palliative care can also support the continuity of care by maintaining connections with formal and informal caregivers across various settings and promoting the timely exchange of information (National Consensus Project for Quality Palliative Care, 2009).

Autonomy and Soft Paternalism The principle of autonomy is often credited with starting what we call a “bioethics revolution.” The story, put simply, is that physicians in the twentieth century consolidated significant political, social, and economic power, both legally and internally, through developing into a profession with complex methods of self-​regulation. Physicians in the midcentury enjoyed almost the ultimate autonomy both as a profession and as individuals within the profession: this supported what some call an extremely paternalistic health-​care system where physicians often made decisions for patients without their consent or knowledge. However, a series of key court cases, public awareness, policy reforms (the 1990 Patient Self Determination Act), and bioethicists’ arguments supporting patient autonomy put the patient (and often the family) at the center of medical decision-​making (Cruzan v. Director, 1990; In re Quinlan, 1976; Patient Self Determination Act, 1990; Baker and McCullough, 2012; Starr, 2007; Steinberg and Youngner, 1998). Through a process of informed consent and, more recently, shared decision-​making, patients and physicians are supposed to work together to find the right treatments that fit with the patient’s values and preferences. Generally, this movement toward patient autonomy and reduced paternalism has been hailed as a success in terms of improving the quality of the process and outcomes in health care—​patients can direct their care and be more involved, even to the point of creating advance directives, which may help direct their care after they lose the cognitive ability to make their own decisions. However, patient autonomy can be critiqued on a practical level. It may not be realistic for patients to make decisions because they may lack the knowledge to make such informed decisions, given that some medical decisions involve complicated scientific evaluations. Therefore we may be undermining the concept of informed consent if the person deciding cannot accurately weigh the information provided. Efforts to improve the informed consent process have included decision aids for patients and physicians;

The Cost of Dying Among the Elderly    251 clinician communication skills training, and the use of patient navigators. (O’Connor, 2007; Baile, 2011; Ubel, 2012; Dohan and Schraig, 2005). Many people who write about how to contain costs focus on arguments based on the principle of justice—​finding ways to fairly ration care to allow people to receive high-​ quality health care in a system with finite resources. When we refer to the fair rationing of care, we mean explicitly rationing or “the systematic allocation of resources within health care systems” ” (Moreira, 2011: 1333) in a manner that is free of bias or has the effect of mitigating inequalities in opportunity (Daniels, 2008). According to Daniels, Saloner, and Gelpi (2009), it is of ethical concern if a nation receives little value, in terms of needs met and money spent, for its health-​care spending, and it is morally important that the money spent on health care is allocated in a way that achieves the most value, because that money may be spent elsewhere, such as on schooling. In this way, the cost effectiveness or value of care is a matter of justice as fairness. We fully support these efforts and find the arguments made by philosophers and bioethicists, such as Daniels (2008) and Sen (2009), compelling. Furthermore, there are effective, practical examples of ethically defensible fair rationing schemes used in other countries, such as the United Kingdom, where the government makes systematic national insurance coverage decisions based on the effectiveness and costs of treatments. In addition to using cost effectiveness as a primary tool to determine coverage decisions, these decisions are made within a context of a budget allocated to health-​care spending—​which puts specific restrictions on what the government can afford. The concept of only offering what it can afford is literally a foreign concept in US health-​care system: the American public is averse to the idea of to the idea of fair rationing schemes, due in part to wariness about governmental intervention (Schlesinger, 2004; Jacobs, 2008). Furthermore, even if public opinion suddenly swayed toward one of imposing limits on health care, creating a consistent rationing scheme would be almost impossible in the US system given the complicated mixture of payers and providers. Furthermore, a national health care budget does not exist. On a theoretical level, unbridled autonomy may conflict with certain theories of justice (Daniels, 2008), in which what an individual wants may not be as important as providing a fair system of health-​care allocation to all. Daniels et al.’s (2009) take on Rawls’s conceptualization of justice as fairness in relation to health care argues that meeting health needs is a way of protecting opportunity, as it promotes the normal functioning of individuals in society. The unequal allocation of health-​care spending may present a problem because it reduces opportunity for others. Another way of conceptualizing this idea would be considering the budget of a small household. Within that household, there may multiple and competing needs to be met with limited money to meet those needs. If one member of that household has the ability to spend as much as he or she wishes on a particular item and consume that item, it not only constrains the ability of others in the household to purchase not only that item but also other goods. Therefore, one could argue that it is sensible to manage how much money is allocated to particular purchases and to consider how spending affects other members of that household. Still, rationing and fair systems seem incompatible with the kind of

252    Susannah L. Rose and Janelle Highland autonomy and individual freedoms that Americans seem to value, creating a public opinion problem. However, there is room within autonomy to establish systems that seek to maximize freedom of choice in end-​of-​life care—​one that may even be palatable to the American public given its libertarian roots. We propose we adopt a “soft paternalism” approach to end-​of-​life care decision-​making, one that provides patients with the autonomy to make decisions as they currently enjoy in this country but make the “defaults” more in line with providing high-​quality end-​of-​life care and reducing societal costs. Soft paternalism (also called libertarian paternalism or asymmetrical paternalism) was made popular by Thaler and Sunstein (Sunstein and Thaler, 2003; Thaler and Sunstein, 2003; Thaler and Sunstein 2009) and is a method of presenting various choice options in ways that enable people to make optimal decisions. In this framework, people maintain autonomy in that they still make the decisions based on their own values and preferences, but the “default” options are constructed in ways that benefit the decision-​maker and society. Thaler and Sunstein use an example of employer systems for retirement savings: instead saving nothing for retirement, the default is rather for employees to save a moderate amount with the choice of reducing or increasing the saving amount. In other words, the default is changed from saving nothing to saving a moderate amount. Saving for retirement is considered to be in almost everyone’s best interests, but if a person does not value such savings, he or she retains the choice to save nothing. Importantly, saving for retirement is not only in the individual’s best interests but also their families’ and society’s: if people do not save enough for retirement, they may not be able to meet family obligations and may be increasingly dependent on societal supports, such as Medicaid or insurance subsidies as part of the Patient Protection and Affordable Care Act. We propose that we utilize such “choice architecture” to help inform people’s decisions regarding end-​of-​life care, including the elderly. As stated previously, people have a wide range of preferences regarding end-​of-​life care; however, the current default of providing the newest technologies and invasive care may not be in people’s best interests or their families or society. In my (SR) clinical practice as a bioethicist, ethicists are often called to the intensive care units or hospital wards to help patients and families make end-​of-​life decisions, and the following message is frequently delivered to patients and family members by physicians: “We [the medical team] will provide all these things [x, y, and z], but we do not think any of these things will help. However, we leave the decision to you about what we should do.” Then the patient and family are left to decide what to do, often with little understanding of what x, y, and z actually are or the risks and benefits, despite the physicians’ best efforts to communicate such information. This is compounded by the circumstances of patients’ and families’ fears, lack of knowledge, and feelings of uncertainty—​in addition to the feelings of being responsible or blameworthy if the wrong decision is made. The cultural norm in acute care is to provide everything possible to treat medical problems, despite the risks and benefits: the current default does not provide the best comfort care and often leaves physicians, patients, and family members ill prepared for death (Nutick Zitter,

The Cost of Dying Among the Elderly    253 2014). The patient or family members may also assume that the team would not be offering options that would not be beneficial, so they follow the current default: continue invasive care. As Thaler, Sunstein, and Balz state: many people will take whatever option requires the least effort, or the path of least resistance. All these forces imply that if, for a given choice, there is a default option—​ an option that will obtain if the chooser does nothing—​then we can expect a large number of people to end up with that option, whether or not it is good for them. These behavioral tendencies toward doing nothing will be reinforced if the default option comes with some implicit or explicit suggestion that it represents the normal or even the recommended course of action. (2010: 4)

When the medical team states that “we will provide all these things [x, y, and z],” it is implied that this is the course of action they recommend, and the default is most often to error on the side of doing everything possible, even if the risk of harm is great and the possible value is miniscule. It is not that the physicians are misrepresenting what they think is the truth. Rather, it is that our health-​care system establishes the defaults of the choice architecture to be to provide high-​technology, invasive care through continued reimbursement structures that favor such interventions. The US health-​care system needs to change these defaults to help everyone make informed decisions, and these defaults should be to provide comfort and reduce harm. Next we focus on one important default option: providing palliative care. As discussed previously, palliative care has been shown to benefit people physically and emotionally (and has demonstrated few harms, if any, that we know of in the literature); the US healthcare system should provide palliative care as a default consultation to people who have chronic and life-​limiting illnesses early in their disease process—​for example, people diagnosed with advanced cancers, advanced heart failure or pulmonary disease, or dementia. Palliative care should be introduced soon after the initial diagnosis, as Temel demonstrated in her randomized clinical trial of outpatient palliative care among lung cancer patients (Temel et al., 2010). Furthermore, palliative care can be provided with other medical interventions, including life-​extending measures, and therefore does not exclude options for people but enhances their access to the full range of options and, potentially increases their ability to exercise autonomous decision-​making. Receiving palliative care concurrently with aggressive treatments has been shown to reduce the use of invasive, nonbeneficial treatments, primarily due to people choosing not to pursue such aggressive treatments—​actualizing their own preferences. This is an important point, because receiving palliative care does not limit other options, as hospice care might. Offering palliative care as a default may be particularly helpful to the elderly, given that they are more likely to have chronic diseases and fewer social supports. This means, however, that significant resources need to be dedicated to providing palliative care in the United States, in addition to the means for people to stay in the

254    Susannah L. Rose and Janelle Highland environments they choose, including home, if this is medically possible. This would require additional training programs and the extension of palliative care into outpatient and home-​care environments. This will require a lot of effort and money and may potentially cost more, particularly in the short term, than is currently being spent on end-​of-​life medical interventions in the United States. However, as has been shown in the palliative care research, this type of care not only enhances the quality of care at the end of life but also reduces costs, on average, in the long term through mechanisms of cost-​avoidance.

Conclusion The elderly cost a lot in terms of health care. Reducing health-​care costs in the United States is a priority, and some have argued that limiting care to the elderly would be one way to save costs. However, as outlined here, being elderly, by itself, is not the cause of high costs at the end of life: the growing total number and percentage of the population with chronic diseases is likely the most direct contributor to rising costs. Additionally, some have argued that people do not want invasive care at the end of life and claim that, for this reason, we should provide less invasive care and more comfort care to help people actualize conceptions of a good death. However, as we discussed, preferences at the end of life vary within individuals over time and circumstances and among individuals in systematic ways, such as certain minorities preferring more invasive end-​of-​life care than whites on average. The US health-​care system is not equipped to implement even the fairest rationing scheme to reduce costs, including end-​of-​life care costs. Given all of these factors, we need to implement a system that improves end-​of-​life care quality for everyone, including the elderly, and maintains patient autonomy for medical decision-​making but best equips people to make decisions in their best interests and the best interests of society. We have outlined a method of decision architecture that would allow people the full range of options currently available to them but would change the default choice. The default choice should be one that includes quality palliative care, which can help people die well and prepare them to make the best choices for themselves. This may be a controversial proposal, and an expensive one, but this option allows people to maintain autonomy, will almost certainly improve quality of care at the end of life, and will most likely reduce costs to society and to individuals in the long term.

Acknowledgments We would like to acknowledge Ms. Ruchi Sanghani and Dr. Paul Ford for their valuable feedback on this chapter.

The Cost of Dying Among the Elderly    255

References Andre, C., and M. Velasquez. (1990). “Aged-​based health care rationing.” Issues in Ethics 3(3). http://​www.scu.edu/​ethics/​publications/​iie/​v3n3/​age.html. Aspinal, F., J. Addington-​Hall, R. Hughes, and I. J. Higginson. (2003). “Using satisfaction to measure the quality of palliative care: a review of the literature.” Journal of Advanced Nursing 42(4): 324–​339. Baile, W. F. (2011). “Training oncology practitioners in communication skills.” Journal of Pediatric Hematology/​Oncology 33(Suppl. 2): S115–​S122. Baker, R., and L. McCullough, eds. (2012). The Cambridge world history of medical ethics. New York: Cambridge University Press. Bakitas, M., K. D. Lyons, M. T. Hegel, et al. (2009). “Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: The Project ENABLE II randomized controlled trial.” JAMA 302(7): 741–​749. Barnato, A. E., D. L. Anthony, J. Skinner, P. M. Gallagher, and E. S. Fisher (2009). “Racial and ethnic differences in preferences for end-​of-​life treatment.” Journal of General Internal Medicine 24(6): 695–​701. Barnato, A. E., M. B. Herndon, D. L. Anthony, P. M. Gallagher, J. S. Skinner, J. P. Bynum, and E. S. Fisher. (2007). “Are regional variations in end-​of-​life care intensity explained by patient preferences? A study of the US Medicare population.” Medical Care 45(5): 386–​393. Bayer, R., D. Callahan, J. Fletcher, et al. (1983). “The care of the terminally ill: Morality and economics.” New England Journal of Medicine 309: 1490–​1494. Blackhall, L. J., G. Frank, S. T. Murphy, V. Michel, J. M. Palmer, and S. P. Azen. (1999). “Ethnicity and attitudes towards life sustaining technology.” Social Science & Medicine 48(12): 1779–​1789. Borum, M. L., J. Lynn, and Z. Zhong. (2000). “The effects of patient race on outcomes in seriously ill patients in SUPPORT: An overview of economic impact, medical intervention, and end-​ of-​ life decisions. Study to understand prognoses and preferences for outcomes and risks of treatments.” Journal of the American Geriatrics Society 48(Suppl. 5): S194–​S198. Brumley, R., S. Enguidanos, P. Jamison, et al. (2007). “Increased satisfaction with care and lower costs: Results of a randomized trial of in-​home palliative care.” Journal of the American Geriatrics Society 55(7): 993–​1000. Callahan, D. (2012). “Must we ration health care for the elderly?” The Journal of Law, Medicine & Ethics 40(1): 10–​16. “Care of chronic illness in the last two years of life.” (2014). In The Dartmouth atlas of health care. Accessed March 28, 2014, http://​www.dartmouthatlas.org/​data/​topic/​topic.aspx?cat=1. Casarett, D., A. Pickard, F. A. Bailey, C. Ritchie, et  al. (2008). “Do palliative consultations improve patient outcomes?” Journal of the American Geriatrics Society 56(4): 593–​599. Centers for Medicare & Medicaid Services. (2013). National health expenditures 2012 highlights. Accessed December 24, 2013, https://​wayback.archive-​it.org/​2744/​20141127165214/​. Colburn, B. (2013). “Autonomy and end of life decisions: A paradox.” In Adaptation and autonomy: Adaptive preferences in enhancing and ending life, edited by J. Raikka and J. Varelius. Pavia, Italy: Springer, 69–​80. Connors, A. F., N. V. Dawson, N. A. Desbiens, et al. (1995). “A controlled trial to improve care for seriously ill hospitalized patients: The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT).” JAMA 274(20): 1591–​1598.

256    Susannah L. Rose and Janelle Highland Crawley, L., R. Payne, J. Bolden, T. Payne, P. Washington, and S. Williams. (2000). “Palliative and end-​of-​life care in the African American community.” JAMA 284(19): 2518–​2521. Crippen, D., and A. E. Barnato. (2011). “The ethical implications of health spending: Death and other expensive conditions.” The Journal of Law, Medicine & Ethics 39(2): 121–​129.

Cruzan v. Director, Missouri Department of Health. (1990). 479 U.S. 261.

Daniels, N. (2008). Just health:  Meeting healthcare needs fairly. New  York:  Cambridge University Press. Daniels, N., B. Saloner, and A. H. Gelpi. (2009). “Access, cost, and financing: Achieving an ethical health reform.” Health Affairs 28: w909–​s916. Degenholtz, H. B., S. B. Thomas, and M. J. Miller. (2003). “Race and the intensive care unit: Disparities and preferences for end-​of-​life care.” Critical Care Medicine 31(5): S373–​S378. Ditto, P. H., J. A. Jacobson, W. D. Smucker, J. H. Danks, and A. Fagerlin. (2006). “Context changes choices: A prospective study of the effects of hospitalization on life-​sustaining treatment preferences.” Medical Decision Making 26(4): 313–​322. Dohan, D., and D. Schraig. (2005). “Using navigators to improve care of underserved patients.” Cancer 104(4): 848–​855. “Effects of health care spending on the U.S.  economy.” (2005). Washington, DC:  US Department of Health and Human Services, Office of Assistant Secretary for Planning and Evaluation. Accessed March 27, 2014, http://​aspe.hhs.gov/​health/​costgrowth/​index.htm. Emanuel, E. J., and L. L. Emanuel. (1998). “The promise of a good death.” The Lancet 351(Suppl. 2): SII21–​SII29. “End of life care.” (2014). In The Dartmouth atlas of health care. Accessed March 28, 2014, http://​ www.dartmouthatlas.org/​data/​topic/​topic.aspx?cat=18. Ferrell, B., J. Paice, and M. Koczywas. (2008). “New standards and implications for improving the quality of supportive oncology practice.” Journal of Clinical Oncology 26(23): 3824–​3831. Field, M. J., and C. K. Cassel, eds. (1997). Approaching death: Improving care at the end of life. Washington, DC: National Academies Press. Finlay, I. G., I. J. Higginson, D. M. Goodwin, et al. (2002). “Palliative care in hospital, hospice, at home: Results from a systematic review.” Annals of Oncology 13(Suppl. 4): 257–​264. Fleck, L. M. (2011). “Just caring: Health care rationing, terminal illness, and the medically least well off.” The Journal of Law, Medicine & Ethics 39(2): 156–​171. Follwell, M., D. Burman, L. W. Le, et al. (2009). “Phase II study of an outpatient palliative care intervention in patients with metastatic cancer.” Journal of Clinical Oncology 27(2): 206–​213. Fried, T. R., A. L. Byers, W. T. Gallo, P. H. Van Ness, V. R. Towle, J. R. O’Leary, and J. A. Dubin. (2006). “Prospective study of health status preferences and changes in preferences over time in older adults.” Archives of Internal Medicine 166(8): 890–​895. Gade, G., I. Venohr, D. Conner, et al. (2008). “Impact of an inpatient palliative care team: A randomized control trial.” Journal of Palliative Medicine 11(2): 180–​190. Gardiner, C., L. Brereton, R. Frey, L. Wilkinson-​Meyers, and M. Gott. (2014). “Exploring the financial impact of caring for family members receiving palliative and end-​of-​life care: A systematic review of the literature.” Palliative Medicine 28(5): 375–​390. Garrett, J., R. P. Harris, J. K. Norburn, D. L. Patrick, and M. Danis. (1993). “Life-​sustaining treatments during terminal illness: Who wants what?” Journal of General Internal Medicine 8(7): 361–​368. Gawande, A. (2010). “Letting go: What should medicine do when it can’t save your life?” The New Yorker (August 2).

The Cost of Dying Among the Elderly    257 Gomes, B., N. Calanzani, M. Gysels, S. Hall, and I. J. Higginson. (2013). “Heterogeneity and changes in preferences for dying at home: A systematic review.” BMC Palliative Care 12: 7. Gomes, B., I. J. Higginson, N. Calanzani, et al. (2012). “Preferences for place of death if faced with advanced cancer:  A  population survey in England, Flanders, Germany, Italy, the Netherlands, Portugal and Spain.” Annals of Oncology 23(8): 2006–​2015. Goodman, D. C., A. R. Esty, E. S. Fisher, and C. Chiang-​Hua. (2011). Trends and variation in end-​of-​life care for Medicare beneficiaries with severe chronic illness. Edited by K. Bronner. A Report of the Dartmouth Atlas Project. Hanover, NH: Dartmouth Institute for Health Policy & Clinical Practice. Granda-​Cameron, C., and A. Houldin. (2012). “Concept analysis of good death in terminally ill patients.” American Journal of Hospice and Palliative Care 29(8): 632–​639. Gries, C. J., J. R. Curtis, R. J. Wall, and R. A. Engelberg. (2008). “Family member satisfaction with end-​of-​life decision making in the ICU.” Chest 133(3): 704–​7 12. Gruenewald, D. A. (2012). “Can health care rationing ever be rational?” The Journal of Law, Medicine & Ethics 40(1): 17–​25. Higginson, I. J., and G. J. Sen-​Gupta. (2000). “Place of care in advanced cancer:  A  qualitative systematic literature review of patient preferences.” Journal of Palliative Medicine 3(3): 287–​300. Hopp, F. P., and S. A. Duffy. (2000). “Racial variations in end-​of-​life care.” Journal of the American Geriatrics Society 48(6): 658–​663.

Patient Protection and Affordable Care Act. (2009). s. H.R. 3590. Patient Self Determination Act of 1990. (1990). s. H.R. 4449.IH.

Hui, D., A. Elsayem, M. De La Cruz, et al. (2010). “Availability and integration of palliative care at US cancer centers.” JAMA 303(11): 1054–​1061. Hunt, S. A., W. T. Abraham, M. H. Chin, et al. (2005). “ACC/​AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult—​Summary article.” Circulation 112(12): 1825–​1852.

In re Quinlan. (1976). 70 N.J. 10, 355 A.2d 647.

Jacobsen, J., V. Jackson, C. Dahlin, et al. (2011). “Components of early outpatient palliative care consultation in patients with metastatic nonsmall cell lung cancer.” Journal of Palliative Medicine 14(4): 459–​464. Kehl, K. A. (2006). “Moving toward peace:  An analysis of the concept of a good death.” American Journal of Hospice and Palliative Medicine 23(4): 277–​286. Kelley, A. S., and D. E. Meier. (2010). “Palliative care—​A shifting paradigm.” New England Journal of Medicine 363(8): 781–​782. Khan, S. A., B. Gomes, and I. J. Higginson. (2014). “End-​of-​life care—​What do cancer patients want?” Nature Reviews Clinical Oncology 11: 100–​108. Levinsky, N. G., W. Yu, A. Ash., M. Moskowitz, G. Gazelle, O. Saynina, and E. J. Emanuel. (2001). “Influence of age on Medicare expenditures and medical care in the last year of life.” JAMA 286(11): 1349–​1355. Medicare Payment Advisory Commission. (2011). Health care spending and the Medicare program: A data book. Washington, DC: Medicare Payment Advisory Commission. Meier, D. E. (2011). “Increased access to palliative care and hospice services: Opportunities to improve value in health care.” The Milbank Quarterly 89(3): 343–​380. Moreria, T. (2011). “Health care rationing in an age of uncertainty: A conceptual model.” Social Science & Medicine 72(8): 1333–​1341.

258    Susannah L. Rose and Janelle Highland Morrison, R. S., J. Dietrich, S. Ladwig, T. Quill, J. Sacco, J. Tangeman, and D. E. Meier. (2011). “Palliative care consultation teams cut hospital costs for Medicaid beneficiaries.” Health Affairs 30(3): 454–​63. National Consensus Project for Quality Palliative Care. (2009). Clinical practice guidelines for quality palliative care, 2nd ed. http://​www.nationalconsensusproject.org/​guideline.pdf. National Priorities Partnership. (2008). National priorities and goals: Aligning our efforts to transform America’s healthcare. Washington, DC: National Quality Forum. Nutick Zitter, J. (2014). “A ‘code death’ for dying patients.” The New York Times (April 10). O’Connor, A. M., J. E. Wennberg, F. Legare, et al. (2007). “Toward the ‘tipping point’: Decision aids and informed patient choice.” Health Affairs 26(3): 716–​725. O’Mahony, S., A. E. Blank, L. Zallman, and P. Selwyn. (2005). “The benefits of a hospital-​ based inpatient palliative care consultation service: Preliminary outcome data.” Journal of Palliative Medicine 8(5): 1033–​1039. Pantilat, S. Z., D. L. O’Riordan, S. L. Dibble, and C. S. Landefeld. (2010). “Hospital-​based palliative medicine consultation: A randomized controlled trial.” Archives of Internal Medicine 170(22): 2038–​2040. Phipps, E., G. True, D. Harris, U. Chong, W. Tester, S. I. Chavin, and L. E. Braitman. (2003). “Approaching the end of life: Attitudes, preferences, and behaviors of African-​American and white patients and their family caregivers.” Journal of Clinical Oncology 21(3): 549–​554. Pritchard, R. S., E. S. Fisher, J. M. Teno, et al. (1998). “Influence of patient preferences and local health system characteristics on the place of death. SUPPORT Investigators. Study to understand prognoses and preferences for risks and outcomes of treatment.” Journal of the American Geriatrics Society 46(10): 1242–​1250. Rabow, M. W., S. L. Dibble, S. Z. Pantilat, and S. J. McPhee. (2004). The comprehensive care team: A controlled trial of outpatient palliative medicine consultation. Archives of Internal Medicine 164(1): 83–​91. Rabow, M. W., E. Kvale, L. Barbour, et al. (2013). Moving upstream: A review of the evidence of the impact of outpatient palliative care. Journal of Palliative Medicine 16(12): 1540–​1549. Rummans, T. A., M. M. Clark, J. A. Sloan, et al. (2006). “Impacting quality of life for patients with advanced cancer with a structured multidisciplinary intervention: A randomized controlled trial.” Journal of Clinical Oncology 24(4): 635–​642. Scitovsky, A. A. (2005). “‘The high cost of dying’: What do the data show?” Milbank Quarterly 83(4): 825–​841. (Original work published 1984) Sen, A. (2009). The idea of justice. Cambridge, MA: Harvard University Press. Shugarman, L. R., S. L. Decker, and A. Bercovitz. (2009). “Demographic and social characteristics and spending at the end of life.” Journal of Pain and Symptom Management 38(1): 15–​26. Singer, P. A., D. K. Martin, and M. Kelner. (1999). “Quality end-​of-​life care: Patients’ perspectives.” JAMA 281(2): 163–​168. Smith, T. J., and J. B. Cassel. (2009). “Cost and non-​clinical outcomes of palliative care.” Journal of Pain and Symptom Management 38(1): 32–​44. Smith, T. J., P. Coyne, B. Cassel, L. Penberthy, A. Hopson, and M. A. Hager. (2003). “A high-​ volume specialist palliative care unit and team may reduce in-​hospital end-​of-​life care costs.” Journal of Palliative Medicine 6(5): 699–​705. Smith, T. J., S. Temin, E. R. Alesi, et al. (2012). “American Society of Clinical Oncology provisional clinical opinion:  The integration of palliative care into standard oncology care.” Journal of Clinical Oncology 30(8): 880–​887.

The Cost of Dying Among the Elderly    259 Starr, P. (2007). The social transformation of American medicine: The rise of a sovereign profession and the making of a vast industry New York: Basic Books. Steinberg, M. D., and S. J. Youngner, eds. (1998). End-​of-​life decisions: A psychosocial perspective Washington, DC: American Psychiatric Press. Steinhauser, K. E., N. A. Christakis, McNeilly, L. McIntyre, and J. A. Tulky. (2000). “Factors Considered important at the end of life by patients, family, physicians, and other care providers.” JAMA 284(19): 2476–​2482. Steinhauser, K. E., E. C. Clipp, M. McNeilly, N. A. Christakis, L. M. McIntyre, and J. A. Tulsky. (2000). “In search of a good death: Observations of patients, families, and providers.” Annals of Internal Medicine 132(10): 825–​832. Sunstein, C. R. and Thaler, R. H. (2003). “Libertarian paternalism is not an oxymoron.” The University of Chicago Law Review 70: 1159–​1202. Temel, J. S., J. A. Greer, A. Muzikansky, et al. (2010). “Early palliative care for patients with metastatic non–​small-​cell lung cancer.” New England Journal of Medicine 363(8): 733–​742. Temel, J. S., V. A. Jackson, J. A. Billings, et al. (2007). “Phase II study: Integrated palliative care in newly diagnosed advanced non-​small-​cell lung cancer patients.” Journal of Clinical Oncology 25(17): 2377–​2382. Teno, J. M., A. Gruneir, Z. Schwartz, A. Nanda, and T. Wetle. (2007). “Association between advance directives and quality of end-​of-​life care: A national study.” Journal of the American Geriatrics Society 55(2): 189–​194. Thaler, R. H., and C. R. Sunstein. (2003). “Libertarian paternalism.” The American Economics Review 93(2): 175–​179. Thaler, R. H., and C. R. Sunstein. (2009). Nudge. New York: Penguin. Thaler, R. H., C. R. Sunstein, and J. P. Balz. (2010). “Choice architecture.” http://​ssrn.com/​ abstract=1583509. Townsend, J., A. O. Frank, D. Fermont, S. Dyer, O. Karran, A. Walgrove, and M. Piper. (1990). “Terminal cancer care and patients’ preference for place of death:  a prospective study.” British Medical Journal 301(6749): 415–​417. Ubel, P. (2012). Critical decisions: How you and your doctor can make the right medical choices together, New York: Harper Collins. Walsh, D. (2001). “The Harry R.  Horvitz Center for Palliative Medicine (1987–​ 1999):  Development of a novel comprehensive integrated program.” American Journal of Hospice Palliative Care 18(4): 239–​250. Wittink, M. N., K. H. Morales, L. A. Meoni, D. E. Ford, N. Y. Wang, M. J. Klag, and J. J. Gallo. (2008). “Stability of preferences for end-​of-​life treatment after 3  years of follow-​up:  The Johns Hopkins precursors study.” Archives of Internal Medicine 168(19): 2125–​2130. Yong, P. L., and L. Olsen. (2010). “Synopsis and overview.” In The healthcare imperative: Lowering costs and improving outcomes: Workshop series summary, edited by P. L. Yong, R. S. Saunders, and L. Olsen. Washington, DC: National Academies Press, 1–​69.

Chapter 16

Death, Dyi ng , and the Di s a bl e d Anita Silvers and Leslie P. Francis

Presuming About Disability There is a familiar presumption that disability abrogates liberty so as to render disabled individuals unfit to decide when to die. People with disabilities, it is said, should not be choosing aid in dying, or withholding or withdrawing care, because they cannot make such choices freely: their decisions are ineluctably nonautonomous. At best, the assumption that people with disabilities cannot act autonomously with respect to these decisions is arguably insultingly paternalistic. At worst, however, presuming that disability distorts or compromises decision-​making capacity subordinates such individuals unjustifiably. Thus our first goal is to alleviate the tension created by supposing that people with disabilities must, in the relevant respect, be less free. We show that crucial misunderstandings of autonomy and its relationship to freedom are at work to propel this presumption. More specifically, ethical problems that characteristically arise in regard to disabled people’s choices at the end of life are of two sorts: (a) issues prompted by presumptions that being disabled compromises autonomy because disability impedes or warps disabled people’s understanding of the good and (b) issues prompted by presumptions that being disabled complicates freedom because disability creates conflicts between the individual’s interest and the interests of others. Further, unmasking these presumptions and the unwarranted metaphysical and epistemological suppositions on which they seem to depend is critical to our second goal: proposing procedures that facilitate abandoning or dismissing such troublesome views so as to vacate hasty judgments that disability preempts individuals from making end-​of-​life decisions of their own, both in cases in which death, anticipated but not impending, is to be hastened and in cases where dying is neither approaching nor expected at all.

Death, Dying, and the Disabled    261

Diminished Capacity and Adaptive Autonomy: Cognitive Disabilities and Deciding to Die We begin this section with a description of the kinds of cognitive disability that almost always have been taken to preclude autonomous decision-​making.1 We employ this strategy because cognitive disability often is thought to impede autonomy in a way that it appears physical disability does not, although in our view the picture is much more complex and nuanced. As courts have increasingly recognized, people with different types of cognitive disabilities are capable of participating in decision-​making in multiple ways.2 Many of the same factors that have been thought to impede participation in decision-​making by people with cognitive disabilities are also at play for people with physical disabilities or for individuals who are apparently normal. Discussions of advance directives illustrate how people with intellectual disabilities are treated as incapable of participating in decision-​making. These discussions commonly suggest that as cognitive abilities decline in old age (or due to certain progressive neurological diseases), people become unable to govern themselves. They then must be replaced as choosers for themselves by either their prior stipulations or preferences (in the exercise of “precedent autonomy”) or the substituted judgments of representatives acting for them.3 Understanding how this picture is problematic, we think, can point the way to what is also problematic about the understanding of autonomy that is prevalent with respect to people with noncognitive disabilities. Why think that people with cognitive disabilities are incapable of self-​governance? Typical answers to this question are that people with cognitive disabilities are either incapable of formulating and communicating or incapable of implementing their conceptions of their good. Which of these answers is advanced varies with the types of cognitive impairment. We examine three cases here: people with life-​long cognitive impairments, people with sudden onset impairments due to illness or injury, and people in states of decline from dementia or other similar conditions. Particularly in the case of dementia, cognitive abilities may fluctuate or may be effected by context, so that individuals may be decision-​capable under some but not other conditions. It is important to note that the view we develop about how people with cognitive disabilities can participate in the exercise of autonomy does have some limiting cases. If individuals are not responsive at all, or their responsiveness is so fragmented and inconsistent that no plausible account of their preferences and therefore no extrapolation of their own idea of their good can be formed, there really is no way for them to participate in decision-​making. For such individuals, we must look elsewhere than their own autonomy for the decisions that need to be made. Decisions may be thrown back to precedent autonomy in cases of individuals who were at one point able to express their wishes. Alternatively judgments may be based on substituted judgment efforts to extract

262    Anita Silvers and Leslie P. Francis an idea of what the individual would want from what is known about what he or she was like in the past. In the absence of any such knowledge, best interest serves as the default standard. Thus when people have cognitive impairments that render them unable to have or express any feelings or likes and dislikes, autonomy will not apply to them. People in permanent vegetative states are a primary example of this. When people have been in such states since birth, there simply is no way to observe or elicit from them the kinds of reactions and interactions that are the basis of our account of individuals’ formulating conceptions of their good presented in this article. In other cases, some people will have lost the capacity to feel or to interact to the extent that they cannot participate in formulation of their good in the way we describe. In this latter situation, any formulation of individuals’ conceptions of their good must rely on prior expressions of preference or what is known about their prior lives—​raising the issues of precedent autonomy that we touch on later. These are limiting cases, however, and it is important to understand the assumptions that have served to preclude the attribution of autonomy to many. Consider first people with severe to profound cognitive disabilities beginning very early in life. Such people typically are judged incapable of reliably formulating conceptions of their good or of exercising the executive capability needed to implement these conceptions with any success. In such cases, nondisabled representatives are authorized to decide what outcomes are in the disabled person’s best interests and to act accordingly. On this account, others construct cognitively impaired persons’ conceptions of the good from what is known about them, or about humans generally, and act as substitute decision-​makers.4 Ideally substitute decision-​makers would judge impersonally, completely free of the influences or biases of their individualized or idiosyncratic experiences and preferences, but whether such detached judgment is possible even for the most meticulous of replacement decision-​makers is unclear. In this section we suggest that the standard approach does not suffice; if people with cognitive disabilities have sufficient capacity to interact and can express preferences in some way, decision-​making need not and should not fully substitute the judgment of others for them. Instead, they can and should participate in formulating their own good. If they do not have sufficient capacities, as in the limiting case referenced earlier, then the only possible standard for decision-​making is the best interests standard. People in the process of cognitive decline from conditions such as dementia may be seen as having fading, fluctuating, or vanishing capacities and as a result having diminishing or no capabilities for self-​governance. In contrast to those with severe life-​long cognitive disabilities, however, such persons are typically seen as having prior conceptions of their good; the issue in decision-​making for them is then judged to be the relevance of these prior conceptions to their present state. A “precedent autonomy” view employs these prior conceptions to direct present choices, but whether precedent autonomy should be judged as present autonomy is complex. Depending on the account of the good postulated, persons’ former views may be taken as their good now, serving as the exercise of present autonomy. Or they may be seen as having fractured identities with perhaps newly emerging preferences and

Death, Dying, and the Disabled    263 choices. For example, persons with dementia frequently are reported to become absorbed in the pleasures of watching television, sitting in the sunshine, or singing simple lyrics repetitively, despite having disparaged such passive pastimes prior to their cognitive decline. On this view, precedent preferences and choices are alien to the individual’s present state, cannot be regarded as the exercise of autonomy for them, and therefore do not authorize the withdrawal of life support or other life-​ ending action. Autonomy after traumatic neurological injury or sudden neurological illness may be addressed on either or both of the precedent autonomy and substituted judgment accounts. After profound traumatic injury, people may appear psychologically discontinuous with their former selves. To the extent that they do have capacities, they may be seen as reemerging as new persons with different identities and conceptions of their good, much as amnesiacs do. Depending on the extent to which they are seen as currently subject to their own prior conceptions of their good, they may be seen as candidates for precedent autonomy—​or as persons with fractured identities for whom the good has changed. On our view, these standard analyses of life-​long cognitive disability and the precedent autonomy and fractured identity pictures of later cognitive disability rest on problematic presumptions about how individuals formulate conceptions of their good. Underlying presumed views about forming ideas of the good are, inescapably, metaphysical commitments about the nature of persons and epistemological suppositions about the nature of people’s knowledge of another person’s good. We have emphasized our approach in previous writings on this subject and continue to do so here: that is, individuals, whether or not they are disabled, formulate their conceptions of their good interactively with other people and do so in the context of social scripts. Thus, we contend, each person’s conception of the good cannot be just a matter of that individual’s mental state and capacity to choose. It is a mistake to think that because an individual cannot fully formulate a conception of the good independently from others, the only options are precedent autonomy, substituted judgment, or the application of the best interests standard. Individualist assumptions about formulating the good are at the heart of the problem. Agnieszka Jaworska’s well-​ known work on valuing and people with dementia attempts to expand the range of persons who can be said to develop their accounts of the good. However, even her view is illustrative of how individualistic presumptions infect accounts of people formulating their conceptions of their good. Jaworska attempts to cut through the sharp dichotomy between precedent autonomy and present preferences that we sketched earlier by developing an account of how people with dementia can be understood as valuing and choosing. Nonetheless, her approach continues to locate valuing solely in the individual’s mental life. Posing the question of decision-​making in cases of dementia, she writes: “The quandary, in a nutshell, is this: Should we, in our efforts to best respect a patient with dementia, give priority to the preferences and attitudes this person held before becoming demented, or should we follow the person’s present preferences?”5

264    Anita Silvers and Leslie P. Francis Jaworska continues by describing two antithetical answers as sharing center stage in the prevailing literature: either listen to the patient’s current preferences or follow the dictates of precedent autonomy as expressing his or her prior preferences. Questioning the authority of precedent autonomy, she says, if a caregiver “is faced with a person—​ or if not a fully constituted person, at least a conscious being capable of pleasure and pain—​who, here and now, makes a claim on the caregiver to fulfill her needs and desires; why ignore these needs and desires in the name of values that are now extinct?” (p. 108) But to put the point in this way begs questions about what it would mean for prior values now to be extinct for a person. Reasons a prior value might be thought now-​extinct could be that it no longer is articulated or otherwise reflected in the person’s conduct and therefore is presumed to no longer be a part of that person’s experienced mental life. These reasons, however, presuppose a questionable individualist account of persons and their good and therefore do not fully resolve questions about the permissibility of life-​ ending action. Jaworska’s own approach emphasizes what it means for a value to exist for a person but does so in a manner that preserves problematic assumptions about the role of individualized mental life in valuing.6 To develop her account of how persons with cognitive disabilities can be said to value, Jaworska turns to Ronald Dworkin’s distinction between experiential and critical interests, a distinction frequently relied on in discussions of end-​of-​life decision-​making.7 Experiential interests perforce require present experiences; thus a person with dementia who finds flowers pretty and pleasant has an experiential interest in looking at flowers. “Critical” interests, however, involve the occurrence of events outside the person’s present experiences and reflect whether life is going well for the individual overall: for example, whether a person’s children are happy or her projects are realized. As people lose cognitive capacities, Jaworska points out, critical interests may remain without the individual’s being able to recognize them or at least to refer to them. In such cases, she says, individuals are not themselves valuing, although in some sense it could be said that their critical interests remain of value for them. On Jaworska’s account, therefore, the more complex critical interests of earlier life likely fade as dementia takes its course. People with dementia may be left with less complex interests and diminished ability to acquire new critical interests; what critical interests they remain capable of acknowledging or even acquiring may be less complex and more oriented toward the present. If so, the problem of the authority of precedent autonomy in regard to ending life is pressed by the possibility that any remaining critical interests may conflict with more complex critical interests developed earlier in life. Jaworska writes: “After all, the most likely scenario of a conflict between a demented person’s current critical interests and her critical interests from the predementia period is not one in which a person’s demented mind generates completely new critical interests, but rather one in which dementia causes a person to lose some of her earlier more complex interests, so that in the new, simpler configuration the remaining interests gain import.”8 On Jaworska’s account, however, people with these more limited critical interests can still be regarded as valuing. She continues: “Critical interests may well be understood to issue

Death, Dying, and the Disabled    265 from something less grand—​simply from convictions about what is good to have, which do not require the ability to grasp or review one’s whole life.”9 As these passages illustrate, on Jaworska’s view critical interests continue to be understood in terms of the cognitive activities of the person to whom they belong. The difference between her view and Dworkin’s is that the cognitive activities she sees as requisite for valuing are less ambitious or grand. They need not involve an understanding of the shape of one’s life as a whole but only sufficient cognitive power to recognize the feel of regret and its effect on self-​esteem, as well as some sense of practical identity. The account Jaworska gives of the features of critical interests remains rooted in individual cognitive activity; indeed, the remainder of her discussion explores how findings from neuroscience support her view, such as findings about the individual mental capacities of people with dementia. In our judgment, Jaworska does get the understanding of autonomy partially right, as she says that persons may be seen as still exercising autonomy even if they are dependent on the help of others to realize their values. In other words, she countenances the role of prosthetic aids, such as someone else making nonintrusive assistance arrangements for a person with dementia who once valued being independent and still, in Jaworska’s sense, does but who cannot now govern daily life activities sufficiently to direct life fully on his or her own.10 Nonetheless, Jaworska fails to extend this analysis to the activity of formulating a conception of the good, which on our view is not solely a matter of individual mental activities. Instead, she puts the problem in narrow terms about the implementation of conceptions of the good: “They cannot be autonomous because, left to their own devices, they cannot lead their lives by their own lights. And this is largely because they have lost the ability to reason from their preferences to the appropriate decisions and actions—​they have lost the adeptness for means-​ends reasoning and planning.”11 We have defended an alternative view that conceptions of the good are not developed either independently or purely cerebrally in several articles. In “Liberalism and Individually Scripted Ideas of the Good: Meeting the Challenge of Dependent Agency,”12 we argued that many defenders of political liberalism hold views of the good that understand it as subjective, plural, and arrived at independently. We then explored how such conceptions of the good are mistaken. One crucial step in our argument was explaining how these are separate theses that need not and should not go together. So, for example, nondisabled as well as cognitively disabled individuals may differ from one another in conceptualizing of the good while not necessarily arriving at these conceptions solely on their own. Most important for our argument here is understanding how the enterprise of constructing individual ideas of the good is often carried on in dependency—​and not only when people have cognitive disabilities. There no bright lines that distinguish people with “mental retardation” and people with later-​life cognitive disabilities from people viewed as cognitively “normal,” and, more important, there are no sharp lines that separate how these groups of people formulate conceptions of their good. These conceptions are formulated in interaction with others: few people create conceptions of their good

266    Anita Silvers and Leslie P. Francis independently; instead, conceptions of individual good draw on social scripts and are worked out both supra-​and interindividually. In another article, “Thinking About the Good: Reconfiguring Liberal Metaphysics (Or Not) for People with Cognitive Disabilities,”13 we explored how the problematic presumption that understandings of the good must be arrived at independently infects liberal approaches to formulating justice. Rawlsian understandings of the construction of justice, for example, assume a basic framework of principles responsive to full cooperators; supposed needs of people with disabilities are to be met only after this framework has been constructed. These understandings are idealizations; as such, they generalize—​ but also obscure—​the variety of forms of agency and abstract from—​but by doing so also lose sight of—​the contexts of injustice within which justice is called on to emerge. In this work we drew on familiar understandings of personhood and agency. Although we cannot fully rehearse the issues here, it is important to recognize that conceptions of personhood do not necessarily tie to cognitive abilities or indeed solely to features located within individuals’ cognitive processes. To take one illustration, Hilde Lindemann recently advanced a conception of personhood consisting of four elements: sufficient mental activity to constitute a personality, bodily expression of this personality, recognition of these manifestations by others as the expression of a personality, and responses by others to what they see.14 In a similar vein, Andrew Sneddon recently defended an externalist view of moral agency.15 Sneddon takes a “wide” view of cognitive processing on which the environment plays a constitutive role. External features of the agent’s environment thus function interactively in action production. Sneddon is agnostic about whether to accept a desire-​belief model of action, instead pressing the reinterpretation of what it means to, on externalist accounts, have a desire or a belief. On views such as these, what it is to be a person or even a moral agent is interactive; the individual is an essential part of the interactive process but not the whole of it. Moreover, individuals themselves do not necessarily bear an epistemologically privileged position with respect to conceptualizing their good, as doing so is not a matter of introspection. Working out conceptions of the good does not depend solely or even primarily on the individual’s own mental capabilities; thus even individuals with diminished cognitive capabilities may participate in development and articulation of their conceptions of their good. Recent developments in international and US law illustrate in practice how agency can be understood as interactive and dependent. The recent United Nations Convention on the Rights of Persons with Disabilities (CRPD)16 presents supported decision-​ making (some forms of which we have analyzed by construing other peoples’ supportive roles as analogous to prosthetic supplements facilitate disabled persons’ own cognitive function) as a model for realizing effective choices by people with disabilities, including those with cognitive disabilities. The critical provision of the CRPD is the Article 12 provision for the recognition of full legal capacity for people with disabilities. In Article 12, the Convention provides that all “States Parties shall take appropriate measures to provide access by persons with disabilities to the support they may require in exercising

Death, Dying, and the Disabled    267 their legal capacity.”17 An important goal of the CRPD is to implement a civil rights model for prohibiting discrimination, in contrast to a medical model of treatment or a welfarist model of meeting needs.18 Such a civil rights approach calls for social frameworks within which people with disabilities can—​with any necessary reasonable and effective supports—​exercise political rights and work out how to live lives that are good for them. Supported decision-​making is a method for giving enhanced legal authority for jointly made decisions. It is voluntary, can be revoked at any time, and does not remove legal capacity. Models for supported decision-​making have been developed in a number of jurisdictions. One example is British Columbia’s Representation Agreement, a contract authorizing a third party to act on someone’s behalf for a range of personal decisions under which the maker retains legal capacity, can revoke the agreement, and must guide the representative in regard to choice of ends.19 The expectation is that such representation is based on regular consultation between the representative and the person; with regard to health care, the representative is permitted to participate only in a decision to end life-​sustaining treatment if the representation agreement gives him or her the right to participate in health-​care decision-​making. Another model is Sweden’s “god man” or mentor; appointment of a mentor does not affect the legal capacity of the recipient, although the mentor can be appointed without the recipient’s consent. Mentors are appointed “If, because of sickness, mental disorder, a weakened state of health or the like, a person needs assistance in safeguarding his rights, administering his property or providing for his needs.”20 In a recent Virginia case, a trial court adopted supported decision-​making for Jenny Hatch, a person with Down syndrome who chose to live independently and objected to her parents’ attempt to impose guardianship on her.21 On this model for deciding her good, Hatch chose, and the court affirmed, outside advisors. Her parents were the most likely candidates for decision-​makers whose choices could be substituted for her own had the judge resolved the case differently. In contrast, her court-​appointed advisers are people with whom she consults; she engages with them in developing an alternative idea of a good way to live her life.22 She chooses where she resides and what life goals she should pursue, with a support system to assist with the instrumental choices required to implement major decisions about how she conducts her daily life.23 As her rejection of her parents’ decision that she live in a group home and labor in a sheltered workshop, rather than live with friends and earn regular wages, and ultimately of their guardianship, shows, Hatch’s conceptualization of the good diverges markedly from the ideas her parents have had for her. Not so long ago, people with Down syndrome and cognitive differences of similar kind and degree would have been denied any voice, let alone the lead voice, in determinations of their destinies. Yet an often-​unrecognized impact of the disability rights movement that expanded so influentially in the last half of the twentieth century has been to extend recognition of personhood to individuals with cognitive disabilities. The CRPD, for example, insists that all persons with disabilities have the right to recognition before the law and are to enjoy legal capacity on an equal basis with others in all aspects

268    Anita Silvers and Leslie P. Francis of life. States parties must take “appropriate measures” to provide access to supports people need in exercising this legal capacity.24 The major question for supported decision-​making is how the practice can ensure that decisions are truly the person’s own, as they must be given the practice’s goal.25 In regard to strategies for implementation, Salzman suggests that individuals must retain legal capacity, the relationship must be freely entered into, there must be active participation in decision-​making, and decisions made with support must be legally enforceable. However, supported decision-​making has not yet been subject to systematic empirical study; there is little evidence about supported decision-​making processes with, and the effects of these relationships on, people with severely limited cognitive capabilities or the quality of resulting decisions. Among other issues, Kohn, Blumenthal, and Campbell identify socially imposed challenges faced by persons with cognitive disabilities that make it difficult for them to engage in analytic tasks: stereotyping and other circumstances that deprive them of opportunities to practice decision-​making skills.26 As empirical questions for investigation, these authors suggest whether and under what circumstances supported arrangements can become as coercive as substituted decision-​making may be and whether supportive arrangements results in different decisions than surrogate arrangements do. Much remains to be learned about how supported decision-​making can work in practice, especially in the context of end-​of-​life decision-​making, but it is an approach that captures the involvement of cognitively disabled individuals in relational determinations of their good. Two critical lessons are to be drawn from the relational autonomy and the CRPD approaches. The first is that, although autonomy is an attribute of individuals or of their singular conduct, autonomous choice need not and usually does not occur in isolation. The second is that any exercise of autonomy, whether by disabled or nondisabled people, requires appropriate political frameworks, which in turn shape legal status and processes, including those that govern the practice or prohibition of life-​ending intervention. We have illustrated these points in discussing cognitive disability, but they apply equally well to physical disability and indeed to decision-​making by nondisabled people as well.

Thwarted Capacity and Expropriated Autonomy: Noncognitive Disabilities and Deciding to Die Disability inflects experiences of living, such as the modes of functioning a person must adopt, the social positioning the person is assigned, and the quality of life the person can enjoy. Being disabled can influence how we die as well. As we have seen, individuals who have cognitive disabilities—​whether congenitally or from trauma, dementia,

Death, Dying, and the Disabled    269 psychiatric or other illnesses, or injuries—​may be judged deficient both in understanding and executive function and thereby be considered insufficiently autonomous to be accorded the right to choose death, which is an element of liberty for those who enjoy sovereignty over themselves. Our exploration of the ways disability may affect dying thus has focused so far on how the reduced capacity for rational self-​governance has been associated with cognitive disability. But disabilities that are not cognitive also have been advanced as reasons to deny the usual freedom to make decisions for one’s self about one’s end of life.27 Some arguments in this vein contend sweepingly that disabilities and illnesses warp reasoning about dying, despite the disabled individual’s well-​balanced thinking in other respects. Other arguments insist that reasonable assessments of their own situations notwithstanding, individuals with disabilities have special obligations that restrict how they may approach their own lives’ ends. Both versions of the objection to affording autonomy suppose that disabled people’s troubled relationship with nondisabled society is liable to warp their ideas of their good. To illustrate, consider the dismissive analysis by disability activists of the thinking of Elizabeth Bouvia, a young woman with extensive dysfunction from cerebral palsy, dependent on others to execute essential activities of daily living and suffering severe pain from arthritis as well. The child of divorced parents who could not care for her, she had been institutionalized at a facility for crippled children from age ten to eighteen, had lived independently through college, and had experienced a brief marriage but had no close personal ties at the time she abandoned her attempt to live independently. Bouvia, who had completed some of the requirements for a master’s degree in social work but was told she would never be hired to do this work, ceased her efforts at independent living and was voluntarily admitted to a California hospital in 1983. She stated that she did not wish to receive any nourishment and preferred to die. When violent nausea made spoon-​feeding ineffective, the hospital administered artificial nutrition, holding her down and inserting a nasogastric tube for each feeding. Bouvia then sought legal representation and became a client of the American Civil Liberties Union (ACLU), which pursued a lawsuit against the hospital to free her from any treatment to which she did not consent. During the three years of litigation that followed, Bouvia received artificial nutrition and was constantly on a morphine drip for arthritis pain.28 After a loss at trial, where the court decided that the state had an interest in saving her from suicide, the appellate court reversed, ruling that Bouvia, as a competent patient, had a fundamental right to refuse any treatment even at the cost of her life. The appellate court judged that Bouvia met the standard of mental competence because she understood the consequences of refusing tube feeding. Further, the court declared that the patient’s interests and desires were key to decision-​making, outweighing the state’s interest in maintaining individual citizens’ lives, discouraging suicide, enabling medical professionals to adhere to the standards of their profession, and, more generally, protecting the general population against medical skills being used to harm rather than help. Finally, the appellate court noted that, despite the appearance that Bouvia had an overriding interest in living for the two decades projected as feasible with

270    Anita Silvers and Leslie P. Francis artificial feeding, the prospect of suffering a painful and isolated life suggested that in her case the seeming interest in staying alive was illusory.29 The Bouvia appellate decision thus invoked a troublesome quality-​of-​life standard, one that resulted in Bouvia’s being cast as an adversary by disability organizations.30 The appellate court described Bouvia’s life in the following terms: “[s]‌he herself is imprisoned and must lie physically helpless subject to the ignominy, embarrassment, humiliation, and dehumanizing aspects created by her helplessness.” As Bouvia was the sufferer, the court opined, she possessed epistemic privilege through direct access to her experience of her own life and therefore her standpoint should prevail.31 Disability activists saw Bouvia differently. Bouvia’s self-​reports about her own situation seemed suspiciously stereotypical to them. They believed Bouvia was being misrepresented as desiring dying by starvation, for they thought it would be simpler for her to ingest sufficient sleeping pills to do the deed. Although she could have received the maximum state payment for In Home Support Services—​about $800 a month or about eight hours a day at the 1984 minimum wage of $3.35 an hour—​disability activists suspected that she had entered the hospital because “she wanted as little assistance at home as possible, that she didn’t want other people around, helping her.”32 They diagnosed Bouvia as having internalized the rejection and repression that society rains on disabled people to obscure them and thereby distance fear of becoming disabled themselves from nondisabled people’s minds. Finally, Bouvia was accused of having embraced nondisabled people’s aspersions that she could not live independently, could not have a marriage or career, could not cope with the multiple hardships of disabled life, and therefore that she—​as a person who had to live such an absent of high-​quality life—​was worthless.33 Bouvia released a statement asking the activists who criticized her choice to accede that, nonetheless, it was hers to make. She acknowledged that “the disabled community is uncomfortable with my choice because of their fear that my decision may have a negative impact on the future decisions of other disabled persons.” But, she contended, “all people, whether or not disabled, should be free to determine their own future—​ personally, privately and individually.”34 In response, her disability community critics dismissed her statement as being the product of her manipulation by nondisabled people, for it seemed to them too much an echo of sophisticated right-​to-​die arguments. Packed into the disability opposition’s arguments were three reasons meant to countermand Bouvia’s authority for decision-​ making by contending that she could not be an individual who has sovereignty over herself. All three of the reasons deprecate her status as a free agent, not by reason of diminished capacity but on moral or political grounds. Two of the reasons, both rooted in hypotheses about nondisabled people’s antipathy toward the disabled, were invoked to characterize Bouvia’s motivation as disingenuous and therefore morally flawed. The first diagnosed her as suffering from pathological self-​ hatred as a result of emotional injury caused by her chronic experiences of nondisabled people’s rejection and disregard. This kind of “happy slave” objection35 is commonplace in discourses on the moral wrongness of slavery, where counterexemplary testimony from individuals content to be slaves is advanced to defend the practice.

Death, Dying, and the Disabled    271 As the standard response to so-​called happy slaves goes, persecution can induce victims to adopt their oppressor’s devaluation of themselves, but were the conditions of injustice with which they constantly contend no longer to hold, they would never embrace such negative self-​assessments. Consequently, their self-​referential motivation where their own well-​being is concerned must be suspect. Applying such a strategy to Bouvia’s refusal of nutrition, disability activists condemned her motivation as morally flawed, a product of hatred she held for herself. This feeling corrupted her expressions about the good for herself, they supposed, for her judgment on that subject was rendered an inauthentic voice lacking competence to represent anyone with a disability. Second, Bouvia’s critics contended that Bouvia’s dependence on nondisabled society for services and to mitigate her loneliness exposed her to being exploited by nondisabled people for their own purposes. Her assessment of her quality of life as insupportable did not reflect that of a self-​confident and proud quadriplegic but appeared instead to voice ideas introduced by ACLU attorneys who meant to make her a tool to show that disability made life expendable. Here again, Bouvia was portrayed as deficient in moral agency, in this case because her disability was construed as rendering her merely a defenseless means for nondisabled people to achieve their own ends. Finally, disability activists advanced their denial of Bouvia’s right to reject tube feeding on political grounds. Her legal pursuit to be free of life-​saving medical intervention could influence how other individuals with disabilities were treated and thus could have a negative impact on the fates of others like herself, especially those who depended on medical support to survive, for her expressed idea of the good for herself might be thought representative of those who shared her group identity. Because her idea of her good did not align—​and even threatened—​others’ idea of the good for their group, other members proposed to expropriate her autonomy. In sum, to discredit Bouvia’s decision that death was in her ultimate interest, disability activists derogated her capacity for self-​governance. Not accidently, the critique they mounted to dispute her capacity to decide for herself echoed the broader society’s customary deprecation of disabled people generally. The characterization of Bouvia as lacking morally sufficient agency reproduced a prevailing stereotype of the era that equated having a disability with being morally flawed, for it was commonplace in that time for nondisabled people to picture disabled individuals who retained executive functioning as being driven by anger and shame.36 Further, construing Bouvia as being obligated to reject her right to freedom from medical intervention reiterated the prevailing paternalistic policy of the era, which did not hesitate to attenuate disabled people’s liberty in the interest of a larger group. As an illustration of this policy, in City of Cleburne, Texas, et al. v. Cleburne Living Center et al. (473 U.S. 432 [1985]), the US Supreme Court characterized cognitively disabled people as a naturally inferior group who “have a reduced ability to cope with and function in the everyday world” and justified deflecting equal protection scrutiny of laws and programs that disadvantaged them in virtue of there being supposed safeguards for this group in singling them out for special treatment.37 Bouvia’s disagreement with her disability critics initiated in a divergence about whether she could live happily: they insisted

272    Anita Silvers and Leslie P. Francis she could be happy by living independently; she believed she had tried to do so with all effort she could muster and it had not worked. Both lines of argument deployed by Bouvia’s critics in the effort to cast her self-​ determination as suspect have analogues in familiar challenges to cognitively disabled people’s capacities to decide by themselves about themselves. For the Bouvia critique, however, incapacity in noncognitively disabled people was traced not to their bad biology’s rendering them incapable of exercising sovereignty over themselves but instead to a defectively oppressive cultural and society climate that had the same effect. Having had her right to determine her dying affirmed on appeal, Bouvia did not proceed to die.38 A similar outcome occurred in the case of Larry McAfee, a ventilator-​ dependent individual with quadriplegia, who was shifted back and forth from hospital to nursing home and in the process was also moved from state to state. Although initially covered for independent living by his own insurance plan, his care rapidly reached the plan’s total cost cap; subsequently, he lost nursing home care when funding under the Medicare Catastrophic Coverage Act ceased with Congress’s decision to eliminate the program. Neither Medicaid nor Medicare paid for care in the community, despite the substantial saving over hospitalization.39 Contending that he did not wish to live with “no control over what’s done to me, how it’s done or by whom,” McAfee sued to enable installation of a device that would enable him to turn off his ventilator himself, thereby relieving himself of dependence on another person’s becoming complicit in his death. (Prior to the motorcycle accident that resulted in his quadriplegia, he had been an engineer.) A Georgia Superior Court judge decided in his favor, and the Georgia Supreme Court affirmed.40 McAfee did not need to turn off the ventilator, however; as a result of his taking legal action that would let him die, he eventually gained access to adaptive equipment to facilitate his managing his own daily life and was also able to move back from the hospital to a community home. Like Bouvia, McAfee sought to end his life by rejecting medical care essential to maintaining it. Bouvia wanted to be free of artificial nutrition; McAfee wanted freedom from artificial respiration. Unlike Bouvia, however, McAfee clearly blamed remediable social circumstances rather than an immitigable biological condition for the intolerable quality of his life. His decision-​making therefore had a different character. Whereas, at least on the disability critique of her thinking, Bouvia’s choice was an emotional rather than a rational result, the outcome of socially induced self-​loathing, McAfee seems to have been engineering a rational plan to rain reproach on a system that failed to fund an endurable living situation for himself. He acted to shame society, not because of shame about self. Had he not enjoyed sovereignty over self, he could not have executed the strategy that raised his public profile sufficiently noticeably for better prospects to become apparent. Granted from a moral point of view he should not have had to face such a risky and extreme choice because of being an individual with a disability, being a person with a disability did not make him incapable of choosing. McAfee had lost the means to execute other liberties due to his disability, leaving deciding when to die as one of the few freedoms available to him. But such narrowing of practicable

Death, Dying, and the Disabled    273 freedom does not mean that a person’s decisions are nonautonomous. It only means that the person’s options are very poor. Of course not every disabled individual who seeks and obtains the freedom to choose to die eventually decides, like Bouvia and McAfee, not to use it. Consider the controversy over the decision of Tim Bowers, who fell sixteen feet and received a high (C-​3 to C-​5) cervical spinal injury that left him paralyzed from the shoulders down and ventilator dependent. The day after his injury, his family requested that sedation be discontinued until he could make a decision about whether to live or die. He chose the latter course, was extubated, and died. Should someone so new to disability be deemed capable of deciding when to die? If not, what about being unused to disability compromises the capacity to do so?41 More than 700 readers expressed judgments about this case. Some commended the result, stating that Bowers, like Bouvia, had a duty to others that overrode any inclination to secure personal desire. Unlike in the Bouvia case, however, Bowers’s commentators thought he had done the right thing by dying rather than living on as a burden on his new wife and imminent child. Many more were disturbed by the haste of Bowers’s decision, as well as the unfavorable conditions under which he reached it. Some declared that they, or people they knew, were living long and well after severe spinal cord injury. Many others opined that Bowers had discounted the possibility of positive outcomes too greatly. They talked about new techniques for healing and the possibility that he could walk again. Some were insistent that he should have waited until after his baby was born. One even proposed that, having lost sensation from the neck down, he nevertheless could compensate for being unable to hold his child in his arms by having the baby lifted up and pressed against his cheek. But the meagerness of satisfaction to be gleaned from such achievements is not reassuring. Furthermore, as Bowers emerged from sedation, his sister, a nurse, had delivered a dire prognosis about the details of his future life. Undoubtedly sudden disability does cramp choice in cases like this, but the source of the constraint comes not from the disabling condition but from the swiftness of its onset. Bowers is described as being a passionate outdoorsman; the experiences of these pursuits looms large in how he envisions the good for himself, we may suppose. Told by his sister that his new future precludes such pleasures presumably stripped away a considerable chunk of the good he had envisioned, an absence that is not immediately replaceable. Developing one’s idea of the good is not, we have pointed out, a solitary process but rather is one in which social interaction must interject at crucial points not only to enrich the emerging conceptualization but to probe and test it. It is implausible to hold that, still on the edge of such a traumatic life-​changing event, an individual can have fully formed a forceful substitute notion; this will be true whether becoming disabled, losing a loved one, having made a catastrophic mistake, or having suffered some other catastrophic loss is the source of the trauma. And of course anyone’s autonomous decision-​making will be hampered—​indeed crippled—​by having nothing but an ill-​ formed or fragmented notion of one’s good. Here, however, it is a disrupted life plan and

274    Anita Silvers and Leslie P. Francis the ensuing confusion and rush to judgment that attenuates autonomy, not difficulties attendant on disability itself.

Supporting Decisions About Dying: Disability and Nondisability Alike Our last observation points the way to discerning how having a disability may affect the autonomy of decision-​making at the end of life. Our analysis of prominent cases in which people with disabilities decided to live their own lives, or to die in their own way for their own reasons, reveals that disability reduces the range of options from which an individual can choose. In some instances the nature of the disability itself delimits options; in others, other peoples’ or social policy responses to the disability create barriers that curb choice. Nevertheless, having an impoverished or undesirable set of alternatives does not equate to having no power of choice at all. Indeed, we not only acknowledge but also admire individuals who exercise their capacity as choosers to make the best of bad situations. Our discussion of misappropriated or thwarted autonomy revealed that the tension disability occasions for liberty arises not from an absence of capacity (whether direct or supported) but rather from alien-​seeming, and thereupon alienating, ideas of the good crafted by circumstances most people cannot imagine themselves having to endure. This misalignment between the conditions shaping the majority of people’s conceptualizations of their good and those implacably affecting disabled people’s ideas calls for careful practice when critical and presumably irreversible decisions about health care are at stake. Although the cases we have selected to illustrate our discussion all are instances in which the disabled person could express a choice about whether to die, this last point about nondisabled people’s limitations in formulating authentic ideas of the good from disabled people’s points of view should abridge nondisabled family members’ and medical professionals’ power to substitute their own inclinations when the patient cannot communicate at all. The following are some prescriptions for fair practices that emerged from our analysis. First, as the circumstances of Bowers’s case illustrate, such decisions should be deliberative, not snap judgments. This consideration is no more than the Humean thought that important decisions should be made in a cool moment, not in the heat of immediate distress or pressure. They should, of course, also be informed by reflection on more than superficial or hasty comprehension of the full complement of facts. Possibly a waiting period such as prescribed in state policies permitting physicians to assist terminal patients’ suicides might be appropriate. However, any waiting period must be structured in such a way as to be tolerable, with adequate care and interaction during the period. Moreover, any waiting period must not predetermine the outcome,

Death, Dying, and the Disabled    275 as when the imposition of a waiting period in effect precludes the discontinuation of a care modality such as a ventilator that is no longer needed at the end of the period. Second, adequate support for decision-​making must be available. This may include help in gathering or understanding information, arranging living conditions or other care needs, articulating or understanding goals, and linking possible alternatives to goals thus articulated. The prosthetic components for successful support may require knowledge and skills well beyond the usual warnings about medical probabilities and risks. Third, as is especially salient in the dispositions of the Bouvia and McAfee cases, a process for identifying and resources for ameliorating intolerable conditions injustices that needlessly narrow the disabled person’s options should be introduced. When people such as Jenny Hatch face the prospect of lives that, according to their view, are useless and isolated, the choices that result from the exercise of their autonomy understandably may seem distorted and defective to the more favored majority whose options are not so confined. Finally, these are not matters of concern for people with disabilities alone. For anyone facing the realities of mortality or morbidity, the exercise of autonomy is centrally linked to socially supported opportunities for reflection, collaboration, and securing justice.

Notes 1. Many types and degrees of cognitive disability are relevant to this section. These include accidents, sudden illness, and lingering declines. They include insults to particular areas of the brain, such as the impact of a frontal lobe injury on emotion or an injury to the hippocampus on memory. Cognitive deficits also may not be uniform: persons with dementia, for example, may lose more recent memories before long-​term memories. Or persons with significant intellectual limits may have remarkable memories for and responses to music. There are limiting cases to our analysis when precedent autonomy or else substituted judgment of best interest are the alternatives. The trend over the past quarter-​century, however, has been for courts to acknowledge the capacity of individuals with intellectual disabilities. See, e.g., Stanley S. Herr, Lawrence O. Gostin, and Harold Hongju Koh, eds., Human Rights of Persons with Intellectual Disabilities: Different but Equal (Oxford: Oxford University Press, 2003). 2. http://​jennyhatchproject.info. 3. For example, John Davis, “The Concept of Precedent Autonomy,” Bioethics 16(2) (2002): 114–​133. 4. The terminology here has also been confused. As it was presented in courts in the United States, the “substituted judgment” standard involved efforts to reconstruct what the person would have wanted from what was known about his or her prior preferences, interests, or life circumstances. The New Jersey courts, primary developers of this account, characterized it as “subjective” because it was tailored to the individual subject. There was something of a legal fiction here: the substitute decision-​maker was thought of as actually exercising autonomy on behalf of the individual, but the individual was not engaged in the choice in any manner. When insufficient information was available about the individual’s prior preferences, these courts adopted the “best interests” standard, relying on

276    Anita Silvers and Leslie P. Francis the interests of human beings generally; New Jersey courts called this an “objective” standard. These courts also recognized the possibility of cases in which there would be some information available about a person’s prior life but insufficient information to reconstruct a choice for them. For such cases, the courts adopted a “limited objective” standard, relying on best interests unless there was clear evidence (such as expressions of pain or strong dislike) that the individual would desire otherwise. See In Matter of Farrell, 529 A.2d 404 (NJ 1987); Matter of Jobes, 529 A.2d 434 (NJ 1987); Matter of Peter, 529 A.2d 419 (NJ 1987); Matter of Conroy, 486 A.2d 1209 (NJ 1985). Other courts, however, applied “substituted judgment” whenever someone else acted for the individual (e.g., Matter of Storar, 420 N.E.2d 64 [NY 1981]). An additional complication is whether the individual selected the stand-​in decision-​maker who is then characterized as a “proxy” or whether the decision-​ maker was selected by some other process such as a list of relatives or friends by proximity. Yet another complication about the terminology is that, in the context of the 2008 United Nations Convention on the Rights of Persons with Disabilities, “substituted judgment” is used to refer to problematic practice that may not comply with Article 12 on disabled people’s right to legal standing. See http://​www.alrc.gov.au/​publications/​2-​conceptual-​ landscape%E2%80%94-​context-​reform/​supported-​and-​substituted-​decision-​making. 5. Agnieszka Jaworska, “Respecting the Margins of Agency: Alzheimer’s Patients and the Capacity to Value,” Philosophy and Public Affairs 28(2) (2006): 105–​138. 6. A parallel problem can be discerned in Jaworska’s account of the moral standing of people with cognitive disabilities. Jaworska writes: “Both proponents and opponents of the commonsense view of [full moral standing] need an account of the mental properties that ultimately justify these special moral requirements” (463). Her own account of full moral standing is found in the capacity for caring; however, why should an account of moral status be based solely on mental properties? Agnieszka Jaworska, “Caring and Full Moral Standing,” Ethics 117(1) (2007): 460–​497. 7. Ronald Dworkin, Life’s Dominion:  An Argument about Abortion, Euthanasia, and Individual Freedom (New York: Alfred A. Knopf, 1993), especially 201ff. Here is Dworkin’s account of a “critical interest”: “Most people think that they also have what I shall call critical interests: interests that it does make their life genuinely better to satisfy, interests they would be mistaken, and genuinely worse off, if they did not recognize. Convictions about what helps to make a life good on the whole are convictions about those more important interests. They represent critical judgments rather than just experiential preferences.” (p. 201) Note the role of individual cognition in this account. 8. Jaworska, “Respecting the Margins,” 112. 9. Jaworska, “Respecting the Margins,” 113. 10. Jaworska, “Respecting the Margins,” 126. 11. Jaworska, “Respecting the Margins,” 128. 12. Leslie P. Francis and Anita Silvers, “Liberalism and Individually Scripted Ideas of the Good:  Meeting the Challenge of Dependent Agency,” Social Theory and Practice 33(2) (2007): 311–​334. 13. Anita Silvers and Leslie P. Francis, “Thinking About the Good:  Reconfiguring Liberal Metaphysics (Or Not) for People with Cognitive Disabilities,” Metaphilosophy 40(3-​4) (2009): 475–​498. 14. Hilde Lindemann, Holding and Letting Go (New York: Oxford University Press, 2014), ix. 15. Andrew Sneddon, Like-​Minded: Externalism and Moral Psychology (Cambridge, MA: MIT Press, 2011), especially ch. 5.

Death, Dying, and the Disabled    277 16. United Nations Convention on the Rights of Persons with Disabilities (CRPD), http://​ www.un.org/​disabilities/​convention/​conventionfull.shtml. 17. CRPD, Article 12.3. 18. For example, Arlene Kanter, “The United Nations Convention on the Rights of Persons with Disabilities and its implications for the Rights of Elderly People Under International Law,” Georgia State University Law Review 25 (2009): 527–​573. 19. British Columbia Representation Agreement Act, http://​www.bclaws.ca/​civix/​document/​ id/​complete/​statreg/​96405_​01. 20. http://​w ww.alzheimer-​europe.org/​Policy-​in-​Practice2/​C ountry-​comparisons/​L egal-​ capacity-​and-​proxy-​decision-​making/​Sweden. 21. The case is described at http://​jennyhatchproject.info/​. 22. Sarah Levis, “Jenny Hatch’s Victory,” GirlWithaCane blog, August 12, 2013, http://​www. girlwiththecane.com/​jenny-​hatch/​. 23. Susan Mizner, “Disability Is No Excuse to Deprive One of Civil Liberties,” ACLU Blog of Rights, August 5, 2013, https://​www.aclu.org/​blog/​human-​rights/​disability-​no-​excuse​deprive-​one-​civil-​liberties. 24. CRPD, Article 12. 25. Leslie Salzman, “Guardianship for Persons with Mental Illness—​A Legal & Appropriate Alternative?” Saint Louis University Journal of Health Law & Policy 4 (2011):  279–​329, especially 306. 26. Nina Kohn, Jeremy Blumenthal, and Amy Campbell, “Supported Decision-​Making: A Viable Alternative to Guardianship?” Penn State Law Review 117 (2013): 1111–​1157. 27. See Katharina Heyer, “Rejecting Rights:  The Disability Critique of Physician Assisted Suicide,” in Austin Sarat, ed., Special Issue Social Movements/​Legal Possibilities, Studies in Law, Politics, and Society 54 (2011): 77–​112, for an analysis of the effects of denying a freedom right on the rhetoric of a civil rights movement. 28. Gregory Pence, Classic Cases in Medical Ethics, 3rd ed. (Boston:  McGraw Hill, 2000), 64–​65. 29. Bouvia v Superior Court, 179 Cal. App. 3d 1127, 1135–​1136, 225 Cal. Rptr. 297. (Ct. App. 1986), review denied (Cal. June 5, 1986). 30. Mary Johnson, “Right to Life, Fight To Die: The Elizabeth Bouvia Saga,” The Ragged Edge Online (January/​February 1997), from 1984 stories in The Disability Rag http://​www. broadreachtraining.com/​advocacy/​artbouvia.htm. 31. Bouvia, 170 Cal. App., 1143. 32. Johnson, “The Elizabeth Bouvia Saga.” 33. Johnson, “The Elizabeth Bouvia Saga.” 34. Johnson, “The Elizabeth Bouvia Saga.” 35. See Dan Haybron, “Happiness,” in The Stanford Encyclopedia of Philosophy, Edward N. Zalta, ed. (Stanford, CA:  Stanford University, 2011), http://​plato.stanford.edu/​archives/​ fall2011/​entries/​happiness/​. 36. P. K. Longmore, “Screening Stereotypes:  Images of Disabled People in Television and Motion Pictures,” in A. Gartner and T. Joe, eds., Images of the Disabled, Disabling Images (New York: Praeger, 1987), 65–​78. 37. See Anita Silvers and Michael Stein, “Disability, Equal Protection, and the Supreme Court:  Standing at the Crossroads of Progressive and Retrogressive Logic in Constitutional Classification,” University of Michigan Journal of Law Reform 35(1–​2) (2001–​2002): 81–​136.

278    Anita Silvers and Leslie P. Francis 38. Associated Press, “Woman Won Right to Die But Stays Alive,” Deseret News, December 16, 1993, http://​www.deseretnews.com/​article/​326259/​WOMAN-​WON-​RIGHT-​TO-​DIE-​ BUT-​STAYS-​ALIVE.html?pg=all This 1993 article describes Bouvia as remaining hospitalized but made comfortable by a morphine drip. She was eating solid food. She did not concede to being happy to have lived, however, explaining that only by living on morphine could she tolerate her existence. Parenthetically, despite her symptoms, Bouvia apparently was never diagnosed with bulimia or another eating disorder. As late as 2012, Bouvia was reported to be alive, in which case she outlived the most vocal of her disability community critics. 39. Peter Applebome, “An Angry Man Fights to Die, Then Tests Life,” Special to The New York Times (February 7, 1990), http://​www.nytimes.com/​1990/​02/​07/​us/​an-​angry-​man-​fights-​ to-​die-​then-​tests-​life.html?src=pm&pagewanted=2). 40. Georgia v. McAfee, 385 S.E.2d 651 (Ga. 1989). 41. Associated Press, “Newlywed Father-​to-​Be, 32, Chooses to Go Off Life Support and Die One Day After He Fell from Tree and Became Paralyzed,” Daily Mail (November 5, 2013), http://​www.dailymail.co.uk/​news/​article-​2487986/​Newlywed-​Tim-​Bowers-​chooses-​life-​ support-​die-​ONE-​DAY-​fell-​tree-​paralyzed.html.

Chapter 17

T he Effect of S o c ia l M edia on En d -​of-​L i fe Decision M a k i ng Jessica Berg

The computer revolution has had an enormous effect on all aspects of the practice of medicine, from scheduling to billing to treatments to research. But with all the attention focused on electronic health records and cybermedicine, little thought has been given to the role of social media in end-​of-​life decision making. Many people have been using e-​media for years and left significant electronic documentation of their medical treatment preferences. As social media users age, questions of how and whether to use such documented evidence to aid surrogate decision making will become more prevalent. This essay considers the ethical and legal issues involved in the use of social media for end-​of-​life decisions.

What Is Social Media? “Social media” is a term that covers multiple categories of Internet-​based social interaction tools. It includes everything from online forums to email to virtual worlds (such as Second Life). One of the most common forms of social media is the “weblog,” which ranges from online personal diaries to commentaries on world events (there are also vlogs1 or video-​logs). BlogPulse, a website that tracked blogs and allowed searches of the “blogosphere,” identified approximately 180 million blogs, with a growth-​rate of about 100,000 per day.2 Social communication tools are also a part of commonly used social media, such as email, instant messaging (e.g., AIM,3 MSN Messenger,4 Gchat5), and Twitter6, which functions both as a mechanism for instant messages and a type of short-​ form commentary on life and world events. Finally, there are social network sites (SNS).

280   Jessica Berg As this latter category has expanded, SNS have begun to incorporate blog and instant messaging features.7 SNS are “web-​based services that allow individuals to (1) construct a public or semi-​ public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) traverse their list of connections and those made by others within a system.”8 There are two primary types of SNS: those that focus on expanding social contacts and those that focus on communication within existing social networks.9 “Networking” sites are those that allow individuals to make new social connections online and may be designed to facilitate eventual face-​to-​face networking efforts. These include such sites as LinkedIn, which is a professional (job) networking community and includes approximately 135 million subscribers worldwide.10 By contrast, social “network” sites facilitate communications between an established network of friends or professionals. While these sites can be used for limited networking, information is primarily shared with an established group (although membership in that group can be fluid and may grow through outside networking efforts). Facebook is a paradigmatic example of an SNS.11 As the company notes on the website, “Facebook helps you connect and share with the people in your life.”12 According to a 2011 Pew Research Center study, two-​thirds of all U.S. adults who are online use social networks.13

What Role Might Social Media Play in End-​of-​L ife Decision Making? Social media might be relied upon in medical decision making in multiple ways. First, social media sites may provide links to online information sources for advance directives. Both the American Academy of Hospice and Palliative Medicine (aahpm.org) and PalliativeDoctors.org provide extensive resources. Second, social media can help identify appropriate surrogates.14 Pope points out that social media may be used to identify the best surrogate in cases of conflict or to locate an appropriate decision maker of an apparently “unbefriended” patient. Third, once an advance directive document (e.g., durable powers of attorney, living wills, or other instructions for health care)15 is legally executed in print format, it may be stored electronically and linked to via social media. Some states have “clearinghouses,” which allow individuals to store instructions in an online database.16 Others have more limited online systems, say for organ donation preferences.17 Recent efforts include online advance-​directive tools that utilize a web-​based decision making process and result in a legally authorized set of instructions that can be printed and/​or stored online.18 In the absence of a national databank for advance directives, patients may choose to post their legally executed documents on their Facebook page or other social network.19 The use of such electronic copies raises few new legal or ethical questions, although there may be some additional security challenges. But there also are advantages to electronic storage, since the documents may be easier to locate

Social Media and End-of-Life Decision Making    281 and update if online (or part of an electronic health records). Few people, however, complete an advance-​directive document in any format, and fewer still post such a document online. In the absence of a legally executed advance-​directive document, surrogate decision makers and health care professionals may seek to use less formal comments, instructions, or statements made on social media regarding medical treatment preferences. In some cases this may include a full advance-​directive document that simply was not legally executed. More likely, this will entail information that will not be in any legally recognizable format, such as general statements about medical care preferences, specific statements about one’s own or another’s medical situation,20 or even links to other sites that may provide an indication of preferences. How should this data be incorporated into the current legal framework for end-​of-​life decision making?

Current Framework for Decision Making All states have legislation allowing some form of medical decision making for incapacitated patients in health-​care settings, and all allow surrogate21 involvement in the majority of circumstances. The surrogate can be formally designated by the patient while competent (e.g., through a health-​care power of attorney or proxy document), formally appointed by a court (guardian or conservator), or informally identified by the health-​ care team (usually a family member). In the latter situation, a state statute may determine the applicable next-​of-​kin hierarchy. Even without such a statute, however, family decision making for incapacitated patients is well accepted. Despite the well-​established legal authority, few statutes provide detailed guidance to surrogates regarding decision making. Nonetheless, there are some prominent court cases in this area that offer some direction to both decision makers and health-​care professionals.22 Two primary standards of surrogate decision making are used: substituted judgment (appropriate when there is some evidence of what the patient would have wanted) and best interests (appropriate when the patient has never previously been competent or there is no evidence of prior wishes). The former situation is of particular relevance here. To apply the substituted judgment standard, a surrogate must consider a variety of factors to determine what choice the patient would have made had he or she been competent at the time of the decision. Evidence may include written documents, oral discussions, previous health decisions, reactions to treatment of others, general personality, religious preferences, and any other pertinent information. There is no explicit bar to using electronic evidence as opposed to paper documents; format does not necessarily have any bearing on usefulness. The key is appraisal of any specific factor of information. Each should be evaluated for specificity, thoughtfulness, consistency, remoteness, maturity, and capacity.23 How do these factors apply to information found on social media?

282   Jessica Berg Social media information can be very nonspecific and difficult to interpret. For example, 185 people have indicated they “like” a Facebook page describing the withdrawal of life-​sustaining treatment from Terri Schiavo (a high-​profile case in Florida).24 While such action provides little insight into individual preferences, one can imagine more helpful information, such as “liking” a site that advocates organ donation.25 This is not to imply that “liking” something on Facebook is determinative of patient inclination, but the information may add to the overall picture of patient preferences. By contrast, in other cases social media information has the potential to be very precise. At one point in time, Facebook indicated that five people were “talking about” the Terri Schiavo page, and thirty-​four more the organ donation page. The “talking about this” tool is a metric designed to measure the number of people over the past seven days who post on, comment, share, or mention a Facebook page.26 Detailed comments made by the individuals engaged in such activity may be extremely specific and potentially relevant for future surrogate decision making. In fact, electronic conversations have advantages regarding specificity over spoken communications. They allow words to be saved in original format (as opposed to relying on memory). Furthermore, links to other documents and websites may be included, possibly clarifying ambiguous statements and providing a fuller picture of the individual’s wishes. For example, a blog discussing the care of one’s father at the end of his life27 may include links to sample advance-​directive forms specifying treatment preferences and thus providing additional insight. Tweets may be less specific due to length limitations; Twitter allows only 140 characters per tweet. But Twitter has other advantages, such as timeliness, discussed later. Moreover, there is some initial evidence that the informal way we communicate via social media better mirrors our everyday conversations and thus may be more understandable than formal advance-​care documents. One of the most significant barriers to the use of formal advance directives is the difficulty interpreting exactly what the author means and applying the formal statements to real-​ life situations. Comments conveyed via social media, by contrast, may function closer to spoken words but with the added benefit of being recorded. One hazard with interpreting electronic communications is the lack of emotional tone. Determining the state of mind of the blogger or poster can be difficult. The advent of emoticons (emotional punctuation marks, such as smiley faces) has helped alleviate some of these problems. But even where no emotional cues are present, electronic communication does not necessarily pose more of a problem than standard written communication. Face-​to-​face communication may best capture emotional tone (telephone to a lesser extent). But this format has other weaknesses, including the lack of documentation. To use past spoken information to make patient decisions, we must rely on the accuracy of the memory of the person conveying the information. Much has already been written about the deficiencies of human memory, and I will not reiterate the arguments here. Suffice it to say that memory is less a recording mechanism than a fluid, ever-​changing instrument of interpretation—​raising significant questions about the validity of using remembered conversations in making treatment care decisions for incompetent patients.

Social Media and End-of-Life Decision Making    283 Thoughtfulness also varies considerably between different types of social media. Tweets are designed to be “instantaneous” (or at least occur within a short time frame after the event), although they may still allow more time for reflection than spoken words. They have the advantage of being a contemporaneous response to a stimulus. But they have the corresponding disadvantage of possibly being a “knee-​jerk” reaction. Instant messaging has similar problems. But the same is true of spoken words. Little, if any, evidence supports the notion that people are more likely to blurt out things they regret in Twitter or instant messaging than in spoken conversations (although they may regret the staying power of the electronic words). Other forms of social media may be especially thoughtful. Before posting a blog entry, the individual may take the time to edit and reflect on it. Even when posted, the entry can be changed or removed. Likewise, Facebook can be constantly updated and altered. Evaluating the information for consistency may alleviate some concerns in this area, giving little weight to single random comments. In addition, contextual clues can be used to determine the meaningfulness of the information in the same way they may be used to determine the weight of remembered oral conversations or the thoughtfulness of printed documents. Remoteness, or length of time between the creation of the information and the point of decision making, is another crucial factor. Information on the Internet rarely disappears and can be undated. While this may also be true of written documentation, we may tend to give more weight to information found in an electronic format (since it does not appear old) than that found written in an old diary or letter. But unlike a paper-​based diary, individuals may regularly update their electronic media, adding to or deleting an irrelevant posting, thus giving weight to the idea that the information remains pertinent (at least in the mind of the individual who chooses to keep it visible). Moreover, blogs and even Tweets are usually dated; in fact, Twitter includes an automatic expiration date for postings. Similarly, determining the maturity and capacity of the individual at the time the information was created may be easier for electronic evidence with date stamps. Not only will the date stamp identify the age when the comments were written, but surrounding comments and postings may highlight concerns about capacity. Determining the capacity of the individual at the time an advance-​directive document was created can be difficult. Determining capacity at the time of a past spoken comment is particularly challenging. These three elements—​remoteness, maturity, and capacity—​hint at a fundamental concern in the use of advance directives more generally: How should previously expressed preferences guide medical decision making? Medical treatment choices can change based on life experiences.28 Some suggest that the currently ill individual could be considered a “different” person compared to the individual at the time the preferences were first articulated.29 These fundamental challenges to advance decision making are a concern for all types of evidence, forcing us to consider the actual relevance of past information for current decision making. Resolving this debate is beyond the scope of this article. For our purposes is it important only to note that electronically stored information is no less relevant than printed information.

284   Jessica Berg In summary, social media information is not necessarily inferior to paper documentation or to spoken words. In fact, it may have some distinct advantages over these other forms of information since it can be easily updated, potentially tracked over time, and, perhaps most important, easily found. One can call up and display a Facebook page in a matter of minutes on a computer, tablet, or smartphone. By contrast, hard-​copy documents need to be tracked down, an effort that could take more time than available for certain treatment decisions. But social media is not a unitary category, and each element of information must be evaluated within its context.

Other Concerns Assuming that social media can play an important role in surrogate decision making, are there other concerns with the use of this information? A recent article in Wired Magazine noted that “every time you post something on Twitter, Facebook, Tmblr, or Instagram, you’re influencing—​or trying to influence—​how the world views you … Social Media isn’t about having a conversation with people you know. It’s about advertising. It’s not social; it’s media … We are all constantly building our brands online.”30 In response, Tom Anderson, the founder of MySpace, stated: “The fact that you think about what you post on social networks may make it seem more like ‘advertising’ and ‘personal branding’ than conversation, but it’s not that simple. At some level, even in person, we’re always advertising ourselves to those who we care about … Being online simply gives us more time to figure out who and what we would like to be and project that to other people.”31 The study of online identity is in its infancy. The little data we have show possible generational differences between how individuals use social media. For example, adolescents32 and even young adults in their twenties are building their identities,33 online and offline. They may try on and discard multiple identities over time. In the same way they might dress preppy at one point and goth at another, adolescents reconfigure their social media sites to reflect outwardly their current internal identity choices. While this tendency to rapidly transform online identity may make us wary of using statements on electronic media to make surrogate decisions, the same is true of other evidence that comes from adolescent years and may be addressed by a careful application of the factors discussed previously. More complicated is the growing use among young adults of multiple online identities. Some of these identities may be particularly unsuited for surrogate decisions, such as avatars. But others are not as clear-​cut. For example, many young professionals have a personal Facebook page separate from their professional Facebook page, in an effort to keep their private and public lives separate.34 Here, the possibility that statements will be made on a public page in order to advertise certain characteristics that do not truly reflect the individual’s wishes regarding medical care can be a concern. Perhaps someone whose wishes are not in accord with their overt religious traditions, or whose wishes are unusual, will hesitate to post their true feelings.

Social Media and End-of-Life Decision Making    285 However, there is no indication that this is currently an issue. To some extent all surrogate decision making evidence suffers from this problem. In all cases we must decide whether the information in question in fact reflects what the person would want done. Our lack of certainty is what drives some states to put in place greater limits on surrogate decisions than competent patient decisions and to scrutinize the level of evidence available. Since such decisions are always an approximation of what the individual would want and not a direct exercise of autonomy, such limitations are appropriate.35 A second concern reflects the legal sufficiency of social media evidence. Social media has been used in family law (e.g., divorce and child custody), employment law, and criminal law (to show criminal behavior and juror misconduct). Courts are unlikely to reject the use of social media in surrogate decision making cases. In fact, they may give it greater weight than remembered spoken conversations. The crucial issue is authentication of the information—​assuring that the author was, in fact, the person claimed to be the author. The National Institute of Standards and Technology publishes guidelines for authenticating online identity (updated in 2011). But these standards do not address voluntary sharing of login IDs and passwords. Like generational differences in identity, there may also be generational differences in the sharing of sign-​in information. Some current data show teenagers share passwords as a sign of trust in romantic relationships.36 Older users of social media may share log-​in information with younger family members who can help them navigate various sites. In both situations there is a risk that postings will be made by someone other than the individual of record. While fraud in the creation of traditional wills is well documented, its prevalence in the living will or advance-​directive context is unknown. The use of electronic media prevents the use of investigational tools such as handwriting analysis. But fraud concerns can also be raised for remembered oral conversations. In all situations the weight given to information should depend on analysis of the factors listed previously. Statements that appear out of character, or that lack any corroboration, should be relied on with caution. Whether any amount of information will suffice to meet legal evidentiary standards for end-​of-​life decisions is another aspect of sufficiency. Most states require a “preponderance of the evidence,” sometimes described as greater than 50% likelihood, that the decision in question is one that the patient would have made if competent.37 But other states require a higher degree of evidence—​“clear and convincing” (described as somewhere between 60% to 75% likelihood).38 No specific bit of information automatically meets the clear and convincing evidence standard. Some courts have rejected what appears to be a significant amount of evidence of spoken communication regarding preferences.39 While a legally executed written advance directive likely meets the higher evidentiary standard, what other information would be sufficient remains unclear. Given the low levels of advance-​directive completion in the general population, social media may be looked at as a valuable tool to supplement remembered conversations. Such tools may be crucial in states that apply higher evidentiary standards. If social media proves to be a valuable resource, perhaps online searches for information should be required. At this point, this is unwarranted. There is no formal requirement to search for specific types or sources of information. Even for legally executed

286   Jessica Berg advance-​directive documents, the law merely requires that a health-​care institution (or health-​care professional) ask about the existence of the document, not search all possible locations. Moreover, is not clear who would bear the burden of a social media search and whether that individual would have access to the online resources. Facebook access is by invitation only, and few (if any) members of the treatment team are likely to be “friended.” Even family members may not have full access to all the information on the individual’s Facebook site since different things can be shared with different people. In some cases the designated surrogate may not be the person who has access to the social media information. The situation can become more complicated where younger generations share broader access to social media between themselves but not with older generations. For example, siblings may have more extensive access to each other’s Facebook pages than parents. Parents, however, may be the ones designated as surrogate decision makers. Here, as in other surrogate decision making situations, coordination of family members will be crucial. Although searches of social media should not be mandated, surrogates should be told that information found on social media may be one of the sources of evidence to consider in decision making. Moreover, treatment teams and ethics committees may query surrogates about the patient’s online activity and encourage them to seek out relevant information if appropriate. But surrogates who are unaware of relevant social media postings should not be penalized for failing to access them. Furthermore, if social media information is used to challenge a particular decision by a surrogate, or if a family member wants to challenge the validity of the social media evidence, the burden is on the challenger to show that the surrogate is not making the appropriate choice (and possibly also that the surrogate is not the appropriate decision maker). This mirrors the current situation for surrogate decision making in the absence of social media information. All relevant information should be taken into consideration when there are challenges—​ social media is only one part—​and each aspect of information should be evaluated according to the factors identified here. In some of these cases, as with other medical decision making conflicts, assistance from ethics consultants or even courts will be necessary to resolve disagreements.

Conclusion Social media already pervades many areas of our lives. As current users age, it is increasingly likely that social media evidence will be presented by surrogate decision makers trying to make decisions for incompetent patients. Those involved in these decisions—​ medical care providers, families, and courts—​should be encouraged to evaluate the information as they would any other information. In some cases social media will prove especially useful in understanding an incapacitated person’s wishes. It is even possible that new social media tools will be developed that provide specific guidance regarding advance-​care instructions or surrogate decision making. For example, some people who

Social Media and End-of-Life Decision Making    287 are reluctant to fill out a hard-​copy form and have a face-​to-​face discussion may be more comfortable with completing an online process and engaging in electronic communications, such social media postings. This may be especially true of younger generations who use electronic media regularly. Evidence from other areas of e-​medicine suggests that certain patient groups are more comfortable communicating in online settings about sensitive information than in face-​to-​face settings.40 It is also possible that social media sites will be proactive about their role in decision making. For example, Facebook could choose to facilitate its role in this context by providing a mechanism to encourage advance-​care planning in the same way that it created a feature to allow users to post their organ donation status.41 The use of social media to facilitate advance medical decision making should be explored in more detail; it may serve an important role going forward. Although more research is needed to understand fully the appropriate role for social media in this context, clinicians and ethics committees must anticipate scenarios like the one posted at the outset of this article and prepare to help surrogates navigate the complexities of medical decision making in the Internet age.

Acknowledgments This article was adapted from Berg J. 2012. “Social Media in the Internet Age.” Am J Bioeth. 12:28–​33.

Notes 1. Teresa B. Burke, “A Disability Response to Surrogate Decision Making in the Internet Age,” American Journal of Bioethics 12, no. 10 (2012): 36–​37. Burke points out that these are often used by hearing impaired individuals. 2. “Farewell to BlogPulse,” accessed March 13, 2013, http://​www.blogpulse.com/​ (accessed December 12, 2011). BlogPulse was shut down on January 14, 2012, at a final count of 182,397,015 blogs. http://​smartdatacollective.com/​node/​44748. 3. http://​www.aim.com/​, accessed December 12, 2011. 4. http://​explore.live.com/​messenger, last modified April 2013. 5. http://​www.google.com/​talk/​, accessed December 12, 2011. 6. http://​twitter.com/​, accessed December 12, 2011. 7. Some have also included email features (e.g., Facebook). The reverse has also been true. Google started Gmail in 2004 and recently started beta testing a new feature, Google+, an SNS. 8. Danah M. Boyd and Nicole B. Ellison, “Social Network Sites:  Definition, History, and Scholarship,” Journal of Computer-​Mediated Communication 13, no.  1 (2007):  210–​230. http://​jcmc.indiana.edu/​vol13/​issue1/​boyd.ellison.html. 9. Boyd and Ellison, “Social Network Sites.” What makes social network sites unique is not that they allow individuals to meet strangers but rather that they enable users to articulate and make visible their social networks.

288   Jessica Berg 10. http://​press.linkedin.com/​about, accessed December 12, 2011. 11. http://​www.facebook.com/​press/​info.php?statistics, accessed December 12, 2011. Facebook is the most popular SNS, with approximately 800 million active users worldwide. Other popular sites include MySpace (http://​www.myspace.com/​), with about 110 million users, and bebo (http://​www.bebo.com/​), a popular site outside the United States. 12. http://​newsroom.fb.com/​company-​info/​ accessed August 27, 2014. 13. Zoe Fox, “Two-​Thirds of Online U.S. Adults Use Social Media—​But Why,” Mashable, last modified November 15, 2011, http://​mashable.com/​2011/​11/​15/​social-​media-​use-​study/​. 14. Thaddeus M. Pope, “Facebook Can Improve Surrogate Decision Making,” American Journal of Bioethics 12, no. 10 (2012): 43–​45. 15. Jessie Berg, “Legal Issues in End-​of-​Life Decision Making,” Journal of Health Care Compliance (2007): 1–​43. All states have some form of advance-​directive legislation. 16. Pamela L. Dolan, “Handful of States Promise Physicians Online Access to Advance Directives,” American Medical News, January 3.  http://​www.ama-​assn.org/​amednews/​ 2012/​01/​02/​bise0103.htm, accessed January 9, 2012. 17. http://​www.donatelifeohio.org/​, accessed December 12, 2011. 18. “Online Living Will Completion for Texans,” TexasLivingWill.org. Developed by faculty at the University of Texas Health Science Center. 19. Harry McCracken, “Did Facebook Just Change Social Networking Forever?” Time, September 29, 2011. http://​content.time.com/​time/​business/​article/​0,8599,2095516,00. html, accessed December 12, 2011. A public search shows one example. Because of the way Facebook is set up, a search will only show results for people who are designated as “friends.” Informal surveys in various settings show that there are people who say they have posted their advance directive on Facebook. While specific numbers are unclear, the use of Facebook to store these kinds of documents may become more prevalent. Facebook announced in September 2011 that the new version would “try to come far closer to replicating your entire life—​and to keep track of it all for as long as both you and Facebook exist.” Jenna Wortham, “Your Life on Facebook, in Total Recall,” The New  York Times, December 15, 2011, http://​www.nytimes.com/​2011/​12/​16/​technology/​ facebook-​brings-​back-​the-​past-​with-​new-​design.html?_​r=1&nl=todaysheadlines&e mc=tha26 (accessed December 16, 2011). This may make Facebook (or a similar SNS) an ideal storage location for all sorts of relevant legal documents, including advance directives. 20. “Million Dollar Baby Tweet-​Along on AMC—​2.27.11,” Cinemit, February 27, 2011, http://​www. cinemit.com/​content/​million-​dollar-​baby-​tweet-​along-​amc-​22711, accessed December 13, 2011. For example, the movie Million Dollar Baby had a “tweet-​along” when shown on television. The final scene raised many questions about end-​of-​life care. Could tweets in response to a movie be used by a surrogate making medical decisions for one of the viewers? 21. I  use the term “surrogate” to cover proxies (designated by the patient) and legally appointed guardians and conservators. In some states, guardians and conservators have specific limitations on decision making for incompetent patients. 22. Cruzan v. Director, 497 U.S. 261 (1990). Terri Schiavo case in Florida. 23. In the Matter of Claire C. Conroy, 486 N.J. A.2d 1209 (1985). 24. http:// ​ w ww.facebook.com/​ f ind-​ f riends/ ​ browser/ ​ ? ref=ler#!/ ​ p ages/ ​ Terri- ​ s chiavo/​ 108246655875925, accessed December 15, 2011.

Social Media and End-of-Life Decision Making    289 25. http:// ​ w ww.facebook.com/​ O rganDonors?ref=ts#!/ ​ p ages/ ​ O rgan- ​ D onation/​ 267705317204, accessed December 15, 2011; 8,315 people “like” this. 26. According to Facebook, the metric includes all of the following: liking a page; posting to a page’s wall; liking and commenting on or sharing a page post or content on the page; answering a question posed; rsvp’ing to an event; mentioning a page in a post; photo tagging a page; liking or sharing a check-​deal; checking in at a place. 27. William Dale, “Reflections on My Father’s Passing … and Writing About It,” The John A Hartford Foundation, May 26, 2011, http://​www.jhartfound.org/​blog/​?p=3625, accessed December 12, 2011. 28. Linda Emanuel et  al., “Advance Directives:  Stability of Patients’ Treatment Choice,” Annals of Internal Medicine 154, no. 2 (1994): 209. Peter Ditto et al., “Context Changes Choices:  A  Prospective Study of the Effects of Hospitalization on Life-​ Sustaining Preferences,” Medical Decision Making 26, no. 4 (2006): 313–​322. 29. Rebecca Dresser and Alan B. Astrow, “An Alert and Incompetent Self: The Irrelevance of Advance Directives,” The Hastings Center Report 28, no. 1 (1998): 28–​30. 30. Erin Biba, Wired Magazine, August 2011. 31. Thomas Anderson, Wired Magazine, September 2011. 32. Adolescence lasts from around age thirteen to age nineteen or even into early twenties. The best-​interests standard is usually applied to minors, so at issue would be statements from adults in their late teens and early twenties. 33. Erik H. Erickson, Identity and the Life Cycle (New  York:  International Universities Press, 1959). 34. Of course the whole notion of a “public” and a “private” life is challenged by the use of social media. 35. Cruzan v. Director, 497 U.S. 261 (1990). 36. Matt Richtel, “Young, in Love and Sharing Everything, including a Password,” The New  York Times, January 17, 2012, http://​www.nytimes.com/​2012/​01/​18/​us/​teenagers-​ sharing-​passwords-​as-​show-​of-​affection.html?_​r=1&nl=todaysheadlines&emc=tha26, accessed January 19, 2012. 37. C.M.A. McCauliff, “Burdens of Proof:  Degree of Belief, Quanta of Evidence, or Constitutional Guarantees,” Vanderbilt Law Review 35 (1982): 1293–​1335. It discusses percentage requirements for burdens of proof and surveying judges’ attitudes. 38. Cruzan v. Director, 497 U.S. 261 (1990). The Supreme Court in the Cruzan case held that a state (there, Missouri) may choose to apply a clear and convincing evidentiary standard for surrogate decision making. Florida, Missouri, New York, Michigan, Texas, and some other states use this higher standard. 39. In re Martin, 538 N.W. 2d 399 (Mich. 1995). 40. Jessica Berg, “Ethics and E-​Medicine,” St. Louis University Law Journal 46 (2002): 61. 41. Pope, “Facebook Can Improve,” 44.

References Anderson, Thomas. Wired Magazine. September 2011. Biba, Erin. Wired Magazine. August 2011. Berg, Jessica. 2002. Ethics and e-​medicine. St. Louis University Law Journal 46:61.

290   Jessica Berg Berg, Jessica. 2007. Legal issues in end-​ of-​ life decisionmaking. Journal of Health Care Compliance: 1–​43. Boyd, Danah M., and Ellison, Nicole B. 2007. “Social Network Sites: Definition, history, and scholarship.” J Comput Mediat Comun. 13(1):210–​230. http://​jcmc.indiana.edu/​vol13/​issue1/​ boyd.ellison.html Burke, Teresa B. 2012. “A disability response to surrogate decision making in the Internet age.” American Journal of Bioethics 12(10):36–​37.

Cruzan v. Director. 1990. 497 U.S. 261.

Dale, William. 2011. Reflections on my father’s passing … and writing about it. The John A Hartford Foundation, May 26. Accessed December 12, 2011. http://​www.jhartfound.org/​ blog/​?p=3625. Ditto, Peter H., Jacobson, Jill A., Smucker, William D., Danks, Joseph H., and Fagerlin, Angela. 2006. Context changes choices: A prospective study of the effects of hospitalization on life-​ sustaining preferences. Medical Decision Making 26(4):313–​322. Dolan, Pamela L. 2012. Handful of states promise physicians online access to advance directives. American Medical News, January 3. Accessed January 9, 2012. http://​www.ama-​assn. org/​amednews/​2012/​01/​02/​bise0103.htm. Dresser, Rebecca, and Astrow, Alan B. 1998. An alert and incompetent self: The irrelevance of advance directives. The Hastings Center Report 28(1):28–​30. Emanuel, Linda L., Emanuel, Ezekiel J., Stoeckle, John D., Hummel, Lacinda R., and Barry, Michael J. 1994. Advance directives: Stability of patients’ treatment choice. Annals of Internal Medicine 154(2):209. Erickson, Erik H. Identity and the Life Cycle. New York: International Universities Press, 1959. Fox, Zoe. Two-​thirds of online U.S. adults use social media—​but why? Mashable.com. Last modified November 15, 2011. http://​mashable.com/​2011/​11/​15/​social-​media-​use-​study/​.

In re Martin. 1995. 538 N.W. 2d 399 (Mich.). In the Matter of Claire C. Conroy1. 1985. 486 N.J. A.2d 1209.

McCauliff, C.M.A. 1982. Burdens of proof: Degree of belief, quanta of evidence, or constitutional guarantees. Vanderbilt Law Review 35:1293–​1335. McCracken, Harry. 2011. Did Facebook just change social networking forever? Time, September 29. Accessed December 12, 2011. http://​content.time.com/​time/​business/​article/​ 0,8599,2095516,00.html. Million Dollar Baby tweet-​along on AMC—​2.27.11. 2011. Cinemit, February 27. Accessed December 13, 2011. http://​www.cinemit.com/​content/​million-​dollar-​baby-​tweet-​along-​amc-​22711. Online living will completion for Texans. n.d. TexasLivingWill.org. Pope, Thaddeus M. 2012. Facebook can improve surrogate decision making. American Journal of Bioethics 12(10):43–​45. Richtel, Matt. 2012. Young, in love and sharing everything, including a password. The New York Times, January 17. Accessed January 19, 2012. http://​www.nytimes.com/​2012/​01/​18/​us/​ teenagers-​sharing-​passwords-​as-​show-​of-​affection.html?_​r=1&nl=todaysheadlines&emc= tha26. Wortham, Jenna. 2011. Your life on Facebook, in total recall.” The New York Times, December 15. Accessed December 16, 2011. http://​www.nytimes.com/​2011/​12/​16/​technology/​facebook-​ brings-​back-​the-​past-​with-​new-​design.html?_​r=1&nl=todaysheadlines&emc=tha26.1.

Chapter 18

C u ltu ral Fac tors Megan Crowley-​M atoka

Problematic Death Death is a problem. Or rather, it provokes a whole host of problems: How do we recognize and respond to its approach? How do we confirm when it has—​definitively—​ arrived? What does death mean, to the person facing it, and to the social group left behind? And what is to be done with what remains, with the physical body, with the worldly property, and with the web of memories and emotional ties and social obligations that formerly constituted the living person? While such questions are universal, their answers most decidedly are not. For most twenty-​first-​century Americans, death has become a largely biological—​indeed medical—​matter, to be resisted for as long as possible with the aid of physicians and pharmaceuticals and biotechnologies. But this, of course, has not always—​or everywhere—​been so. Among the Yolmo Buddhists of Nepal, for example, the approach of death is best met with a cultivated acceptance rather than resistance, with a gradual, ritual relinquishment of the ties of this particular life in order to move gracefully into a good rebirth (Desjarlais, 2003). Captured in such contrasting responses to the approach of death is the fact that while is death is always a problem, just what sort of problem it is thought to be, and what sorts of responses are deemed proper and fitting, is profoundly shaped by culture. Indeed, across different times and places people have come up with radically different solutions to the range of material, social, and existential conundrums posed by the fact of our mortality. Reports across the historical and anthropological record reveal enormous variety in how death is defined and recognized, in what is understood to constitute “good” versus “bad” forms of death, and in what the appropriate responses to death are understood to be. Definitions of death may be made to hinge on various bodily states, from lack of consciousness, to cessation of heartbeat, to evidence of putrefaction. Thus for the Vanatinai of Papua New Guinea, even transitory unconsciousness is considered to be a form of death, such that someone may be said to have died multiple times in a lifetime (Lepowsky, 1985). In eighteenth-​century Europe, the physician Jean-​Jacques

292   Megan Crowley-Matoka Winslow’s own experience of premature burial as a child prompted him to contend that death was not certain until the body began to visibly decay (Alexander, 1980). And, of course, our own medicalized era has seen the boundaries of death pushed and pulled in various ways. With the widespread adoption of brain-​death criteria, death can now be pronounced once the brain has ceased to function—​even if the workings of the rest of the body can, at least for a time, be technologically maintained (Wijdicks, 2001). And that same technological maintenance—​via ventilator, feeding tube, and pharmaceuticals of various kinds—​can also be used to stave off death for longer than ever before, maintaining patients in a persistent vegetative state for years or even decades of an existence that some have termed a form of “living death” (Kaufman, 2000). Recognizing death is not just a matter of pinpointing when it has taken place, of course. It is also a matter of determining what kind of death has occurred, for deaths may be differently deemed good or bad, natural or unnatural, worthy of full mourning or not. Such determinations turn not just on how the death itself has unfolded but on the status of the being to which it has come. Thus in times and places where infant mortality has been especially high, infants may go unnamed for some ritual period of time as a mark of their still liminal personhood, and their deaths are marked with more attenuated forms of grieving, what the anthropologist Nancy Scheper-​Hughes has called “death without weeping” (1992). In contrast, reproductive technologies such as uterine imaging and in vitro fertilization have made it possible for present-​day couples in the United States to name, mourn, and memorialize miscarried embryos only weeks old (Reagan, 2003). At the other end of the life course, the death of an elderly person may be regarded as proper, fitting, and even a gift to the social group released from the burden of ongoing care for a member who can no longer contribute to collective survival (Koenig and Marshall, 2014). In different times and places, such a “good death” might be marked in widely disparate ways, among them the Amazonian Wari’ practice of a ritual form of “compassionate cannibalism” that reverently loosens the ties between the living and the dead by dissolving the corpse into the social group through its ingestion (Conklin, 2001). Or, as has become increasingly common in the United States, dying at any age may come to seem unjust and “unnatural” as even the very elderly—​those in their 80s and 90s—​are expected to fight unceasingly for more life through a whole host of medical technologies such as organ transplantation and implantable cardiac devices (Kaufman et al., 2006; Kaufman et al., 2011). Ultimately, such stark contrasts in how humans identify and respond to death serve as incontrovertible evidence of how death is a deeply cultural—​not just biological—​matter.

Problematic Culture Yet just what does it mean to say that death is “cultural”? First, it seems useful to have a working definition of this common yet complex concept. Put simply, we can understand culture as a set of shared meanings, norms, and values that underlie and shape the

Cultural Factors   293 way members of a social group understand and act in the world around them. Or, as the anthropologist Clifford Geertz has described it in more vividly evocative terms: “Man is an animal suspended in webs of significance he himself has spun—​I take culture to be those webs” (1973: 5). As such, culture operates as the taken-​for-​granted lens through which we interpret our experiences, rendering those meanings, norms, and values often tacit and largely invisible—​it is difficult, after all, to see those webs in which we ourselves are held. It is this powerful feature that makes culture often most evident in the unfamiliar, sometimes unsettling ways of “other” people, while our own cultural beliefs, attitudes, and practices seem instead only commonsense and even “natural.” Powerful, yet not deterministic in any straightforward sort of way, culture must also be understood as both multiple and mutable, because. Individuals living within any given cultural system may take up, reject, and engage with its norms and values in varying ways, according to their own positionings of gender, class, race, age, sexuality, and many other forms of identification. Moreover, most people in the contemporary world belong to multiple social groups of one kind or another and so navigate constantly and creatively between the beliefs and expectations of multiple cultural worldviews. Neither monolithic nor static, cultural systems themselves are also always changing, shifting in response to the pressures of internal societal debate and the influence of contact with other cultural ways of being in the world. While its taken-​for-​granted character is indeed one of culture’s key hallmarks, contemporary American society—​and perhaps American medicine in particular—​has seen increasing efforts to make culture more deliberately visible as a matter of discussion and a site for potential intervention and change. Such efforts emerge, for instance, in the growing body of medical literature attentive to the role of culture in shaping patient beliefs and behavior—​evident in what is often called the cultural competency movement, for example (Bettancourt, 2004). Over time this attention has expanded beyond the patient, to include the ways in which medicine is itself a cultural system that powerfully shapes the attitudes and practices of its professional practitioners as well (Fox, 2005; Boutin-​Foster et al., 2008). Within medicine the notion of culture is thus most often invoked as a matter of influence on individuals—​be they patients or clinicians—​ whose beliefs, attitudes, and behaviors are shaped by the particular webs of significance within which they reside. Less well explored, at least in the medical literature, is the way that cultural meanings and values also get sedimented into the structures and institutions of medicine, shaping the way that medical care is physically organized, politically regulated, and financed, for example. It is this more expansive understanding of culture in medicine as not just a matter of beliefs, attitudes, and practices but of the institutional structures and flows of power and money that make up medicine as well that must be brought to bear on any exploration of the relationship between culture and dying. Reflecting this more general move to consider culture within medicine, the particular shape of dying in contemporary America—​with its often technologically protracted character—​has increasingly come to be identified as a problem of culture. American cultural values of individualism and unswerving faith in the inevitable march of scientific progress are thought to combine with the culture of medicine itself—​in which death

294   Megan Crowley-Matoka represents defeat, cure is valued over care, and the technological drive to do something is enormously powerful—​to produce troubling patterns at the end of life (Callahan, 1995). Particularly at issue is the apparent disjoint between what many Americans describe as a “good death” (peaceful, dignified, often orchestrated to occur at home, surrounded by family and friends) and the reality that death in the United States is so frequently preceded instead by lengthy hospitalizations, invasive surgical procedures, and intensive technological support. Partially in response to this posing of the problem, a number of reviews in recent years have surveyed the role of culture at the end of life in American medicine, often with an eye toward improving the capacity of health-​care providers to communicate across and accommodate cultural differences in dying patients and their families (e.g., Koenig and Gates-​Williams, 1995; Kagawa-​Singer and Blackhall, 2001; Crawley et al. 2001; Coolen, 2012). Others have trained the analytic eye more directly on the death-​denying and -​defying features of the medical culture itself, seeking to make these tacit values more visible and hence, perhaps, more available to open discussion and deliberate change (e.g., Callahan, 1995; Kaufman and Morgan, 2005; Chapple, 2010; Timmermans, 2010). The question of culture as a crucial part of the problem in contemporary forms of death and dying is thus rather well-​covered ground at this point. What more might we add to what has already been said? Perhaps one opportunity to enrich the ongoing conversation presents itself in the serendipitous existence of two widely acclaimed, highly skilled, and historically spaced ethnographic explorations of the process of dying in American hospitals. Bringing together these two studies—​published precisely forty years apart—​in close reading offers an intriguing opportunity to examine how understandings of the problem(s) of death, as well as the cultural features that shape how it unfolds, have (and have not) shifted over a particularly critical period of time in the United States.

Forty Years of Dying in American Hospitals: A Tale of Two Ethnographies In 1965 the sociologists Barney Glaser and Anselm Strauss published Awareness of Dying, an ethnographic study of interactions between health-​care providers, dying patients, and their families in six California hospitals (Glaser and Strauss, 1965). The study was in part a response to a growing sense at the time that death and dying had been a neglected area of study and a largely silenced topic of conversation in society at large (Faunce and Fulton, 1958, Timmermans, 2010). Groundbreaking both substantively and methodologically, Awareness of Dying was an almost instant classic, invigorating new lines of both research and activism around the issue of dying in American society. Forty years later, in 2005, the anthropologist Sharon Kaufman published And a Time to Die: How American Hospitals Shape the End of Life, also an ethnographic study of dying

Cultural Factors   295 in several California hospitals (Kaufman, 2005). In contrast to the social silence around death that motivated Glaser and Strauss in their time, for Kaufman it was instead the almost oppressively pervasive cultural conversation about what constitutes a good death in contemporary American society, rife with notions of “death with dignity” and “quality of life,” that called out for urgent study. And though not quite defining a new field of inquiry in the same way, Kaufman’s award-​winning research has also had significant impact far beyond its home discipline, garnering wide attention and acclaim in both the medical literature and the popular press. Not just influential each in their own right, however, taken together these two studies offer a strategic lens through which to consider questions of culture and the problematic end of life in American medicine for (at least) three specific reasons: their focus on hospitals, their use of ethnographic research methods, and the very particular period of history elapsed in the time between them. First, although only about one-​third of deaths in the United States occur in hospitals, it is clear that the hospital is a key site of anxiety about a particular form of “bad death” in the contemporary United States, one that is technologically intensive and temporally drawn out in a ways that are often thought to both produce suffering and harm dignity—​ as well as drive up health-​care costs (Hall et al., 2013). These sorts of deaths register as particularly troubling in a setting where the overwhelming majority of people report that they want to die at home (Cloud, 2000). Moreover, even when the actual moment of death does not occur in a hospital, recent research demonstrates that the process of dying in the United States—​the days and weeks leading up to death itself—​is increasingly likely to involve multiple hospitalizations and the use of intensive care (Teno et al., 2013). Together, all of this makes hospitals a central location in the landscape of death and dying in contemporary America, both practically and imaginatively. Hospitals are not only central, however; they are also enormously complex. And they are complex not just in terms of the dizzying variety of medical specialization and forms technological and pharmaceutical care that they offer. They are also institutionally and socially complex, with bureaucratic requirements, financial arrangements, hierarchies of power, and specialized forms of ritual and language that are largely opaque to the patients and families who enter them. In short, hospitals are a distinctive social world unto themselves. In trying to study the cultural shape of dying in such a setting, the ethnographic approach—​employed by both Kaufman and the Glaser and Strauss team—​ offers some distinct advantages. Ethnography, the signature method of anthropology, typically involves a combination of in-​person observation and in-​depth interviewing, often supplemented with the collection of other forms of quantitative, textual, and/​or visual data (Bernard, 1994). Developed originally as a means of studying small-​scale societies, ethnography can be described as a flexible and full-​contact form of research designed to enable the researcher to understand, as fully, deeply, and experientially as possible, another social group’s way of being in the world. Because of its emphasis on immersive, long-​term, and multiple-​method engagement with the social world under study, ethnography is particularly good at producing fine-​grained, naturalistic data not just on isolated beliefs or behaviors but on the complex ways that beliefs, behaviors, and institutional structures combine and interact to form a whole cultural system. Thus if

296   Megan Crowley-Matoka the complex social world of hospitals offers a key site in which to examine the problem of death in America, then ethnography offers a methodological tool particularly well attuned to the job. Finally, and perhaps most compellingly, the particular arc of time captured between these two studies has seen a number of crucial historical changes with direct relevance for the cultural shape of death and dying in the United States. Most obviously pertinent, of course, is the rise of hospice beginning in the 1970s, a movement with the explicit goal of improving the management and experience of dying through promoting deinstitutionalization of the dying person from hospital to home, highly attentive symptom management, and self-​aware preparation for death (Siebold, 1992). The notion, embedded in hospice, that the dying can—​and should—​take charge of the process of their own demise signals a second key cultural shift going on during this period, a move away from paternalism in medicine and toward a framework of “patients’ rights.” Arising out of the larger set of rights’ movements around race and gender during the same period in the United States, and shaped also by the emerging field of bioethics, the notion of patients’ rights sought to rebalance power relations between doctor and patient, staking a claim for the rights of patients to know their own medical information and to make “autonomous” decisions about their medical care (Rothman, 2001). Finally, the emphasis on decision-​making and choice intrinsic to the very concept of patients’ rights has, over this same historical period, played into a third shift critical to the way death is understood, managed, and experienced in contemporary America: the increasing corporatization of health care (Rylko-​Bauer and Farmer, 2002). Indeed, as health care in the United States has become ever-​more market driven over the past several decades, business logics of marketing and “consumer choice,” coupled with the drive for more efficiency, have come to powerfully shape the very possibilities for how, where, and when death unfolds.

Shifting Dilemmas of Dying Despite the four decades—​ and significant social change—​ that stretch between Awareness of Dying and And a Time to Die, one immediately striking effect of bringing these two studies together is to reveal an almost eerily similar framing of what constitutes “the problem” of dying in America. Glaser and Strauss preface their study by describing the pressing need to delve into American attitudes toward death and dying in light of “what seems a senseless prolonging of life within hospital walls by medical technology run wild” (1965: vii). And noting the general silence and even outright denial that tended to surround death in 1960s’ America, they ask whether this silence contributes to the problem by denying “the dying person the opportunity… . to control his style of dying, much as he controlled his style of living” (6). Forty years later, Kaufman characterizes the context that produced the pervasive cultural conversation about the problem of death in America in this way: “Ordinary individuals were disturbed about the ways their loved ones died in hospitals—​connected to tubes and machines, following endless

Cultural Factors   297 procedures, disoriented and in pain. In general, they died with too much technological intervention, and with too little personal say” (2005: 2). The staunch persistence of these two central features of American dying—​too much technology and not enough personal control—​across this tumultuous, change-​filled, forty-​year period seems both surprising and somewhat sobering. Yet a tremendous amount has also changed in the attitudes, practices, and institutional structures of dying traced across these two studies. Indeed, it is precisely the comparative contrast between them that makes some of the central cultural features of contemporary death and dying stand out more clearly. Seen from this comparative angle, three critical shifts in what the process of dying was in American society and what it has become thus emerge with particular power, poignancy, and import—​from no clear cultural “script” for what would constitute a good death to one that is ubiquitous yet nearly impossible to achieve; from physician control to patient responsibility; and from death being a matter of “nothing more to do” to one of “nothing more we should do.” Each of these changes in the culture of death and dying is explored in more detail in the following.

From No Script to an Impossible One In the 1960s when Glaser and Strauss conducted their study, death was a rather taboo subject not just in general American society but even in medical practice—​as they note, despite dying being an inevitable part of caring for ill people, medical professionals received virtually no training in how to manage or communicate about death with either patients themselves or their family members. In Glaser and Strauss’s study, the central drama surrounding the dying process in the hospital was simply the question of who knows that death is approaching—​this is the “awareness of dying” that titles their book—​and in their era it was commonplace for medical staff to actively conceal from patients and family members when death was imminent. This was done in part because it was thought that those who “really wanted to know” would find a way to learn the truth, either through persistence, careful attunement to the “clues” sometimes dropped by staff, or even sometimes outright guile. But concealment was common also because medical staff often feared that informing the dying patient would provoke a “falling to pieces” reaction that would prove too disruptive to their own work, as well as too upsetting for both staff and the patient’s family to witness. As Glaser and Strauss rather bluntly put it: “Unaware patients who die quickly tend to do so without fuss, so the hospital’s work routine is delayed less” (1965: 45). The management of death was thus geared, in many ways, to ease the experience of those around the dying person—​rather than that of the dying person him or herself. And yet implicit in this assumption that many patients would react badly (and, hence, disruptively) to the news of their own approaching death was also the possibility for its opposite, for a “good” or “proper” form of reaction, presumably one that would not unduly burden either the staff or the family members caring for the patient during his or her final hours.

298   Megan Crowley-Matoka Indeed, for those patients who did become aware of their dying status, Glaser and Strauss make clear that medical staff held a certain set of expectations about how the dying should behave, for, in their words, “ ‘facing’ an impending death means that the patient will be judged, and will judge himself, according to certain standards of proper conduct. These standards pertain to how a man handles himself during his final hours, and to his behavior during the days he spends waiting to die” (1965: 82). Highlighted here is the way that knowing about death’s approach thus conveys a certain responsibility for acting, a responsibility for responding and behaving in the socially considerate and tacitly sanctioned ways that would not overly tax others with one’s own suffering. And yet, living up to these expectations was considerably complicated precisely by their tacit nature in a society where death was an intensely private and largely silent matter, an event often isolated in hospitals where death was so frequently treated like a dangerous secret. As Glaser and Strauss put it: “People are supposed to live correctly while dying, providing they understand that they are dying, but Americans have no clear rules for this behavior. No script has been carefully laid out for the dramatis personae” (85). In contrast to the many times and places where dying has been a process accomplished in the presence of family and community and surrounded by well-​elaborated social ritual, Glaser and Strauss contend that, for Americans in the 1960s, death had become such a private, isolated matter that people had few chances to directly experience—​and hence understand and practice—​what a “proper” form of dying might look like. This situation of dying patients—​even for those who know they are dying—​left without a clear script for how the process might best be accomplished stands in stark contrast with the state of affairs described forty years on by Kaufman, whose own study was in part motivated by curiosity about the very pervasiveness of the open, sometimes even strident, cultural conversation about contemporary death in both “good” and “bad” forms. Growing out of some of the historical trends identified previously, in particular the hospice and the patients’ rights movements, Kaufman traces the emergence in contemporary America of a highly elaborated “script” for dying well. Developed in explicit reaction to the posing of the “problem of death” in terms of too much technology and not enough personal control, this model for a good death advocates for both acceptance and intentionality in how death is met, such that patients who know they are dying can—​and should—​orchestrate the process to best express their own values. As Kaufman describes it: “The emphasis is on individual patient control of the style of death, and ‘good’ mostly indicates a death that is aware, pain free and in which psychological and worldly business are completed” (2005: 71). Critically implicit in the notions of “acceptance” and “awareness” at work here is the idea that patients, once they know that they are dying, should exercise their right to control the process so as to avoid its futile, technological prolongation. The epitome of a good death in contemporary America has thus become a dignified death, gracefully accomplished at home, with symptoms and suffering well controlled and with time for self-​reflection and the gathering of family and friends for final farewells. Although this vision has become culturally dominant in many ways, one must not misconstrue this as meaning that it is universally shared. Culture, after all, is both

Cultural Factors   299 multiple and mutable, and this is a vision, as many have pointed out, with particular dimensions of race and class and religiosity (Crawley et al., 2000, Waters, 2001, Greiner et al., 2003, Johnson et al., 2008). Some patients’ faith, for instance, may deem it inappropriate to try to anticipate, much less control, their own death, leaving this as a matter entirely in God’s hands. For others, who may have long experienced medical care as a largely inaccessible good, the expectation that they ought to be refusing medical intervention as death approaches may seem unthinkable at best and a horrifying continuation of a lifetime of inequitable treatment at worst. Yet the power of dominant cultural ideas, like the contemporary notion of a good death, is not that everyone in a culture uniformly shares them but that everyone must contend with them in some way. Thus those patients and families who—​for reasons of faith, or experiences of discrimination, or still other reasons entirely—​do not aspire to a death with less technology and more personal control must still deal with the expectations of others, medical staff often among them, who may believe that they should. Yet how the dominant script for how a good death should proceed in contemporary America is complicated not only by the fact that not everyone cares to follow it. As Kaufman’s careful ethnographic stories poignantly reveal across the unfolding of numerous individual deaths, it turns out to be a script that is painfully hard to enact even for many of those who believe in it deeply. This is so, in large part, because that question of knowing when someone truly—​and irreversibly—​is dying is far from straightforward. And death’s approach, of course, can hardly be accepted, much less thoughtfully prepared for and orchestrated, if it has not yet been recognized. The contemporary difficulties in knowing—​definitively—​when the dying process has begun stem in part from the ever-​increasing scope of the intensive technologies of life support, such that there is almost always “something else” to try. Ventilators can be adjusted and readjusted to keep respiration going, vasopressors can be tinkered with to maintain blood pressure, dialysis can be applied to support failing kidneys—​in short, a whole host of techniques and technologies exist to help support or sustain isolated bodily functions, even when the overall picture is one of inexorable decline. As Kaufman describes the tenor of medical activity surrounding the patient (which then guides how patients and families imagine the possible decisions before them): “death is not spoken of or foreseen until shortly before it occurs… . Disease is treated until there is no more physiologic response to therapy. Only then is death expected” (2005:  28). This seductive possibility for heroic rescue through the logic of “there’s always something else to try” is coupled with the increasingly common experience of dying from forms of chronic illness that unfold over a long period of slow decline, punctuated by episodes of crisis and (relative) recovery. Such experiences can make it enormously difficult for patients and their family members—​ as well as medical staff—​to know when this time is different and at least some form of recovery is truly no longer possible. What all of this means is that, in contrast to the era described by Glaser and Strauss, contemporary Americans do have available to them a relatively clear and, for many, very compelling script for how to “die well.” However, having a script does not solve the complex problem of knowing just when to use it. Moreover, as the next section discusses,

300   Megan Crowley-Matoka endorsing that script while in good health—​saying that one wants a “natural,” peaceful death without “heroic” technological intervention—​does not necessarily predict what it will feel like to bear responsibility for choosing death from amidst the myriad uncertainties of illness. As Kaufman succinctly puts it: “While many claim to want that elusive ‘good death’ for themselves and their loved ones, they also want—​equally or more strongly—​that their loved ones not die” (2005: 4). And in the tension between these two desires, an inability to live up to the “good death” many feel they should want may become an additional source of suffering, revealing a painfully ironic underside to the very notion of patient choice and control in the face of death, one that bears further examination.

From Physician Control to Patient Responsibility Indeed, a second key shift in the cultural shape of death across these two studies emerges around just these issues of control and choice and responsibility in both naming and directing the dying process. In the era of which Glaser and Strauss were writing, there was no question: physicians ruled. The ultimate power of physicians to control both information and decision-​making in the 1960s emerges perhaps most strikingly—​ even from the present-​day perspective, shockingly—​in the very commonplaceness of the practice of actively lying to patients about the imminence of their own death. As Glaser and Strauss describe: “In American hospitals, the attending physician is the only one who can legitimately define the patient’s condition… . Ordinarily, only he may tell patients that they are dying” (1965: 19). Thus it was clearly and entirely up to doctors to mark the moment when dying had truly begun—​the moment when, in the common parlance of the hospital, there was “nothing more to do.” In the 1960s this moment was often named among staff only, so that physicians and nurses might share an understanding—​ still unbeknownst to patient and family—​that recovery was no longer to be hoped for or worked for and the deathwatch should begin. In some situations, when the physician deemed them able to handle the news appropriately, this turning point could also be made evident to the patients and/​or their families. But in all cases, responsibility for determining death’s approach lay squarely on the shoulders of physicians alone. This is not to say, of course, that others—​nurses, family members, and patients—​ were not also interpreting the physical signs of decline and coming to their own private conclusions about the state of things. Indeed, Glaser and Strauss report that one common source of tension between nurses and physicians emerged precisely when nurses believed that a physician had failed to recognize the point of no return and was continuing to aggressively, and inappropriately, treat a dying patient. In addition, patients might have believed themselves to be dying long before a physician was ready or willing to openly confirm their terminal status. Yet the power to publicly, formally declare death’s approach resided solely with physicians. This was a power with consequences, of course, for it meant that only physicians could initiate that fundamental shift in both medical practice and emotional tone, the shift from “treating” the patient—​from continuing to

Cultural Factors   301 hope as well as to look for that “something else” to try—​to instead caring for the comfort of the patient during the wait for death’s arrival. The years elapsed since Awareness of Dying have, as noted earlier, seen massive changes in American cultural norms surrounding physician power and patients’ rights, and the kind of absolute power over knowledge of and decisions about the dying process exercised by physicians in the 1960s is nowhere to be seen in the hospital world described by Kaufman forty years later. Instead, she describes a world in which the notion of patient autonomy has become a guiding ethical—​and legal and bureaucratic—​ framework for the practice of medicine. Indeed, medical practice is now thoroughly permeated with the idea that patients (and their families) have the right—​meaning also the responsibility—​to know their own medical information and make decisions about their own treatment. In the context of death and dying, this has come to mean that it is no longer solely physicians who determine just when that point of “nothing more to be done” has been reached. Now patients (and/​or the family members who may be charged with making decisions on their behalf) are expected to “choose” when it is time to stop treatment. That is, they are expected to choose when to die. It is, to be sure, a gravely demanding expectation to put upon dying patients and/​or their families, yet it is the harsh reality that the push for greater personal control in the realm of death and dying has produced. It is also an expectation congruent in many ways with larger trends in the wider American culture toward values of “self-​actualization” and personal responsibility for health, such that dying well becomes the final chapter of in a career of self-​making for which the individual bears both practical and moral responsibility (Rose, 1998). On the one hand, this power now placed in the hands of patients and families is both real and weighty. As revealed in the many patient stories that Kaufman unfolds, patients and families can now choose to press on with more treatment and more technology even when the medical team believes it to be a meaningless prolongation of the inevitable. They can also, of course, choose to step off the treatment pathway and accept the approach of death before their physicians agree that it is time. Patients and families feel the weight of this responsibility acutely, sometimes agonizingly, in the face of complex medical information they are often ill equipped to interpret and “choices” they would never willingly choose. The appealing vision of the good death that pulls back from needless medical intervention in the name of a self-​aware and dignified ending can seem hard to hold on to in the fear-​filled panic of a respiratory or cardiac crisis, for intervention must often be decided on while the potential for reversibility still seems uncertain. That is, just what “choice” one is making is often all too unclear in the moment of its making: whether intubation for mechanical ventilation is a form of short-​term support or a “heroic measure” may become clear only over time. Even when things do seem clear, people’s ideas about what seems tolerable, “dignified,” and a worthwhile form of life often change dramatically when the question of how to die is no longer merely future speculation but a here-​and-​now reality. As Kaufman learned, for many patients and families in the midst of medical uncertainty, more life—​even life sustained by invasive, intensive technology—​can quickly come to seem preferable to death in any form, good or otherwise. Ultimately, Kaufman is describing a world where the dynamics of power

302   Megan Crowley-Matoka and decision-​making between doctors and patients are, indeed, starkly different from that earlier era studied by Glaser and Strauss. Yet this form of patient empowerment carries with it a sometimes dire set of costs, for “the onus of responsibility for deciding what to want often is on the patient or family, yet they rarely have an informed sense of what is best. Scenarios in which families are offered a choice about goals of medical treatment when death is near reveal the dark side of autonomy—​full of anguish, guilt, and above all, absence of knowledge about medical outcomes” (Kaufman, 2005: 28). On the other hand, for all of its consequential—​sometimes terrible—​weight in contemporary death and dying, patient autonomy is also enormously constrained, because patients, of course, may only “choose” from among the choices made available to them—​ and framed for them—​by medical staff. Although the days of widespread, outright lying to patients are no longer, medical professionals remain the gatekeepers of medical information, and they still have power to decide precisely what (and when and how), among the dizzying array of data now produced about a hospitalized patient’s condition, to share. Yet physicians too, Kaufman is careful to point out, are themselves importantly constrained, limited in what they can offer by a whole host of institutional forces that have come so powerfully into play in health care over the past several decades. As she puts it: “The notion of patient autonomy is actually applied only within a narrow sphere—​decision-​making about specific medical treatments offered by individual physicians. Further constraint comes from a constellation of institutional imperatives. Thus, while patient and family decision-​making is considered paramount, decisions, when they are made at all, are constrained by hospital rules, reimbursement mechanisms, and standards of care” (2005: 28). What this creates, in the end, is a situation where patients and families are made to feel, sometimes crushingly so, far more responsibility for control over life, death, and the course of medical care than they perhaps ever really had.

From “Nothing More to Do” to “Nothing More We Should Do” A final key shift in the culture of death and dying in American hospitals to trace through the comparison across these two ethnographic studies is a move from understanding death as an event that comes when there is “nothing more to do” to experiencing it instead as a choice to be exercised when there is “nothing more we should do.” In the era that Glaser and Strauss describe, recognizing that there was “nothing more to do” was a largely practical question of knowing the limits of medical ability. The moment of recognizing that this limit had been reached was understood to mark the beginning of the dying process, a turning away from treatment and hope for recovery and toward the goal of easing the patient as comfortably (if often unwittingly) into death as possible. Importantly, the time leading up to this acknowledgement that dying had begun was typically characterized by a kind of waiting, as physicians and nurses carefully watched the patient for cues as to the certainty and the timing of death’s approach—​as well as for countercues pointing to the possibility for some form of recovery after all. And in a

Cultural Factors   303 context where death was so often treated as a secret to be concealed from patients and their families, this watchful waiting involved being alert not only to cues about whether and when death would come, of course, but to signs of the patient’s own state of awareness of the situation. Medical staff in Glaser and Strauss’s study described the sense of tension that characterized this period but also the feeling that the eventual outcome would—​and could—​unfold itself over time, that it was something to be observed rather than orchestrated. At this point, as one nurse succinctly put it: “Everybody is simply waiting” (1965: 23). Once the physician—​who was, in those times, uniquely empowered to do so—​ declared that the point of “nothing more to do” had been reached, a different form of waiting began: the deathwatch. Central to the notion of the deathwatch was a sense of inevitability, yet it is important to note that inevitability did not imply inactivity, for this was a form of waiting that could nonetheless require considerable work. Indeed, such waiting might be quite clinically and socially intensive, particularly for a patient believed to be only hours or days away from death, with nurses sometimes organizing shifts to be sure that a frightened patient would not die unaccompanied or in pain. Other patients, particularly those deemed to be responding to their coming death in inappropriate or unpleasant ways, might be left largely alone, deliberately isolated in a private room as far away as possible from the ongoing, life-​directed work of the ward. And for still other patients thought to be inevitably—​but not imminently—​dying, efforts might be made to shift them out of the hospital entirely, to their home or a nursing care facility. Such efforts reflected a familiar economic but also moral logic that hospital space and medical expertise should be dedicated to curative healing and reserved for those patients for whom recovery is still a possibility. Yet that logic did not always prevail. Patients themselves, as well as their family members, might push back fearfully and fiercely against such efforts to relocate the dying process, and waiting for death in these cases was sometimes allowed to unfold over weeks or even months of continuous hospitalization. Such drawn-​out forms of dying could raise questions of not just location but of timing as well, and various forms of hastening death might be used in cases where death was perceived to be taking “too long,” inflicting unnecessary suffering on both patients and those around them. Strategies such as withholding treatment for a secondary condition (antibiotics for pneumonia in a terminal cancer patient, for instance) or nudging up the dosages of pain medications to a quietly lethal level might be privately and independently employed by a physician (or, more rarely, a nurse) to influence the timing although not alter the certainty of death. In these days before the widespread use of life-​support technologies of ventilator and the like, such judgments about the appropriate temporal trajectory of dying (what counts as “too long”) were primarily a moral and emotional matter, rather than a matter of the technological entanglements, financial calculations, and bureaucratic requirements that—​as we shall see—​came to shape contemporary forms of dying in powerful ways. Whereas the 1960s’ era of “nothing more to be done” was marked by the limits of medicine to stave off death, in today’s contemporary hospital world, death is shaped instead by the need to set limits on medicine’s ability to unduly prolong life (Kaufman, 2005). This is

304   Megan Crowley-Matoka a world where death can no longer simply be waited for but rather must be decided. And it must be decided amidst the competing push and pull between those now considerable technological powers to extend the dying process (in some cases, almost indefinitely) and a relentless bureaucratic imperative to keep things “moving along” with ever greater efficiency through the hospital system. A matter both of morals (at what point does extending life become extending suffering?) and of money (at what point is “more life” no longer worth the cost?), this situation creates what Kaufman identifies as enormous “pressures to choose death ‘on time,’ ” pressures that operate on patients and families, as well as on medical professionals (2005: 174). Powerfully at work in the very notion of a proper “timeliness” for death are those pervasive cultural ideas, discussed previously, about what constitutes “good” and “bad” forms of death in the contemporary United States, such that technologically drawn-​out forms of dying are often perceived as unnatural, undignified, and filled with unnecessary suffering. These cultural ideas then frame decision-​making at the end of life, which has become less a question of whether there is anything else we can do (for, as mentioned, there is almost always some technology or medication to be tinkered with) and more a question of whether there is anything more we should do. Posed thus as a moral matter of what is right and not just a technical issue of what is medically possible, determining when dying has begun has now shifted in many ways from being a privilege of physician power to a weighty responsibility of patient autonomy. Yet also powerfully—​and perhaps somewhat more invisibly—​at work in those pressures to choose death “on time” are the institutional rules that govern what kinds of hospital care will be paid for, and for how long, and under what conditions of treatment patients are allowed to remain in the hospital. Contemporary structures of financing and bureaucratic management in medicine prioritize discrete, curative clinical procedures in ways that can work to push the dying process out of the hospital. The problem, succinctly put, is that “dying is not billable, you can’t treat it” (Kaufman, 2005: 91). Evident here are a set of cultural values about the goals of medicine as focused on cure rather than care. Although such values are hardly new in the culture of medicine—​indeed, they were expressed informally by those 1960s’ physicians trying to shift dying patients out of their hospitals—​what is new over the past several decades is the way that they have been formalized and sedimented into the institutional structures of how medical care is organized, delivered, and financed. Medical professionals are now subject to powerful institutional imperatives to keep dying patients moving through the hospital, to make sure that the patients who remain keep receiving some form of reimbursable treatment, and to guide patients and families toward the “choices” that will allow them to meet these imperatives. Perhaps paradoxically, however, these imperatives may work to help produce precisely the technologically intensive and drawn-​out forms of dying that have come to epitomize the problem of death in American hospitals, for in privileging treatment, procedures, and heroic intervention as the (reimbursable) goals of medicine, the institutional structures of health care have created a contemporary hospital world where waiting for death is no longer an option—​a world where there is never “nothing more to do.” Rather, as one intensive care unit nurse in Kaufman’s study so eloquently put

Cultural Factors   305 it: “The deathwatch has given way… . to heading full bore into strategies and techniques for mastering death by staving it off or choosing the time for it” (2005: 93).

Everything Changes and Everything Stays the Same? Thus through the strategic lens made possible by the comparative contrast over the forty-​ year span between Awareness of Dying and And a Time to Die, some of the key cultural features of contemporary death and dying in the United States do indeed come into sharper focus. Clearly a great deal has changed in the culture of death and dying in America, even profoundly so. The lack of a cultural script for dying in the 1960s highlights the power and consequences of our current compelling—​but often hard-​to-​live-​up-​to—​notions of what constitutes a good death. The fact that the image of physicians in the 1960s outright lying to their dying patients is now so shocking shows just how seismic the shift in medical culture from a presumption of physician control to a framework of patients’ rights (and the responsibilities they entail) has truly been. Perhaps most fundamentally, the way in which death is now a matter of deciding rather than of waiting—​and the degree to which this has come to be framed as a personal, moral choice rather than a professional, technical one—​emerges powerfully in that deceptively simple historical shift from death as a matter of “nothing more to do” to one of “nothing more we should do.” Yet, for all of this considerable, consequential cultural change in the beliefs, practices, and institutional structures of American death and dying, it is not at all clear that we have made as much progress as we might like to think across this forty-​year span. The “problem of death” posed by Glaser and Strauss so long ago still rings all too true in the complex contemporary hospital world that Kaufman evokes and explores: we continue to die today, it seems, in ways that still feel like too much technology and too little personal control.

References Alexander, M. 1980. The narrow embrace of the rigid house: Premature burial and the signs of death. The Hastings Center Report 10(3): 25–​31. Bernard, H. R. 1994. Research methods in anthropology. Thousand Oaks, CA: SAGE. Bettancourt, J. R. 2004. Cultural competence—​marginal or mainstream movement? New England Journal of Medicine 351(10): 953–​954. Boutin-​ Foster, C., J. C. Foster, and L. Konopasek. 2008. Physician, know thyself:  The Professional culture of medicine as a framework for teaching cultural competence. Academic Medicine 83(1): 106–​111. Callahan, D. 1995. Frustrated mastery:  The cultural context of death in America. Western Journal of Medicine 163(3): 226–​230.

306   Megan Crowley-Matoka Chapple, H. S. 2010. No place for dying: Hospitals and the ideology of rescue. Walnut Creek, CA: Left Coast Press. Cloud, J. 2000. A kinder, gentler death. Time 156(12): 60–​74. Conklin, B. 2001. Consuming grief compassionate cannibalism in an Amazonian society. Austin: University of Texas Press. Coolen, P. R. 2012. Cultural relevance in end-​of-​life care. Ethnomed.org. Accessed June 5, 2014. http://​ethnomed.org/​clinical/​end-​of-​life/​cultural-​relevance-​in-​end-​of-​life-​care Crawley, L., P. Marshall, and B. Koenig. 2001. Respecting cultural differences at the end of life. In Physicians’ guide to end of life care, edited by L. Snyder and T. Quill. Philadelphia, PA: American College of Physicians, 33–​55. Crawley, L., R. Payne, J. Bolden, T. Payne, P. Washington, and S. Williams. 2000. Palliative and end-​of-​life care in the African-​American community. JAMA: The Journal of the American Medical Association 284(19): 2518–​2521. Desjarlais, R. 2003. Sensory biographies:  Lives and deaths among Nepal’s Yolmo Buddhists. Berkeley, CA: University of California Press. Faunce, W. A., and R. L. Fulton. 1958. The sociology of death: A neglected area of research? Social Forces 36: 205–​209. Fox, R. C. 2005. Cultural competence and the culture of medicine. New England Journal of Medicine 353: 1316–​1319. Geertz, C. 1973. Interpretation of cultures. New York, NY: Basic Books. Glaser, B., and S. Strauss. 1965. Awareness of dying. Chicago: Aldine. Greiner, K. A., S. Perera, and J. S. Ahluwalla. 2003. Hospice usage by minorities in the last year of life:  Results from the National Mortality Followback Survey. Journal of the American Geriatric Society 51(7): 970–​978. Hall, M. J., S. Levant, and C. DeFrances. 2013. Trends in in-​hospital deaths: National hospital discharge survey, 2000–​2010. National Center for Health Statistics Data Brief 118. Hyattsville, MD:  US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. Johnson, K., M. Kuchibahtla, and J. Tulsky. 2008. What explains racial differences in the use of advance directives and attitudes toward hospice care? Journal of the American Geriatrics Society 56(10): 1953–​1958. Kagawa-​Singer, M. and L. J. Blackhall. 2001. Negotiating cross-​cultural issues at end-​of-​ life: “You got to go where he lives.” JAMA: The Journal of the American Medical Association 286(23): 2993–​3001. Kaufman, S. 2000. In the shadow of death with dignity: Medicine and cultural quandaries of the vegetative state. American Anthropologist 102(1): 69–​83. Kaufman, S. 2005. And a time to die:  How American hospitals shape the end of life. New York: Scribner. Kaufman, S., and L. Morgan. 2005. The anthropology of the beginnings and ends of life. Annual Review of Anthropology 34: 317–​341. Kaufman, S., P. S. Mueller, A. L. Ottenberg, and B. Koenig. 2011. Ironic technologies: Old age and the implantable cardioverter defibrillator in U.S. health care. Social Science and Medicine 72(1): 6–​14. Kaufman, S., A. Russ, and J. Shim. 2006. Aged bodies and kinship matters. American Ethnologist 33(1): 81–​99. Koenig, B., and J. Gates-​Williams. 1995. Understanding cultural difference in caring for dying patients. Western Journal of Medicine 163(3): 244–​249.

Cultural Factors   307 Koenig, B., and P. Marshall. 2014. Death: Cultural perspectives. In Bioethics (4th ed.), edited by B. Jennings. Farmington Hills, MI: Macmillan Reference, 727–​743. Lepowsky, M. 1985. Gender, aging and dying in an egalitarian society. In Aging and its transformations—​Moving toward death in Pacific societies, edited by D. A. Counts and D. R. Counts. Lanham, MD: University Press of America, 157–​178. Reagan, L. J. 2003. From hazard to blessing to tragedy:  Representations of miscarriage in twentieth-​century America. Feminist Studies 29(2): 356–​378. Rose, N. 1998. Inventing our selves:  Psychology, power and personhood. Cambridge, MA: Cambridge University Press. Rothman, D. 2001. The origins and consequences of patient autonomy: A 25-​year retrospective. Health Care Analysis 9: 255–​264. Rylko-​Bauer, B., and P. Farmer. 2002. Managed care or managed inequality? A call for critiques of market-​based medicine. Medical Anthropology Quarterly 16(4): 476–​502. Scheper-​Hughes, N. 1992. Death without weeping:  The violence of everyday life in Brazil. Berkeley: University of California Press. Siebold, C. 1992. The hospice movement: Easing death’s pains. Boston: Twayne. Teno, J., P. Gonzalo, J. Bynum, N. Leland, S. Miller, N. Morden, T. Scupp, D. Goodman, and V. Mor. 2013. Change in end-​of-​life care for Medicare beneficiaries: Site of death, place of care and health care transitions in 2000, 2005 and 2009. JAMA: The Journal of the American Medical Association 309(5): 470–​477. Timmermans, S. 2010. There’s more to dying than death: Qualitative research at the end of life. In The SAGE handbook of qualitative methods in health research, edited by R. Bourgeault, R. Dingwall, and R. deVries. Thousand Oaks, CA: SAGE, 19–​33. Waters, C. 2001. Understanding and supporting African-​Americans’ perspectives of end-​of-​ life care planning and decision making. Qualitative Health Research 11(3): 385–​398. Wijdicks, E. F.  M. 2001. The diagnosis of brain death. New England Journal of Medicine 344: 1215–​1222.

Chapter 19

Et hnicit y as a  Fac tor Kimberly S. Johnson and Ramona L. Rhodes

Introduction If projections are correct, by 2050 non-​Hispanic whites will no longer make up a majority of the US population. Due primarily to higher birth rates and immigration, the growth in the minority population will substantially outpace that of non-​Hispanic whites over the next four decades. Persons of African American (or black) race and Hispanic (or Latino) ethnicity, the two largest minority groups, currently make up 13% and 16% of the US population, respectively. By 2050, Hispanics are expected to represent 30% of the US population (Ortman and Guarneri, 2009). This projected increase in racial and ethnic diversity is not limited to children and young adults but also includes older Americans. Over the next two decades, the proportion of older adults who are racial and ethnic minorities will increase from 21% to 28%, with the growth in the proportion of African Americans and Hispanics expected to be two to three times higher than that of older non-​Hispanic whites (US Department of Health and Human Services, 2013a). With an increasingly diverse racial and ethnic population comes a myriad of cultures. “Culture” refers to integrated patterns of human behavior that include the language, thoughts, communications, actions, customs, beliefs, values, and institutions of racial, ethnic, religious, or social groups. Because culture significantly influences the experience of serious illness and death, including explanatory models of illness, communication, preferences for care, use of health-​care services, and quality of life, attention to cultural beliefs is fundamental to delivering high-​quality palliative care. The issue is especially challenging for a number of reasons. First, as the population grows more diverse, clinicians will increasingly find themselves caring for patients whose cultural backgrounds differ from their own. When clinicians lack skills at acknowledging and respecting the importance of cultural considerations in clinical care, such cross-​ cultural encounters are ripe for misunderstandings, conflict with patients and families, and lower quality care. Another issue is the wide variation in cultural beliefs, values,

Ethnicity as a Factor    309 and practices that exist even among individuals within the same racial or ethnic group. That is, the extent to which patients and families adhere to common cultural beliefs and practices will vary based on other factors, such as acculturation or socioeconomic status. Because of this variation, there is no substitute for attempting to learn about the values and preferences of individual patients and families in real-​time clinical encounters through thoughtful inquiry. Clinicians who make assumptions about individuals based solely on racial or ethnic identity risk stereotyping. However, even with this caveat, given the profound influence of culture on the experience of serious illness, some knowledge of beliefs, practices, and challenges that may be more common for one racial or ethnic group than for another provides an important framework for clinicians caring for culturally diverse groups of patients (Kagawa-​Singer and Blackhall, 2001; Crawley et al., 2002). This chapter focuses on the experience of those of African American (or black) race and Hispanic (or Latino) ethnicity, the two largest minority groups in the United States. For the most part, the term “African Americans” or “blacks” in this chapter refers to those with a shared history who are descendants of enslaved blacks within the boundaries of the present United States. However, some immigrants from African, Caribbean, Central American, and South American nations, and their descendants, may be identified or self-​identify with the term. The term “Hispanic” or “Latino” refers to native or foreign-​born persons from Latin America. Nearly two-​thirds of US Hispanics or Latinos are native born. Those of Mexican ancestry make up two-​thirds of Hispanices in the United States with significantly fewer reporting Puerto Rican, El Salvadoran, Cuban, or Dominican Republican, ancestry (Pew Hispanic Center, 2011). Although Latin American culture is diverse and influenced by country of origin and social class, there are some important commonalities of language and culture among US Hispanics or Latinos. In this chapter, we discuss access to and quality of palliative care for African Americans and Hispanics; highlight cultural beliefs, values, and preferences that may influence care; and suggest strategies and future directions for health-​care providers, organizations, policymakers, and researchers that may improve care for racially and ethnically diverse populations.

Disparities in the Quality of Health Care Racial and ethnic disparities in access to and quality of health care are widely documented for African Americans and Hispanics compared to non-​Hispanic whites. For example, compared to non-​Hispanic whites, both African Americans and Hispanics receive lower quality care and have higher rates of morbidity and mortality from HIV, diabetes mellitus, and some cancers (lung, prostate, stomach, and breast for African Americans; cervical and liver for Hispanics). They are also less likely to receive

310    Kimberly S. Johnson and Ramona L. Rhodes appropriate cardiac medications or procedures for coronary artery disease, and African Americans have a lower life-​expectancy than non-​Hispanic whites (National Research Council, 2003). Although both minority groups are less likely to have health-​care insurance or a usual source of care, these factors do not entirely account for health-​care disparities. Even when income, health insurance, and access to care are accounted for, many racial and ethnic disparities in health care persist, suggesting that other factors, such as health-​care providers’ bias, stereotyping, prejudice, and clinical uncertainty, may contribute (National Research Council, 2003). Despite national initiatives to reduce and eliminate health-​care disparities, there has been little improvement in many areas and worsening disparities in others (US Department of Health and Human Services, 2013b). Racial and ethnic disparities in care extend throughout the health-​care continuum and include not only disease prevention, early diagnosis, and curative treatment but also the care of patients living with advanced, serious illness. For example, compared to whites, bereaved caregivers of African Americans report less satisfaction with the overall quality of end-​of-​life care (Welch, Teno, and Mor, 2005). Other research documents lower quality care for minorities across multiple domains, including communication and pain management, lower rates of advance care planning, and lower rates of hospice enrollment (Hofmann, 1997; Gordon, 2006; Smith, 2007; Mack, 2010; Smith, 2009; Cintron, 2006; Carr, 2011; NHPCO, 2013; Medicare Payment Advisory Commission, 2013).

Communication Racial and ethnic minorities with serious illness and their families experience lower quality communication than non-​Hispanic whites. Compared to whites, bereaved family members of African Americans report more concerns with family support and provider communication, including the extent to which providers share information (Welch et al., 2005). Seriously ill African Americans (versus whites) rate physician communication as less informative, supportive, and partnering and are less likely to report that physicians elicited treatment preferences or listened to their concerns (Hofmann et al., 1997; Gordon et al., 2006; Smith, Davis, and Krakauer, 2007). Even when communication occurs, the outcomes may differ by race. For example, in a study of patients with advanced cancer, end-​of-​life discussions between African Americans and their doctors were less likely to result in care consistent with patients’ stated preferences compared to similar conversations between whites and their doctors (Mack et al., 2010; Loggers et al., 2009). Highlighting the importance of cultural considerations in communication in palliative care and similar to research in primary care settings, patients’ ratings of communication quality tend to be higher in race-​concordant (African American patient and African American physician) than race-​discordant (African American patient and white physician) encounters (Gordon, Street, Sharf, Kelly, et al., 2006; Cooper et al., 2003). Because minorities are significantly underrepresented in the clinician workforce

Ethnicity as a Factor    311 compared to the general population, these data further underscore the importance of clinicians’ knowledge and skills in navigating cross-​cultural encounters. Issues related to the quality of communication are further magnified in encounters between physicians and their Hispanic patients with limited English proficiency (LEP). Nearly three-​fourths of Hispanics who live in the United States speak Spanish at home, and fewer than one in four Hispanic immigrants reports being able to speak English very well. In contrast, the vast majority of US-​born Hispanics and later generations of Hispanics speak English fluently (Pew Hispanic Center, 2007b). Persons with LEP have less access to health care and are less satisfied with the quality of care. (Ngo-​Metzger et  al., 2007). In the absence of a readily available interpreter, clinicians may either attempt to communicate in English with Spanish-​speaking patients with LEP or use family members or untrained staff. Such practices are more likely to lead to errors in communication than when professional interpreters are used (Flores et al., 2010). These errors, which may include intentional or unintentional omissions of information and alterations in content, are particularly problematic in the types of conversations that are common in palliative care, including those involving breaking bad news and eliciting goals of care or treatment preferences. Therefore, whenever possible and especially during conversations about goals and preferences, encounters with patients of LEP should be facilitated by a professional interpreter (Smith, Sudore, and Perez-​Stable, 2009). Although the use of trained interpreters improves communication, even when an interpreter is involved clinicians may still provide less information and emotional support in encounters with language-​discordant seriously ill patients and their families, and errors in interpretation may still occur (Ngo-​Metzger et al., 2007; Thornton et al., 2009; Pham et al., 2008). As in race-​concordant patient–​physician interactions, patients with language-​concordant physicians report higher satisfaction with care than those with language-​discordant physicians (Ngo-​Metzger et al., 2007; Fernandez et al., 2004).

Pain Management Compared to whites, African Americans and Hispanics across age groups, diagnoses, and settings are less likely to have their pain appropriately assessed and treated. Prescribers are more likely to underestimate pain and less likely to treat pain based on established guidelines in seriously ill patients in these minority groups. Also, opioids may not be as accessible to patients who live in predominantly minority neighborhoods since pharmacies in these areas are less likely to stock adequate supplies of opioids than those in predominantly non-​Hispanic white neighborhoods (Anderson, Green, and Payne, 2009; Cintron and Morrison, 2006). Even when opioids are prescribed and available, African Americans and Hispanics may still be unable to obtain them because of lack of insurance and inability to afford the medications. Concerns related to addiction, hastening death, or fear of other side effects may also serve as a barrier to the appropriate use of these medications in seriously ill minorities (Rhodes et al., 2015).

312    Kimberly S. Johnson and Ramona L. Rhodes

Advance Care Planning African Americans and Hispanics are less likely than whites to complete an advance directive, to discuss their preferences for end-​of-​life care with family and friends, or to have a do not resuscitate order during hospitalization (Eleazer et  al., 1996; Kiely et al., 2001; Kwak and Haley, 2005; Morrison et al., 1998; Murphy et al., 1996; Carr, 2011; Carrion, Nedjat-​Haiem, and Marquez, 2013). One barrier to advance care planning for minorities is lack of knowledge. Compared to whites, African Americans and Hispanics have less knowledge of advance directives, and health-​care providers are less likely to engage them in advance care planning (Hofmann et al., 1997; Morrison et al., 1998; Murphy et al., 1996; Carr, 2011; Carrion et al., 2013; Carrion et al., 2013; Cohen et al., 2010; Carr, 2012; Volandes et al., 2008a, b; Fischer et al., 2012; Braun et al., 2008; Morrison and Meier, 2004; Kelley, Wenger, and Sarkisian, 2010; Waters, 2001). Even when minorities want to have discussions about their preferences for end-​of-​ life care, they may not do so if such communication is not initiated by healthcare providers (Hofmann et al., 1997; Morrison and Meier, 2004; Kelley, et al., 2010; Waters, 2001). Language barriers and limited education may further increase the difficulty in completing advance directives for Hispanics (Carr, 2012; Volandes et al., 2008a, b; Fischer et al., 2012). Additionally, certain beliefs that are more common among African Americans than whites may serve as a barrier to advance care planning, including the belief that planning for death may initiate the process of dying; discussions about preferences for end-​of-​life care should occur close to death; their health-​care wishes even if documented in an advance directives will not be honored; and they will receive lower quality care if they have an advance directive (Phipps, True, Harris, et al., 2003; Torke et al., 2005; Caralis et al., 1993; Bullock, 2006). As described later, mistrust of the health-​care system, spiritual beliefs, and preferences for care may also contribute to lower rates of advance care planning among minorities. African Americans and Hispanics place great value on the importance of family, which may include not only nuclear family but also distant relatives and fictive kin–​ friends or nonrelatives who have similar status to blood relatives. Both groups are more likely than whites to feel that decisions about end-​of-​life care should be made by their family, which is counter to the emphasis on patient autonomy in US culture (Smith et al., 2009; Kwak and Haley, 2005; Carrion, Nedjat-​Haiem, Martinez-​ Tyson, et al., 2013; Carr, 2012; Phipps, True, Harris, et al., 2003; Torke et al., 2005; Bullock, 2006; Born et al., 2004; Phipps, True, and Murray, 2003). Because African Americans and Hispanics more often defer decision-​making to family, they may believe that formal documents are not needed and that family will use their own judgment “when the time comes” (Smith et  al., 2009; Torke et  al., 2005; Phipps, True, and Murray, 2003). As such, a large number of family and friends may be present at the bedside or in family meetings and may make decisions about the patient’s care collectively.

Ethnicity as a Factor    313

Use of Hospice and Palliative Care Services African Americans and Hispanics enroll in hospice at lower rates than whites. Although African Americans currently make up 13.1% of the population, they account for only 8.6% of hospice patients nationwide. Hispanics make up 16.9% of the US population but only 6.9% of total hospice patients (US Census Bureau, 2012; National Hospice and Palliative Care Organization, 2013). Similarly, among Medicare beneficiaries, the largest group of hospice enrollees, 47% of whites who died in 2011 used hospice compared to 35% of African Americans and 38% of Hispanics (Medicare Payment Advisory Commission, 2013). A number of factors may explain the lower use of hospice services by minorities. First, African Americans and Hispanics have less knowledge of hospice services than whites, including information about services provided, eligibility, and how hospice care is funded (Born et al., 2004; Colon, 2005; Selsky et al., 2012; Rhodes, Teno, and Welch, 2006; Reese et al., 1999; Ludke and Smucker, 2007; Johnson, Kuchibhatla, and Tulsky, 2009). Research also suggests that African Americans are often not informed about hospice as an option for end-​of-​life care and as a group have less favorable beliefs about hospice care, including the belief that hospice causes one to die before one’s time and the beliefs that those who enroll in hospice get “no treatment” (Rhodes et al., 2006; Reese et al., 1999; Ludke and Smucker, 2007; Johnson et al., 2009; Rosenfeld et al., 2007). For Hispanics with LEP, language may limit providers’ abilities to communicate with and provide comfort and support to patients and families, which may lead to underutilization of services. For both groups, the expectation that family members and friends will provide care at the end of life may also serve as a barrier to hospice enrollment because of concerns about hospice staff interfering with this care (Born et al., 2004). Additionally, although they are a tremendous source of assistance with physical care and provide a great deal of psychosocial and emotional support, sometimes family members themselves may serve as a barrier to hospice enrollment, especially when their preferences for care differ from that of the patient. For example, even when patients are ready to move forward with care focused on comfort, they may continue to pursue aggressive treatment because they do not wish to disappoint family members. Finally, because of mistrust of the health-​care system and spiritual beliefs, African American and Hispanic patients and their family members may be less willing to accept a poor prognosis, which may make discussions about end-​of-​life care and hospice even more challenging (Smith et al., 2009; Rhodes et al., 2015). Given improved outcomes for patients and families who enroll in hospice compared to those who receive conventional care, increasing hospice enrollment for minorities may reduce some disparities in the quality of end-​of-​life care (Baer 2000; Teno, Clarridge, et al., 2004)). For example, among African Americans who enroll in hospice, some disparities in care, such as family ratings of overall satisfaction and concerns about spiritual or emotional support and communication, are reduced when compared to the

314    Kimberly S. Johnson and Ramona L. Rhodes general population (Rhodes, Teno, Connor, 2007). Additionally, hospice enrollment may provide minorities with access to services that may otherwise be difficult for them to obtain. African American cancer patients, for example, endorse a greater need than whites for the kinds of services that hospice provides (Fishman, 2009). Although African American and Hispanic hospice enrollees have similar or longer lengths of stay than whites, there is some evidence that the experience of minority hospice enrollees, specifically African Americans, may differ from that of whites (Park et al., 2012; Colón and Lyke, 2003; Johnson, Kuchibhatla, and Tulsky, 2011). African Americans are more likely to disenroll from hospice to pursue potentially life-​prolonging interventions like CPR, mechanical ventilation, and invasive medical therapies (Johnson, Kuchibhatla, Tanis, et al., 2008; Kapo, MacMoran, and Casarett, 2005). This is important because there is evidence that receipt of such interventions, while common among hospice disenrollees, may not significantly increase life expectancy but may reduce quality of remaining life (Carlson et al., 2010). Also, hospices with a large proportion of African American enrollees receive lower ratings from bereaved family on care coordination and overall quality (Rhodes, Xuan, and Halm, 2012). While there is currently little data, small studies of home-​based and in-​patient palliative care suggest similarly favorable outcomes for whites and African Americans who use these services, including increased satisfaction, greater rates of home deaths and hospice referrals, and increased documentation of treatment preferences (Ciemins et al., 2006; Zaide et al., 2012; Holley et al., 2009).

Cultural Beliefs, Values, and Preferences As in other areas of health care, racial and ethnic disparities in the quality of palliative and end-​of-​life care are multifactorial. Provider factors such as bias, stereotypes, and poor communication, as well as health-​care policies that fail to consider cultural diversity, may contribute to disparities in health-​care access and quality (National Research Council, 2003). Patient factors are also important. Compared to whites, African Americans and Hispanics have lower incomes, lower levels of education, lower rates of health insurance coverage, and lower levels of health literacy—​all of which are associated with poorer health outcomes and less access to care (National Research Council, 2003; US Department of Health and Human Services, 2014a, 2014b; National Center for Education Statistics, 2006). Health literacy is the degree to which individuals are able to obtain, process, and understand basic health information and services needed to make appropriate health decisions (National Center for Health Statistics, 2012). Lower levels of health literacy are associated with less use of preventive services, less knowledge of medical conditions and treatments, higher rates of uninsurance, higher rates of hospitalization, and poorer self-​reported health status (Williams et al., 1998; Baker et al.,

Ethnicity as a Factor    315 1998; Baker et al., 1997; Scott et al., 2002; Bennet et al., 1998). In fact, there is some evidence that education and health literacy are more strongly associated with preferences for end-​of-​life care than race and that the use of video images in advance care planning may address these factors (Volandes et al., 2008a, 2008b). Patient factors, which limit access to palliative care, are further compounded by language barriers and immigration status for some Hispanics. Those who are undocumented may not be as forthcoming with medical information or health concerns because of fear of deportation (Smith et al., 2009). In addition to sociodemograhic factors, cultural beliefs, values, and preferences are also important determinants of health and use of health-​care services. These factors often shape the experience of illness and death and guide medical decision-​making (US Department of Health and Human Services, 2014c). Among those that are especially relevant to palliative and end-​of-​life care for minorities are preferences for care, spiritual beliefs, and mistrust of the health-​care system (Kagawa-​Singer and Blackhall, 2001; Crawley et al., 2002; Smith et al., 2009; Crawley et al., 2000; Krakauer, Crenner, and Fox, 2002; Johnson, Kuchibhatla, and Tulsky 2008).

Preferences for End-​of-​Life Care Compared to whites, African Americans and Hispanics are more likely to want life-​ sustaining therapies, such as cardiopulmonary resuscitation and mechanical ventilation, in the face of poor prognosis, including brain damage or if experiencing a terminal illness with very limited life-​expectancy (Krakauer et  al., 2002; Barnato et  al., 2009; Blackhall et al., 1999). Additionally, these minority groups are more likely than whites to want to die in the hospital and less likely to want potentially life-​shortening palliative drugs. These differences persist even when controlling for sociodemographic factors and beliefs about effectiveness of resuscitation (Barnato et al., 2009). Similar preferences for more aggressive care in the event of catastrophic illness exist among African American physicians compared to white physicians, further highlighting how issues of shared culture rather than solely differences in sociodemographics and medical knowledge may impact end-​of-​life decision-​making (Mebane et al., 1999). Consistent with preferences for more aggressive therapies at the end of life, African Americans and Hispanics incur higher costs and receive more expensive, life-​prolonging care in the last months of life despite less expensive, preventive, or cure-​directed therapies earlier in the life-​course. Specifically, they are more likely to die in the hospital, be admitted to the intensive care unit (ICU), and receive intensive therapies such as ICU care and gastrostomies than whites (Hanchate et al., 2009). Because hospice focuses on comfort rather than cure and advance directives tend to direct healthcare providers to limit life-​prolonging therapies, greater preferences for life-​prolonging therapies may partly explain the lower use of hospice care and completion of advance directives by these minority groups (Johnson, Kuchibhatla, and Tulsky, 2008).

316    Kimberly S. Johnson and Ramona L. Rhodes Although much has been written about the more aggressive treatment preferences of minorities at the end of life, a few important points are worth noting. First, while African Americans and Hispanics as a group are more likely to endorse preferences for more aggressive end-​of-​life care than whites, many endorse preferences consistent with the hospice and palliative care philosophy of care. For example, in a national study of Medicare beneficiaries, more African Americans and Hispanics than whites reported wanting to die in the hospital; however, 82% of African Americans and 85% of Hispanics did not want to die in the hospital (Barnato et al., 2009). Second, expressed preferences for care may not always match the care received, especially for minorities. Research suggests that African Americans may be less likely to receive the end-​of-​life care they state they would prefer regardless of whether those preferences are for aggressive care or care focused primarily on comfort (Loggers et al., 2009; Borum, Lynn, and Zhong, 2000).

Spiritual Beliefs Spirituality and faith are tenants that are especially central to many members of the African American community, and some African American patients rely heavily on their faith in God during times of illness and distress. African Americans are more likely than whites to participate in religious activities and to use religion to cope with illness (Sahgal and Smith, 2009). African Americans are also more likely to believe that God is responsible for physical and spiritual health, that divine intervention and miracles occur, and that there are religious prohibitions against physician-​assisted death or advance directives limiting life-​sustaining therapies (Johnson, Elbert-​Avila, and Tulsky, 2005). As such, they may believe that completing an advance directive conflicts with their belief that God will take care of them or that it will go against God’s will. Spiritual beliefs may also guide African Americans in their decisions about hospice and may be perceived to conflict with the hospice philosophy of care (Johnson, Kuchibhatla, and Tulsky, 2008). Choosing hospice may be viewed as giving up on or losing faith in God’s ability to heal or to perform miracles, and poor health or death may be viewed as a consequence of this loss of faith (Bullock, 2006; Reese et al., 1999; Crawley et al., 2000; Krakauer et al., 2002; Johnson et al., 2005). Similar to African Americans, a majority of Hispanics endorse the importance of religion, and some religious beliefs may influence preferences for more aggressive care at the end of life (Pew Hispanic Center, 2007a). For example, common to both African Americans and Hispanics is a belief that suffering is redemptive and should be endured as part of a test of faith (Smith et al., 2009; Crawley et al., 2000). Such beliefs may seem to conflict with the hospice philosophy of care, which emphasizes the importance of aggressive symptom management and comfort. In general, research suggests that those who endorse greater religious coping or faith receive more aggressive care at the end of life, even in analyses adjusting for racial and ethnic differences (Phelps et al., 2009).

Ethnicity as a Factor    317

Mistrust in the Health-​Care System Mistrust in the health-​care system has been identified as a barrier to use of hospice care and completion of advance directives by some African Americans (Crawley et al., 2002; Crawley et al., 2000; Johnson, Kuchibhatla, and Tulsky, 2008; Cort, 2004). Compared to whites, minorities endorse less trust in the health-​care system (Smith et al., 2009; Born et al., 2004; Boulware et al., 2003; LaVeist, Nickerson, and Bowie, 2000; Rose et al., 2004; Lillie-​Blanton et al., 2000). African Americans and Hispanics are more likely to believe that they receive lower quality care than whites (Lillie-​Blanton et al., 2000). For African Americans, historical occurrences such as the Tuskegee Syphilis Study and other negative experiences may create a sense in some patients that health-​care providers do not always have their best interests in mind (Crawley et al., 2000; Gamble, 1997). As a result, some African Americans may feel that completion of an advance directive or living will is an indication that options for treatment will not be offered, that they will be withdrawn, or that their wishes, though documented, will not be honored or respected (Wicher and Meeker, 2012). Historical racial segregation, discrimination, denial of services, and ongoing disparities in care may also influence African Americans patients to choose aggressive treatment despite its futility, instead of care that is focused on comfort, symptom management, and support such as hospice. They may feel that health-​care providers are not offering all options available or limiting treatment because of their race (Crawley et al., 2000; Krakauer et al., 2002; Wicher and Meeker, 2012). A significant proportion of Hispanics also report that they have experienced discrimination in health care and describe negative patient–​provider interactions. This perceived discrimination along with additional challenges imposed by language barriers and cultural insensitivity may engender mistrust in hospice providers and lead to underutilization of hospice services and a reluctance to complete advance directives (Smith et al., 2009; Lauderdale et al., 2003).

Navigating Cross-​Cultural Encounters: Strategies for  Health-​C are Providers Communication with patients about advance care planning, palliative care, and end-​of-​life care requires an examination of patient preferences and overall goals of care. This can be further nuanced by the consideration of specific racial, ethnic, or cultural beliefs, values, and practices. Given that a lack of cultural sensitivity is a potential barrier to high-​quality end-​of-​ life care for members of the African American and Hispanic/​Latino communities, health-​ care providers must make efforts to provide culturally sensitive care to patients of diverse racial and ethnic backgrounds (Kagawa-​Singer and Blackhall, 2001; Crawley et al., 2002; Smith et al., 2009). The following recommendations are also summarized in Table 19.1.

Table 19.1 Strategies for Effective Cross-​Cultural Care Goals of Care and Treatment Strategies Options ACP (living will, advance directive, medical power of attorney)

Describe the goals of ACP, emphasizing that • ​  ACP helps patients to make their wishes for care known to their loved ones and health-​care providers. •​  ACP helps make decision-​making less difficult and stressful for loved ones because the patient’s wishes are known and documented. •​  ACP does not hasten death. Everyone should have an advance care plan regardless of age or health status. Initiate discussions as early as possible and before a crisis develops (i.e., early in the disease trajectory, at routine health-​care visits) Even if patients are reluctant to complete a formal document, • ​  Include information about their stated preferences for care in the medical record. •  Encourage them to choose a family member or trusted friend to make decisions for them if they become incapacitated and to share their wishes for care with that person. •​  Emphasize the importance of having an advocate or spokesperson.

Palliative Care

Describe the goals of palliative care, emphasizing that •   Palliative care helps to treat symptoms that are associated with certain illnesses and their treatment, such as pain, shortness of breath, nausea, and vomiting. •​  Palliative and curative care can be complementary; they can work together to improve health and well-​being.

Hospice

Describe the goals of hospice, emphasizing that •   Hospice is not a withdrawal of all care; it is a change in the focus of care from cure to comfort. •​  Hospice uses a team approach to provide care and support to the patient and the patient’s family, including doctors, nurses, social workers, nurse aids, etc. •​  Hospice is not “giving up” or “losing faith” in God. In fact, spiritual support is a main component of hospice care. A chaplain is available for spiritual support should the need arise. •​  Hospice staff will not take the place of care provided by family but will support the family as they care for the patient

Specific Barriers Conflict with Spiritual Beliefs

Take a spiritual history. Ask the patient about spiritual beliefs that may influence decision-​making. Ask the patient and/​or family how the health-​care team can best support and meet their spiritual needs. Offer a visit from pastoral care or a chaplain Be open to involvement of members of patients’ religious community if the patient requests this. (continued)

Ethnicity as a Factor    319 Table 19.1 Continued Goals of Care and Treatment Strategies Options Cultural Beliefs

Ask the patient and/​or family if there are specific cultural traditions that they would like to practice or observe and allow them to observe those practices if possible.

Family Involvement

Ask patients how they would like health information and medical decision-​making communicated to them. If patients defer to family, identify one family member who will act as point person for communication about the care plan and medical decision-​making. Arrange a family meeting to discuss prognosis, goals of care, and treatment strategies.

Language Barriers/​Health Literacy

Ask about preferred language for communication. Use a medical interpreter for all communication if not fluent in the patient’s primary language. When available, provide written materials in the patient’s preferred language. Avoid complicated medical terminology when explaining things to the patient and/​or family. Ask the patient and/​or family if they have questions about their condition, prognosis, treatment, etc. Ask the patient and/​or family their understanding of what is going on (i.e., severity of illness, treatment plan, etc.). This may help to gauge their level of comprehension.

Desire for Aggressive Treatment Despite Poor Prognosis/​Medical Mistrust

Have open and honest communication with the patient and family about prognosis and goals of care. Make the patient and/​or family members aware of their options for care at the end of life (palliative care, hospice, etc.). Do not pressure patients into making decisions but revisit the discussion periodically, particularly as the patient’s condition worsens or symptoms progress. Respect the patient and/​or family’s wishes for treatment. The goal is for them to make an informed decision, which may not always be congruent with the provider’s recommendations. Negotiate common goals of care Discuss and consider time-​limited trials of aggressive care based on patient/​family goals and preferences.

Note: ACP = advance care planning.

Avoiding Stereotyping: A Tailored Approach The strategies discussed here are general recommendations that health-​care providers may employ to address the concerns of patients of diverse racial and ethnic backgrounds who may benefit from palliative or hospice care. However, providers should

320    Kimberly S. Johnson and Ramona L. Rhodes tailor their approach to the specific cultural, spiritual, or religious beliefs, values, needs, and concerns of individual patients and families in real time. To identify these important cultural and spiritual beliefs, providers should take a cultural and spiritual history (Kagawa-​Singer and Blackhall, 2001). This includes asking patients and families how the medical team can best meet their cultural and spiritual needs, inquiring about specific cultural practices that are observed, and enlisting the services of pastoral care or chaplaincy if needed or desired. Providers should also be open to involvement of members of the patient’s religious community (i.e., the patient’s pastor, priest, rabbi, etc.) and allow or facilitate patients’ observance of specific cultural or spiritual practices if possible.

Discussing Advance Care Planning Advance care planning and treatment options such as palliative care and hospice services should be discussed with patients and families in a way that informs them of appropriate options for care, allays fears, and addresses previously identified barriers. For instance, patients should know that the goal of advance care planning is to assist them in making their preferences for care known not only to family members but to health-​ care providers. Providers should emphasize that advance care planning does not hasten death. Patients may be more open to advance care planning when it is addressed early in the disease trajectory and during routine office visits—​preferably with a provider with whom the patient has had a long-​standing, trusting relationship (Waters, 2001). Even if patients are reluctant to complete formal documents (i.e., living will, advance directive, or medical power of attorney), health-​care providers should document their discussions with patients in the medical record. They should also encourage patients to choose a family member or trusted friend to make decisions for them, should they not be able to do so. To demonstrate support and respect for patients’ interests, providers may explain the importance of having someone who can advocate on their behalf, knows their preferences for care, and can communicate those wishes to the health-​care team.

Discussing Hospice and Palliative Care Services Some strategies may also help health-​care providers communicate with patients and families about palliative care and hospice care and dispel some of the myths that are associated with these services. Providers should emphasize that the goals of palliative care are to alleviate symptoms that are associated with certain illnesses and their treatment, including pain, dyspnea, nausea, vomiting, and anxiety. Palliative care can be complementary to curative or ongoing treatment; these treatment alternatives can work together to improve the patient’s overall health and well-​being. If and when the patient’s disease progresses beyond curative treatment, hospice can be introduced to the patient and family as an option for care. Hospice should never be referred to as a withdrawal of all care, as patients may view this as being abandoned or fear that clinicians are not

Ethnicity as a Factor    321 offering appropriate care because of their race or ethnicity. Instead, it should be referred to as a change in focus of care from cure to comfort. Patients and families should be made aware that hospice consists of a multidisciplinary team of doctors, nurses, nurses aids, and social workers who work together to provide the patient and family with care and support. Those who have an objection to hospice based on religious or spiritual beliefs should be informed that spiritual support is also a component of hospice care and that chaplains are available for spiritual support when needed.

Caring for Patients with Limited English Proficiency and Low Health Literacy Because of the importance of adequate communication to the care of patients with serious illness, health-​care providers must work to overcome barriers imposed by limited health literacy and LEP, both of which are more common for minorities. The preferred language for communication should be identified for patients who are not native English speakers. For those with LEP or who prefer to communicate in a language other than English, a medical translator—​not a family member or friend—​should be used for all communication. Patients should be informed of their legal right to a trained medical interpreter at no cost (Smith et al., 2009; US Department of Health and Human Services, 2001). If printed information is available in the patients’ preferred language, it should be provided. In all instances, complicated medical terminology should be avoided. To gauge patients’ and families’ knowledge of the treatment plan, providers should ask the patients their understanding of the disease process or plan of care. One way to do this is to ask the patient to describe the treatment plan in his or her own words.

Responding to Requests for Aggressive Treatment Despite Poor Prognosis and Limited Life Expectancy Patients and families sometimes have a desire for aggressive treatment despite limited life expectancy and poor prognosis. Health-​care providers must engage in open and honest communication with them about the overall prognosis and goals of care and attempt to negotiate common goals of care when conflicts arise. Patients and families should be made aware of their options for end-​of-​life care; however, they should not feel pressured into making decisions. Goals of care should be revisited periodically—​ particularly as the patient’s condition worsens. Sometimes a mutually agreed-​on, time-​ limited trial of aggressive care may help to build trust and reduce conflict between the medical team and family members regarding the use of life-​sustaining therapies. Arrangement of a family meeting may be helpful and allow providers to discuss prognosis, goals or care, and treatment strategies with family members involved in decision-​ making. Any consultants or subspecialist providers, clergy, social workers, and other

322    Kimberly S. Johnson and Ramona L. Rhodes health-​care team members should be invited to participate, as they can provide an overview of the care plan from their perspectives. Health-​care providers should work to accommodate the preferred decision-​making process of patients and families, which in some cases may include consensus of a large number of relatives and nonrelatives. While large numbers of family members may attend family meetings for formal updates, it is appropriate for clinicians to identify one person to contact for more frequent updates. Ultimately, the goal should be for patients and families to make an informed decision about the care that they receive; health-​care providers must understand and realize that, in some instances, this will not be congruent with their recommendations.

Future Directions Disparities in end-​of-​life care among members of the African American and Hispanic/​ Latino communities have been identified throughout the literature, and, as noted previously, many reasons for these disparities exist. Health-​care provider strategies for addressing these racial and ethnic disparities in care have been discussed; however, moving forward, a concerted effort should be made by health-​care organizations, researchers, and policymakers to address these barriers as well.

Health-​Care Organizations Organizations and health systems can aid in reducing barriers to end-​of-​life care among persons of diverse racial and ethnic backgrounds. The US Department of Health and Human Services has developed Standards for Culturally and Linguistically Appropriate Services, which mandate that health-​care organizations work to improve access to care for diverse populations by establishing policies that reduce barriers to care, practices that enhance patients’ knowledge of available services, staff that reflect the diversity of the service area, and community partnerships that facilitate service delivery (US Department of Health and Human Services, 2001). Based on these standards, organizations and health systems should provide educational programs to staff on cultural sensitivity at the end of life and institute hiring practices that will increase racial and ethnic diversity among health-​care providers across the spectrum of care. Organizations and health systems may also consider creating outreach programs that target members of diverse communities that will educate them about advance care planning and options for care at the end of life or working with community leaders to create these programs. These initiatives may increase cultural sensitivity among providers and address issues of medical mistrust and have been conducted successfully in some settings (Phipps, True, and Murray, 2003; Lyke and Colon, 2004; Reese, Melton, and Ciaravino, 2003).

Ethnicity as a Factor    323

Research and Policy Although many studies that have identified racial and ethnic disparities in completion of advance directives, use of palliative care, and hospice enrollment and identified reasons for the underutilization of these services in African Americans and Hispanics/​Latinos communities, few studies have focused on the design and implementation of interventions that can combat disparities in end-​of-​life care. Some research has been promising, and targeted interventions have been developed to address disparities in advance care planning and palliative care for seriously ill African Americans and Latinos, including patient navigation, peer support, and video education materials (Volandes et al., 2008a, b; Bullock, 2006; Volandes et al., 2008a, b; Fisher, Sauaia, and Kutner, 2007; Hanson et al., 2013); however, more efforts are needed to design, implement, test, and disseminate effective, culturally sensitive interventions that these disparities. These interventions could then inform policy regarding end-​of-​life care for the underrepresented and underserved.

References Anderson, K. O., C. R. Green, and R. Payne. (2009). “Racial and ethnic disparities in pain: Causes and consequences of unequal care.” Journal of Pain 10(12): 1187–​1204. Baer, W. M., and L. C. Hanson. (2000). “Families’ perception of the added value of hospice in the nursing home.” Journal of the American Geriatrics Society 48(8): 879–​882. Baker, D. W., R. M. Parker, M. V. Williams, and W. S. Clark. (1997). “The relationship of patient reading ability to self-​reported health and use of health services.” American Journal of Public Health 87(6): 1027–​1030. Baker, D. W., R. M. Parker, M. V. Williams, and W. S. Clark. (1998). “Health literacy and the risk of hospital admission.” Annals of Internal Medicine 13(12): 791–​798. Barnato, A. E., D. L. Anthony, J. Skinner, P. M. Gallagher, and E. S. Fisher. (2009). “Racial and ethnic differences in preferences for end-​of-​life treatment.” Annals of Internal Medicine 24(6): 695–​701. Bennet, C. L., M. R. Ferreira, T. C. Davis, et al. (1998). “Relation between literacy, race, and state of presentation among low-​income patients with prostate cancer.” Journal of Clinical Oncology 16(9): 3101–​3104. Blackhall, L. J., G. Frank, S. T. Murphy, V. Michel, J. M. Palmer, and S. P. Azen. (1999). “Ethnicity and attitudes towards life sustaining technology.” Social Science & Medicine 48(12): 1779–​1789. Born, W., K. A. Greiner, S. E. Butler, and J. Ahluwalia. (2004). “Knowledge, attitudes, and beliefs about end-​of-​life care among inner-​city African Americans and Latinos.” Journal of Palliative Medicine 7(2): 247–​256. Borum, M. L., J. Lynn, and Z. Zhong. (2000). “The effects of patient race on outcomes in seriously ill patients in SUPPORT: An overview of economic impact, medical intervention, and end-​of-​life decisions. Study to understand prognoses and preferences for outcomes and risks of treatments.” Journal of the American Geriatrics Society 48(Suppl. 5): S194–​S198. Boulware, L. E., L. A. Cooper, L. E. Ratner, T. A. LaVeist, and N. R. Powe. (2003). “Race and trust in the health care system.” Public Health Report 118: 358–​365.

324    Kimberly S. Johnson and Ramona L. Rhodes Braun, U. K., R. J. Beyth, M. E. Ford, and L. B. McCullough. (2008). “Voices of African American, Caucasian, and Hispanic surrogates on the burdens of end-​of-​life decision making.” Annals of Internal Medicine 23: 267–​274. Bullock, K. (2006). “Promoting advance directives among African Americans: A faith-​based model.” Journal of Palliative Medicine 9: 183–​195. Caralis, P. V., B. Davis, K. Wright, and E. Marcial. (1993). “The influence of ethnicity and race on attitude toward advance directives, life-​prolonging treatments, and euthanasia.” Journal of Clinical Ethics 4(2): 155–​165. Carlson, M. D., J. Herrin, Q. Du, et al. (2010). “Impact of hospice disenrollment on health care use and Medicare expenditures for patients with cancer”. Journal of Clinical Oncology 28(28): 4371–​4375. Carr, D. (2011). “Racial differences in end-​of-​life planning: Why don’t blacks and Latinos prepare for the inevitable?” Omega 63: 1–​20. Carr, D. (2012). “Racial and ethnic differences in advance care planning: Identifying subgroup patterns and obstacles.” Journal of Aging and Health 24: 923–​947. Carrion, I. V., F. R. Nedjat-​Haiem, and D. X. Marquez. (2013). “Examining cultural factors that influence treatment decisions: A pilot study of Latino men with cancer.” Journal of Cancer Education 28: 729–​737. Carrion, I. V., F. R. Nedjat-​Haiem, D. Martinez-​Tyson, and H. Castaneda. (2013). “Advance care planning among Colombian, Mexican, and Puerto Rican women with a cancer diagnosis.” Support Care in Cancer 21: 1233–​1239. Ciemins, E. L., B. Stuart, R. Gerber, J. Newman, and M. Bauman. (2006). “An evaluation of the Advanced Illness Management (AIM) Program: Increasing hospice utilization in the San Francisco Bay area.” Journal of Palliative Medicine 9: 1401–​1411. Cintron, A., and R. S. Morrison. (2006). “Pain and ethnicity in the United States: A systematic review.” Journal of Palliative Medicine 9(6): 1454–​1473. Cohen, M. J., J. B. McCannon, S. Edgman-​Levitan, and W. A. Kormos. (2010). “Exploring attitudes toward advance care directives in two diverse settings.” Journal of Palliative Medicine 13: 1427–​1432. Colon, M. (2005). “Hospice and Latinos: A review of the literature.” Journal of Social Work in End-​of-​Life & Palliative Care 1(2): 27–​43. Colón, M., and J. Lyke. (2003). “Comparison of hospice use and demographics among European Americans, African Americans, and Latinos.” American Journal of Hospice & Palliative Care 20(3): 182–​190. Cooper, L. A., D. L. Roter, R. L. Johnson, D. E. Ford, D. M. Steinwachs, and N. R. Powe. (2003). “Patient-​centered communication, ratings of care, and concordance of patient and physician race.” Annals of Internal Medicine 139(11): 907–​915. Cort, M. A. (2004). “Cultural mistrust and use of hospice care:  Challenges and remedies.” Journal of Palliative Medicine 7(1): 63–​7 1. Crawley, L. M., P. A. Marshall, B. Lo, and B. A. Koenig. (2002). “Strategies for culturally effective end-​of-​life care.” Annals of Internal Medicine 136(9): 673–​679. Crawley, L., R. Payne, J. Bolden, T. Payne, P. Washington, and S. Williams. (2000). “Palliative and end-​of-​life care in the African American community.” JAMA 284(19): 2518–​2521. Eleazer, G. P., C. A. Hornung, C. B. Egbert, et al. (1996). “The relationship between ethnicity and advance directives in a frail older population.” Journal of the American Geriatrics Society 44: 938–​943.

Ethnicity as a Factor    325 Fernandez, A., D. Schillinger, K. Grumbach, et al. (2004). “Physician language ability and cultural competence. An exploratory study of communication with Spanish-​speaking patients.” Journal of General Internal Medicine 19(2): 167–​174. Fischer, S. M., A. Sauaia, S. J. Min, and J. Kutner. (2012). “Advance directive discussions: Lost in translation or lost opportunities?” Journal of Palliative Medicine 15: 86–​92. Fisher, S. M., A. Sauaia, and J. S. Kutner. (2007). “Patient navigation: A culturally competent strategy to address disparities in palliative care.” Journal of Palliative Medicine 10: 1023–​1028. Fishman, J., P. O’Dwyer, H. L. Lu, H. R. Henderson, D. A. Asch, and D. J. Casarett. (2009). “Race, treatment preferences, and hospice enrollment:  Eligibility criteria may exclude patients with the greatest needs for care.” Cancer 115(3): 689–​697. Flores, G., M. Abreu, C. P. Barone, R. Bachur, and H. Lin. (2010). “Errors of medical interpretation and their potential clinical consequences: A comparison of professional versus ad hoc versus no interpreters.” Annals of Emergency Medicine 60(5): 545–​553. Gamble, V. N. (1997). “Under the shadow of Tuskegee: African Americans and health care.” American Journal of Public Health 87: 1773–​1778. Gordon, H. S., R. L. Street, Jr., B. F. Sharf, P. A. Kelly, and J. Souchek. (2006). “Racial differences in trust and lung cancer patients’ perceptions of physician communication.” Journal of Clinical Oncology 24(6): 904–​909. Gordon, H. S., R. L. Street, Jr., B. F. Sharf, and J. Souchek. (2006). “Racial differences in doctors’ information-​giving and patients’ participation.” Cancer 107(6): 1313–​1320. Hanchate, A., A. C. Kronman, Y. Young-​Xu, A. S. Ash, and E. Emanuel. (2009). “Racial and ethnic differences in end-​of-​life costs: Why do minorities cost more than whites?” Archives of Internal Medicine 169(5): 493–​501. Hanson, L. C., T. D. Armstrong, M. A. Green, et al. (2013). “Circles of care: Development and initial evaluation of a peer support model for African Americans with advanced cancer.” Health Education & Behavior 40: 536–​543. Hofmann, J. C., N. S. Wenger, R. B. Davis, et al. (1997). “Patient preferences for communication with physicians about end-​of-​life decisions. SUPPORT Investigators. Study to understand prognoses and preferences for outcomes and risks of treatment.” Annals of Internal Medicine 127(1): 1–​12. Holley, A. P.  H., R. Gorawara-​Bhat, W. Dale, J. Hemmerich, and D. Cox-​Hayley. (2009). “Palliative access through care at home: Experiences with an urban, geriatric home palliative care program.” Journal of the American Geriatrics Society 57: 1925–​1931. Johnson, K. S., K. I. Elbert-​Avila, and J. A. Tulsky. (2005). “The influence of spiritual beliefs and practices on the treatment preferences of African Americans: A review of the literature.” Journal of the American Geriatrics Society 53(4): 711–​7 19. Johnson, K. S., M. Kuchibhatla, and J. A. Tulsky. (2008). “What explains racial differences in the use of advance directives and attitudes towards hospice care?” Journal of the American Geriatrics Society 56(10): 1953–​1958. Johnson, K. S., M. Kuchibhatla, and J. A. Tulsky. (2009). “Racial differences in self-​reported exposure to information about hospice care.” Journal of Palliative Medicine 12(10): 921–​927. Johnson, K. S., M. Kuchibhatla, and J. A. Tulsky. (2011). “Racial differences in location before hospice enrollment and association with hospice length of stay.” Journal of the American Geriatrics Society 59(4): 732–​737. Johnson, K. S., M. Kuchibhatla, D. Tanis, and A. Tulsky. (2008). “Racial differences in hospice revocation to pursue aggressive care.” Archives of Internal Medicine 168(2): 218–​224.

326    Kimberly S. Johnson and Ramona L. Rhodes Kagawa-​Singer, M., and L. J. Blackhall. (2001). “Negotiating cross-​cultural issues at the end of life: ‘You got to go where he lives.’” JAMA 286(23): 2993–​3001. Kapo, J., H. MacMoran, and D. Casarett. (2005). “ ‘Lost to follow-​up’: Ethnic disparities in continuity of hospice care at the end of life.” Journal of Palliative Medicine 8(3): 603–​608. Kelley, A. S., N. S. Wenger, and C. A. Sarkisian. (2010). “Opiniones: End-​of-​life care preferences and planning of older Latinos.” Journal of the American Geriatrics Society 58: 1109–​1116. Kiely, D. K., S. L. Mitchell, A. Marlow, K. M. Murphy, and J. N. Morris. (2001). “Racial and state differences in the designation of advance directives in nursing home residents.” Journal of the American Geriatrics Society 49: 1346–​1352. Krakauer, E. L., C. Crenner, and K. Fox. (2002). “Barriers to optimum end-​of-​life care for minority patients.” Journal of the American Geriatrics Society 50(1): 182–​190. Kwak, J., and W. E. Haley. (2005). “Current research findings on end-​of-​life decision making among racially or ethnically diverse groups.” The Gerontologist 45: 634–​641. Lauderdale, D. S., M. Wen, E. A. Jacobs, and N. R. Kandula. (2003). “Immigrant perceptions of discrimination in health care: The California Health Interview Survey 2003.” Medical Care 44(10): 914–​920. LaVeist, T. A., K. J. Nickerson, and J. V. Bowie. (2000). “Attitudes about racism, medical mistrust, and satisfaction with care among African American and white cardiac patients.” Medical Care Research and Review 57 (Suppl. 1): 146–​161. Lillie-​Blanton, M., M. Brodie, D. Rowland, D. Altman, and M. McIntosh. (2000). “Race, ethnicity, and the health care system:  Public perceptions and experiences.” Medical Care Research and Review 57 (Suppl. 1): 218–​235. Loggers, E. T., P. K. Macjejewski, E. Paulk, et al. (2009). “Racial differences in predictors of intensive end-​of-​life care in patients with advanced cancer.” Journal of Clinical Oncology 27(33): 5559–​5564. Ludke, R. L., and D. R. Smucker. (2007). “Racial differences in willingness to use hospice services.” Journal of Palliative Medicine 10(6): 1329–​1337. Lyke, J., and M. Colon. (2004). “Practical recommendations for ethnically and racially sensitive hospice services.” American Journal of Hospice & Palliative Care 21: 131–​133. Mack, J. W., M. E. Paulk, K. Viswanath, and H. G. Prigerson. (2010). “Racial disparities in the outcomes of communication on medical care received near death.” Archives of Internal Medicine 170(17): 1533–​1540. Mebane, E. W., R. F. Oman, L. T. Kroonen, and M. K. Goldstein. (1999). “The influence of physician race, age, and gender on physician attitudes toward advance care directives and preferences for end-​of-​life decision making.” Journal of the American Geriatrics Society 47(5): 579–​591. Medicare Payment Advisory Commission. (2013). “Report to the Congress:  Medicare payment policy.” Washington, DC: MedPAC. Accessed May 22, 2015, http://​www.medpac.gov/​ -​documents-​/​reports. Morrison, R. S., and D. E. Meier. (2004). “High rates of advance care planning in New York City’s elderly population.” Archives of Internal Medicine 164: 2421–​2426. Morrison, R. S., L. H. Zayas, M. Mulvihill, S. A. Baskin, and D. E. Meier. (1998). “Barriers to completion of health care proxies: An examination of ethnic differences.” Archives of Internal Medicine 158: 2493–​2497. Murphy, S. T., J. M. Palmer, S. Azen, G. Frank, V. Michel, and L. J. Blackhall. (1996). “Ethnicity and advance care directives.” Journal of Law, Medicine & Ethics 24: 108–​117.

Ethnicity as a Factor    327 National Center for Education Statistics. (2006). The health literacy of America’s adults: Results from the 2003 National Assessment of Adult Literacy. Washington, DC: US Department of Education. National Center for Health Statistics. (2012). Health People 2010 final review. Hyattsville, MD: National Center for Health Statistics. National Hospice and Palliative Care Organization. (2013). “NHPCO’s Facts and Figures: Hospice Care in America.” Alexandria, VA: National Hospice and Palliative Care Organization. Accessed April 22, 2014, http://​www.nhpco.org/​sites/​default/​files/​public/​ Statistics_​Research/​2013_​Facts_​Figures.pdf. National Research Council. (2003). Unequal treatment: Confronting racial and ethnic disparities in health care (with CD). Washington, DC: National Academies Press. Ngo-​Metzger, Q., D. H. Sorkin, R. S. Phillips, et al. (2007). “Providing high-​quality care for limited English proficient patients: The importance of language concordance and interpreter use.” Journal of General Internal Medicine 22(Suppl. 2): 324–​330. Ortman, J. M., and G. E. Guarneri. (2009). “United States population projections: 2000 to 2050.” Accessed April 22, 2014, www.census.gov/​population/​projections/​files/​analytical-​ document09.pdf. Park, N. S., I. V. Carrion, B. S. Lee, D. Dobbs, H. J. Shin, and M. A. Becker. (2012). “The role of race and ethnicity in predicting length of hospice care among older adults.” Journal of Palliative Medicine 15(2): 149–​153. Pew Hispanic Center. (2007a). “Changing faiths: Latinos and the transformation of American religion.” Accessed April 22, 2014, http://​pewhispanic.org/​reports/​report.php?ReportID=75. Pew Hispanic Center. (2007b). “English usage among Hispanics in the United States.” Accessed April 22, 2014, www.pewhispanic.org/​2007/​11/​29/​english-​usage-​among-​hispanics-​in-​the-​ united-​states/​. Pew Hispanic Center. (2011). “Statistical portrait of Hispanics in the United States, 2011.” Accessed April 22, 2014, www.pewhispanic.org/​files/​2013/​02/​Statistical-​Portrait-​of-​ Hispanics-​in-​the-​United-​States-​2011_​FINAL.pdf. Pham, K., J. D. Thornton, R. A. Engelberg, J. C. Jackson, and J. R. Curtis. (2008). “Alterations during medical interpretation of ICU family conferences that interfere with or enhance communication.” Chest 134(1): 109–​116. Phelps, A. C., P. K. Maciejewski, M. Nilsson et al. (2009). “Religious coping and use of intensive prolonging care near death in patients with advanced cancer.” JAMA 301(11): 1140–​1147. Phipps, E. J., G. True, and G. F. Murray. (2003). “Community perspectives on advance care planning:  Report from the Community Ethics Program.” Journal of Cultural Diversity 10: 118–​123. Phipps, E., G. True, D. Harris, et al. (2003). “Approaching the end of life: Attitudes, preferences, and behaviors of African-​American and white patients and their family caregivers.” Journal of Clinical Oncology 32(3): 549–​554. Reese, D. J., R. E. Ahern, S. Nair, J. D. O’Faire, and C. Warren. (1999). “Hospice access and use by African Americans: Addressing cultural and institutional barriers through participatory action research.” Social Work 44(6): 549–​559. Reese, D. J., E. Melton, and K. Ciaravino. (2003). “Programmatic barriers to providing culturally competent end-​of-​life care.” American Journal of Hospice & Palliative Care 21: 357–​364. Rhodes, R. L., K. Batchelor, S. C. Lee, and E. A. Halm. (2015). “Barriers to end-​of-​life care for African Americans from the providers’ perspective: Opportunity for intervention development.” American Journal of Hospice & Palliative Care 32(3): 137–​143.

328    Kimberly S. Johnson and Ramona L. Rhodes Rhodes, R. L., Teno, J. M., and S. R. Connor. (2007). “African American bereaved family members’ perceptions of the quality of hospice care: Lessened disparities, but opportunities to improve remain.” Journal of Pain and Symptom Management 34(5): 472–​479. Rhodes, R. L., J. M. Teno, and L. C. Welch. (2006). “Access to hospice for African Americans: Are they informed about the option of hospice?” Journal of Palliative Medicine 9(2): 268–​272. Rhodes, R. L., L. Xuan, and E. A. Halm. (2012). “African American bereaved family members’ perceptions of hospice quality: Do hospices with high proportions of African Americans do better?” Journal of Palliative Medicine 15(10): 1137–​1141. Rose, A., N. Peters, and J. A. Shea, and K. Armstrong. (2004). “Development and testing of the health care system distrust scale.” Annals of Internal Medicine 19(1): 57–​63. Rosenfeld, P., J. Dennis, S. Hanen, et al. (2007). “Are there racial differences in attitudes toward hospice care? A study of hospice-​eligible patients at the Visiting Nurse Service of New York.” American Journal of Hospice & Palliative Care 24(5): 408–​416. Sahgal, N., and G. Smith. (2009). “A religious portrait of African-​Americans.” Washington, DC:  Pew Research Center. Accessed April 22, 2014, www.pewforum.org/​2009/​01/​30/​a-​ religious-​portrait-​of-​african-​americans/​#overview. Scott, T. L., J. A. Gazmararian, M. V. Williams, and D. W. Baker. (2002). “Health literacy and preventive health care use among Medicare enrollees in a managed care organization.” Medical Care 40(5): 395–​404. Selsky, C., B. Kreling, G. Luta, et al. (2012). “Hospice knowledge and intentions among Latinos using safety-​net clinics.” Journal of Palliative Medicine 15(9): 984–​989. Smith, A. K., R. B. Davis, and E. L. Krakauer. (2007). “Differences in the quality of the patient-​ physician relationship among terminally ill African-​American and white patients: Impact on advance care planning and treatment preferences.” Journal of General Internal Medicine 22(11): 1579–​1582. Smith, A. K., R. L. Sudore, and E. J. Perez-​Stable. (2009). “Palliative care for Latino patients and their families: Whenever we prayed, she wept.” JAMA 301(10): 1047–​1057. Teno, J. M., B. R. Clarridge, V. Casey, et al. (2004). “Family perspectives on end of life care at the last place of care.” JAMA 291(1): 88–​93. Thornton, J. D., K. Pham, R. A. Engelberg, J. C. Jackson, and J. R. Curtis. (2009). “Families with limited English proficiency receive less information and support in interpreted intensive care unit family conferences.” Critical Care Medicine 37(1): 89–​95. Torke, A., N. S. Garas, W. Sexson, and M. T. Branch. (2005). “Medical care at the end of life: Views of African American patients in an urban hospital.” Journal of Palliative Medicine 8(3): 593–​602. US Census Bureau. (2012). “State and county quick facts 2012.” Accessed April 22, 2014, http://​ quickfacts.census.gov/​qfd/​index.html#. US Department of Health and Human Services, Administration on Aging Administration for Community Living. (2013a). “A profile of older Americans: 2012.” Accessed April 22, 2014, www.aoa.gov/​Aging_​Statistics/​Profile/​2012/​docs/​2012profile.pdf. US Department of Health and Human Services, Agency for Healthcare Research and Quality. (2013b). “National healthcare disparities report 2012.” Accessed April 22, 2014, www.ahrq. gov/​research/​findings/​nhqrdr/​nhdr12/​2012nhdr.pdf. US Department of Health and Human Services, Office of Minority Health. (2001). “National standards for culturally and linguistically appropriate services in healthcare.” Accessed April 23, 2014, http://​minorityhealth.hhs.gov/​templates/​browse.aspx?lvl=2&lvlID=15.

Ethnicity as a Factor    329 US Department of Health and Human Services, Office of Minority Health. (2014a). “Minority populations: African American profile.” Accessed April 23, 2014, http://​minorityhealth.hhs. gov/​templates/​browse.aspx?lvl=3&lvlid=23. US Department of Health and Human Services, Office of Minority Health. (2014b). “Minority populations: Hispanic/​Latino profile.” Accessed April 23, 2014, http://​minorityhealth.hhs. gov/​templates/​browse.aspx?lvl=3&lvlid=31. US Department of Health and Human Services, Office of Minority Health. “What is cultural competency?” (2014c). Accessed April 23, 2014, http://​minorityhealth.hhs.gov/​templates/​ browse.aspx?lvl=2&lvlid=11. Volandes, A. E., M. Ariza, E. D. Abbo, and M. Paasche-​Orlow. (2008a). “Overcoming educational barriers for advance care planning in Latinos with video images”. Journal of Palliative Medicine 11: 700–​706. Volandes, A. E., M. Paasche-​Orlow, M. R. Gillick, et al. (2008b). “Health literacy not race predicts end-​of-​life preferences.” Journal of Palliative Medicine 11(5): 754–​762. Waters, C. M. (2001). “Understanding and supporting African Americans’ perspectives of end-​ of-​life care planning and decision making.” Qualitative Health Research 11(3): 385–​398. Welch, L. C., T. M. Teno, and V. Mor. (2005). “End-​of-​life care in black and white: Race matters for medical care of dying patients and their families.” Journal of American Geriatric Society 53(7): 1145–​1153. Wicher, C. P., and M. A. Meeker. (2012). “What influences African American end-​of-​life preferences?” Journal of Health Care for the Poor and Underserved 23: 28–​58. Williams, M. V., D. W. Baker, R. M. Parker, and J. R. Nurss. (1998). “Relationship of functional health literacy to patient’ knowledge of their chronic disease. A study of patients with hypertension and diabetes.” Archives of Internal Medicine 158(2): 166–​172. Zaide, G. B., R. Pekmezaris, C. N. Nouryan, et al. (2012). “Ethnicity, race, and advance directives in an inpatient palliative care consultation service.” Palliative and Support Care 6: 1–​7.

Chapter 20

Reframing C a re i n End-​o f-​L i fe C a re  Helpful Themes from a Catholic-​Christian Understanding of Death Michael McCarthy and Mark Kuczewski

It has been noted by some scholars that Roman Catholic medical ethics provided many of the starting points for the development of contemporary mainstream clinical ethics, especially in regard to end-​of-​life care (Kelly, 1991). Concepts such as ordinary and extraordinary treatment, respect for the dignity of the person, and the principle of double effect provided fertile ground from which developed our national consensus on forgoing life-​sustaining treatment (Meisel, 1992, 2005). Of course, as the secular development of this framework has evolved, the emphasis on patient autonomy, patient choice, and patient control has grown and displaced some of the pillars on which it was built. This essay examines several key concepts in the Catholic-​Christian tradition that provide an important supplement or corrective to the current overemphasis on patient autonomy, an autonomy that almost exclusively centers on preference or choice. While a Christian sense of autonomy incorporates the importance of choice, the simple freedom to choose does not constitute the totality of autonomy. Rather this freedom is rooted in a sense of who one is and is becoming by the actions that he or she chooses. Thus a choice not to forego treatment or to continue with it cannot be made in isolation from the totality of who the person is. The care that this person is receiving at the end of life from a Catholic-​Christian perspective emphasizes an understanding of the human person as embodied and worthy of compassionate care within a community that acts both prudentially and lovingly. Patient wishes remain important, but their discretion is not unlimited. The dying process is a part of life to which all are subject. In short, the Christian understanding articulated here sees people not primarily as individuals but as persons in relation to each other, to nature, and to God. It entails responsibilities of compassion on caregivers and, because the patient is embodied, that is by its nature an experience shared with close others.

Reframing Care in End-of-Life Care    331 While this argument focuses on the Christian tradition from a Catholic perspective, one need not embrace specific religious doctrines or theological presuppositions to find this worldview helpful, particularly in the context of palliative care. While these perspectives are grounded in a rich theological background, hopefully all who find spiritual worldviews helpful appreciate these insights. And, of course, hospice and palliative care have tended to embrace such assumptions in their general approach to dying patients. Such care neither hastens nor impedes death but rather provides support for the entire family and social unit by attending to spiritual needs and managing symptoms in an effort to maximize the relational potential of the dying patient. Make no mistake about it: this kind of religious thought is normative. It posits that there are good and not-​so-​good ways to die. The good ways are in accordance with our natures as created beings who work with nature rather than seek to control and overcome it. Dying in accordance with our natures is not simply morally preferable; it is believed to be more conducive to one’s happiness.

Reorienting Autonomy In the development of the bioethical frameworks that govern end-​of-​life issues in the United States, particular importance has been given to individual autonomy. At the heart of these decisions and actions lies patient choice. The accent on patient choice, moreover, not only stems from the respect for autonomy principle elaborated in the Belmont Report and developed by Beauchamp and Childress but has also coincided with the growing commercialization of healthcare, particularly in the United States. The focus on patient autonomy has become so prominent that some physicians pass off difficult decisions to the patient, stating that it would be unethical for them to interfere with their medical choices (Moreno, 2007). This way of proceeding can leave the patient feeling isolated in these unfamiliar waters. Despite this seeming abdication of the support inherent in the doctor–​patient relationship, increasing patient autonomy continues to be a focus in the agenda to improve the patient’s end-​of-​life experience. This is particularly true in reference to the physician-​assisted suicide movement. Assisted suicide has become legal in several states and is a fairly well-​established practice in the state of Oregon. This represents a significant expansion of the role of patient autonomy. Autonomy had generally been conceived as legitimately expressed in negative rights (i.e., a right to refuse medical treatments), but assisted suicide establishes a role for autonomy to exercise a positive right (i.e., a right to choose a treatment explicitly to hasten death). This role extends autonomy beyond being a guarantor of freedom from burdensome treatment to establishing control over the time and manner of death. Autonomy is no longer simply about enabling the disease to “take its course.” Control becomes the driving value. However, some data suggests that this desire for control is less about a

332    Michael McCarthy and Mark Kuczewski particular vision of the end of life or a right guaranteed by freedom and more about a reassurance that the patient will not be allowed to suffer a death characterized by loss of autonomy, loss of control of bodily functions, and uncontrolled pain (Sullivan et al., 2000). Though the current bioethics discourse in the United States has made significant advances toward facilitating a good death by giving primacy to patient autonomy and asserting patient rights and choice, this framework has proven insufficient to meet its goal of being free from burdensome treatments. Routinely the technological imperative to use treatments simply because we can becomes the standard in the clinical setting. “The seriously ill, the aged, and their families seek out hospitals to stave off death but often find themselves trapped by a medical momentum that seems immune to their own sense of dignity and need for understanding and companionship” (Cahill, 2005, p. 71). When the technologically possible drives the clinical situation, the perspective on quality of life, relationality, and even patient choice can seem lost. The focus centers almost exclusively on treating the disease, while the needs of the person become a secondary concern. This often leaves one feeling that models of end-​of-​life care in our era is often split between two extremes, one in which agency seems to be lacking and technology is dominant and another in which individual autonomy is asserted as an intrinsic good so as to overcome the dominance not only of technology but of the experience and vicissitudes of the dying process itself. However, a Christian anthropological understanding from Catholic thought can provide a third way that sees a role for moral agency within a communal context that values an ideal of a natural death. This approach calls for care that extends beyond an autonomy-​driven care model and toward one that navigates between the paternalism of yesterday and the control-​driven freedom of today. A Catholic-​Christian model of care considers the understanding of the human person as an embodied being, that is, a unity of body and soul, deserving of compassion in the context of a caring community. The embodied patient at the center of the caring community is understood as more than a disease or a case of medical futility at the end of life but rather as a biologically, spiritually, and socially conditioned being. This multidimensional approach to care at the end of life is not unique to a Catholic-​ Christian perspective. Many medical schools and physicians ask and care for a patient biologically, psychologically, socially, and even spiritually (Borrell-​Carrió et al., 2004; Puchalski et al. 2014; Kuczewski et al., 2014). However, the motivations for delivering care in this manner come from a variety of sources. This section describes a Catholic-​ Christian understanding of care given at the end of life in a way that enables others to access and integrate this particular way of caring. While the description is rooted in a particular theological framework, it lends itself to discussions about what type of care is best and motivations underlying the care provided at the end of life. Reflections on care at the end of life begins with an understanding of the human person as embodied, argues why care ought to be given compassionately, and, finally, describes the communal dimension of care that incorporates the various relationships present at the end of life.

Reframing Care in End-of-Life Care    333

Embodiment as the Basis of Compassionate Care Within Community Embodied care counsels a reconciliation with the dying process, in which one’s mind is not simply placed over and against one’s body. It is precisely such a dualistic assumption that can justify virtually any attempt to assert control over the body and its processes such as dying. But if one is essentially embodied, then one’s fullest life entails acceptance of the processes that are intrinsic to embodiment. Such a foundation is potential support for planks of the hospice platform that aim to neither hasten nor impede death. Similarly, bodies are by their nature things in the world and part of communities. Minds can sometimes be characterized as individual and removed from and opaque to others, while characterizing people as embodied supports a vision of the dying process as something that will be important to and have an effect upon the community. A notion of embodiment from a Christian perspective lends itself to keeping in tension an understanding of the human person as body and soul, a tension not new to Christianity. Such a dualistic understanding can be traced back neo-​Platonism, a heavy influence of St. Augustine. Even today, a concern about dualism and end-​of-​life care can be seen in the debate surrounding a 2005 statement of John Paul II on artificial nutrition and hydration (John Paul II, 2004). The papal allocution addresses the language of “quality of life” and “vegetative state” to argue that nutrition and hydration form a “natural means of preserving life, not a medical act. Its use, furthermore, should be considered, in principle, ordinary and proportionate, and as such morally obligatory, insofar as and until it is seen to have attained its proper finality …” (John Paul II, 2004, Section 4). This is the statement that received the most attention by a variety of Catholic bioethicists asking if this statement represented a “change” in papal teaching. One of the more prominent theologians to speak on the matter was Dominican Father Kevin O’Rourke. O’Rourke focused on the way in which the Pope used the term “quality of life” and the implicit understanding of the human life. In this case, he was concerned that some would interpret that “life in PVS [persistent vegetative state] is an ‘intrinsic good’ and a ‘great benefit’ ” (O’Rourke, 2006, p. 90). O’Rourke and other, however, want to put forward the idea prolonging a life in PVS state is not a “great benefit,” and to speak of it as such is to speak of the body isolated from its relational or spiritual dimension. For a patient in PVS, the spirit, which in a Christian understanding is typically thought of dynamic and relational, is rendered mute. O’Rourke’s argument offers a reminder of the embodied nature of the human person that is both present physically in the human body but necessarily connected to the spirit, or soul, of the human person. By recognizing the reality of a deteriorating body, particularly in PVS, as an intrinsic good, regardless of context or consideration of “burdens and benefits” at the end of life, a disproportionate emphasis can be placed on the physical existence and the overuse of seemingly unlimited technology to delay the dying process (O’Rourke, 2005, 2006; Paris et al., 2006). As the body continues to age and deteriorate at the end of life, a physicalism poses a kind

334    Michael McCarthy and Mark Kuczewski of temptation to keep the body going as long as possible. The Christian Scriptures, likewise, emphasize a unified view as seen in the various stories of healings and cures, which aim at restoring wholeness for those whom Jesus cares for—​not just physical health. It is this holistic and embodied care articulated in Ethical and Religious Directives for Catholic Health Care Services (USCCB, 2001) that lies at the heart of Catholic-​Christian healthcare. The Directives articulate a vision of care for an embodied being as part of the mission and expectations that one ought to have on admission to a Catholic hospital. While the directives are not without debate with respect to particular directives, the overall perspective lends itself to an understanding of healing as beyond the physical (Aita et al., 2009; O’Rourke and Boyle, 1999). Part two of the directives points to the importance that “the care offered is not limited to the treatment of a disease or bodily ailment but embraces the physical, psychological, social, and spiritual dimensions of the human person” (USCCB, 2001, Part II). One of the contributors to the directives, James Keenan, notes that this embodied tradition has always been a part of the Christian and Catholic understanding of the human person and has been gradually separated by an underlying Greek philosophic understanding of the human person present in medicine (Keenan, 1996). The increasing efficiency of medical technology at keeping human beings alive has led some to refer to a “new dualism” in which the body is seen as raw material for choice and intervention, “exercised via the technical and instrumental rationality guiding much of modern science … Defining instrumental reason as the calculation of maximum efficiency in pursuing means to a given end …” (Lysaught, 2012, p. 412). The technological imperative and assisted suicide are seen from this point of view as the proverbial two sides of one coin. When one sees the body and the mind as two separate things, unbridled dominion over the body is easily justified. One might choose to preserve it at all costs or to end its being in order to prevent the ravages that technology cannot stave off. In either event, there is no response to the body that is naturally fitting whereby the mind may proceed according to its own preferences. This new dualism operates antithetically to the Christian practices in healthcare. A nondualistic view of the person aims to provide care that supports a person spiritually through pastoral care services such as offering of sacraments to patients whose spirituality is so mediated and listening to the needs of all patients, both physical and emotional (USCCB, 2001, Part II). When a patient arrives at the end of life, the embodied care of the person remains central to the overall treatment that one might receive. The Directives establish that the task of medicine must persist in caring measures, even if they prove not curative (USCCB, 2001, Part V). Care incorporating the use of medical technology should be used insofar as it does not prove useless or overly burdensome to the patient, offering some means of reasonable hope (USCCB, 2001, Part V). Yet often the burdens and rationality of treatment or foregoing treatment are difficult to discern, particularly at the end of life. Thus Catholic-​Christian healthcare must be approached in the spirit of compassion in order to assist the patient, the family, and those caring for patients and families with the challenges of caring for an embodied person at the end of life.

Reframing Care in End-of-Life Care    335

The Call to Compassionate Care A spiritual approach to end-​of-​life care entails compassion. Compassion means literally a “suffering with” or “enduring with,” according to its Latin origin. Again the religiously based perspective is not individualistic but emphasizes that, as embodied beings, each human being is affected by the suffering of others. Compassion calls for a response to such suffering in an effort to alleviate it. The theologian-​philosopher Thomas Aquinas describes compassion as a synonym for misericordia, an action in which one aims to remove an undue burden or stress that befalls another human being (Ryan, 2010). Aquinas is, of course, an Aristotelian who believes that human beings by nature seek happiness and happiness is predicated on living a certain kind of life. Aquinas’s notion of suffering contrasts with the happiness that human beings ultimately desire. Suffering frustrates human desires of attaining “the good life.” Aquinas follows Aristotle in characterizing happiness as resulting from virtuous activity but adds a Christian dimension in claiming one’s ultimate happiness is union with God. From this perspective, the dying process necessarily plays a part in achieving ultimate happiness. Part of the compassionate care indicative of the Christian tradition is the recognition of death as a passing unto God. An equally important aspect is the valuing of the relationships and contributions of human beings in this life. In the Christian tradition, achieving the ultimate happiness as union with God continues to take place, as it always has, through and with the people gathered with the individual—​friends, family, and/​or those providing care. Here it is important to note what the religious perspective adds to the Aristotelian. An Aristotelian perspective would characterize happiness in terms of excellence in certain activities. Activities are associated with capabilities, and happiness seems most commonly associated with one’s middle years. But the theological perspective of Aquinas would see similar value in all stages of life, as they are all part of the same journey toward God. Margaret Farley, in commenting on the aging process, draws attention to the union between the reality of deterioration as bodies begin to age and one’s spiritual dimension. She notes, “Both body and spirit are profoundly affected by aging; aging is both biological and spiritual. Whether experiences are of diminishment or enhancement (in wisdom and grace), they are tied to the body and spirit in the process of aging” (Farley, 2006, p. 124). Thus Farley and others call the attention to the process of aging and deteriorating as not something to be fought; rather there is value to each stage, and one cannot simply see life as “not worth living” and seek to end it. One must work with nature and allow nature to follow its course. But this also does not mean allowing pain to ride roughshod over a patient. The care provided ought to aim at alleviating the suffering that exists outside of one’s control, opening the possibility to most fully enter into that ultimate state of happiness. Care that is rendered takes on both a practical alleviation of the undue suffering but also highlights the transcendental nature of providing care, from the perspective of staff or family members who necessarily transcend themselves in providing care.

336    Michael McCarthy and Mark Kuczewski Compassion operates as a function of mercy, exists as an attribute of God, and is a virtue that can be practiced (Aquinas, ST II-​II 30.1). The Directives provide a helpful framework through which to explore this notion of compassion. Compassion incorporates Aquinas’s sense of mercy as compassio through “a listening presence; help in dealing with powerlessness, pain, and alienation; and assistance in recognizing and responding to God’s will with greater joy and peace” (Aquinas, ST II-​II, 14.2). Thus the compassionate care articulated in the Directives focuses on being present for patients in a way that acknowledges their physically undue suffering but also seeks to treat not just the symptoms but the embodied person with the teleological aim of happiness. This holistic goal, however, is not the sole responsibility of the pastoral care staff and the local pastors who provide the “spiritual” care but is extended to the health-​care team. The focus of the professional–​patient relationship, a relationship that ought to transcend the professional treatment of a patient’s physical discomforts, aims to treat the whole. For Christians this takes on a deeper significance at the end of life, given the tension of the relationship between the immanent end of the physical life and the transcendent end. “The health care professional has the knowledge and experience to pursue the goals of healing, the maintenance of health, and the compassionate care of the dying, taking into account the patient’s convictions and spiritual needs, and the moral responsibilities of all concerned” (Directives, Part III). The Directives signal that Catholic-​ Christian care should be rooted in compassion as the alleviation of physical pain and the furthering of happiness through symptom management, as well as provide space to honor the spiritual and human needs of the patient. This is particularly significant with respect to the patient’s human dignity and the human relationships that provide meaning at the end of life. These webs of relationships between patient, health-​care provider, and surrogate or family and friends involved in caring for the individual point to the communal nature of compassion.

The Communal Dimension of End-​of-​L ife Care Healthcare, Christian or not, necessitates a communal dimension in which members of the health-​care team work in concert with each, supporting the patient in providing quality and patient-​centered care. A Catholic-​Christian approach builds on this fundamental team approach by emphasizing the importance of socially responsible care of patients and the importance of the self-​care of the members for the health-​care model itself. First, from a Christian anthropological perspective, the social dimension places an emphasis on the human person as inherently social and also on health-​care providers as mutually responsible for all—​regardless of socioeconomic or political status. The Directives, moreover, provide a framework through which to view the social dimension of care.

Reframing Care in End-of-Life Care    337 In accord with its mission, Catholic health care should distinguish itself by service to and advocacy for those people whose social condition puts them at the margins of our society and makes them particularly vulnerable to discrimination: the poor; the uninsured and the underinsured; children and the unborn; single parents; the elderly; those with incurable diseases and chemical dependencies; racial minorities; immigrants and refugees. In particular, the person with mental or physical disabilities, regardless of the cause or severity, must be treated as a unique person of incomparable worth, with the same right to life and to adequate health care as all other persons. (Directives, Part I)

This quote from the directives points not only to the larger social conscious aspect of care but also to a mission-​oriented vision that shapes the health-​care community. A Catholic-​Christian care community approaches each patient, regardless of social status, as a person worthy of the best possible care as decided with input from the patient, the health-​care staff, and, if possible, the surrogate decision makers as necessary. This communal dimension reinforces a contrast with the individualistic and autonomy-​ driven approach to healthcare. While the decisions may ultimately rest with the patient, the process of arriving at any decision is a consultative process. Unlike the individual models of patient autonomy or physician paternalism, a consultative model allows for a model of care that draws from the freedom of the patient and the expertise of the health-​ care team. It is an ideal of shared decisionmaking as communal but patient centered. An important dimension that is underemphasized, though an essential aspect of a Catholic-​ Christian community, focuses on the self-​care of the health workers. A communal approach to Catholic-​Christian care depends on the supportive network within the health system or hospital capable of developing the importance of self-​care. The obstacles to providing patient-​centered care with a communal focus are pluriform. At any given time, an unforeseen medical event, different professional opinions on how care should be delivered, already tense family relationships strained by a loved one’s impending death, and several other components play into the delivery of care. Many players but particularly nurses are left to confront the challenging aspects of care at the end of life, which includes helping families and patients admit that they have entered the dying process. The constant focus on the patient and the various considerations at the end of life can weigh heavy on care providers. Burnout affects the way in which care is delivered. Thus part of the communal approach to Catholic-​Christian care is self-​care. In calling for a rethinking of Thomas Aquinas’s Cardinal Virtues, Keenan argues that self-​care ought to form an essential virtue if human beings are to maintain their self-​understanding as inherently relational beings (Keenan, 1995). Often this aspect of being relational in an inherently service/​other-​oriented profession can seem unimportant. Yet fundamental to this ability to care for others is the recognition of the need to care for oneself (Keenan, 1995). Without the recognition of the self in the process of providing care, the potential exists for the self-​ worth of the health-​care worker to become lost. At some level it is the recognition of oneself in the other that allows for one to approach the patient, regardless of social status, as a member of the same human community worthy of compassionate/​merciful care.

338    Michael McCarthy and Mark Kuczewski Aquinas’s understanding of mercy includes pity, and pity is inherently communal in that it necessitates the recognition of one’s self in the place of the sufferer. The notion of entering into the suffering of another is fundamental to the Christian story, in which God is made human in the person of Jesus and willingly enters into human existence, life and death. In Christian-​inspired healthcare, the health-​care provider’s ability to see him or herself in the patient and recognize that life as worthy of care is essential. Theologically, then, this patient–​physician relationship can mirror the Christian story in which God’s compassionate mercy allows for God to enter into the human story, which includes suffering. Thus God first recognizes humanity as worthy of love and enters into relationship with it in and through history through the gift of a fully embodied Jesus. Jesus’s entrance into human history most fully points to God’s presence in the midst of human social relationships from the perspective of faith. These human relationships point to and in some ways mirror one’s relationship with God. As a part of this faith relationship with God, a responsibility exists to recognize one’s social interactions with the human other—​the dying patient—​as a symbol of one’s faith relationship with God. The application of God, self, and other relationships to healthcare provides a unique lens through which to view the interactions between a health-​care team and the patient. In providing Catholic-​Christian care for the other, one functions as a witness to and minister of God’s in and through human relationships.

Conclusion While these three aspects of care—​ embodied, compassionate, and communal—​ have strong roots in the Catholic-​Christian theological tradition, the impact on how to approach the challenges at the end of life seem worth considering. Technology will continue to improve, and people will be physically able to be sustained for longer. The lingering question that becomes explicit in a theological context is: to what end? An autonomy-​driven model seems to reduce human existence to the ability to make decisions. At the end of life, why should this capacity to choose be held up as the ultimate value? When a life is viewed as embodied, decisions and questions of how best to proceed take into account not only the possibility of one’s physical well-​being but also the way in which one relates to the other—​both human and divine. The importance of the ultimate ends of human existence in framing medical decisions seems to be the point Kevin O’Rourke emphasizes in his response to John Paul II’s statement on artificial nutrition and hydration. O’Rourke saw the force of an embodied understanding of the human person and also the importance of compassion at the end of life. Compassion, as mercy, is not only the responsibility to be present to and care for those who suffer but to enter into their suffering. From a theological perspective, compassion enables one to continue to pursue “happiness” as including the capacity to live in relationship. This relationality of the patient emphasizes the final

Reframing Care in End-of-Life Care    339 dimension of care, the communal. A Catholic-​Christian model of care acknowledges that all human beings are a part of and shaped by the community. While the theological rationale for an embodied, compassionate, and communal focus in caring for dying persons at the end of life may not resonate with everyone, it is our hope that the importance of practicing this type of care touches not only the best aspects of patient care but also emphasizes values that lie at the heart of the human person at all stages of life.

References Aita, M., Bennett-​Woods, D., Clark, P.M., DuBois, J., Haddad, A., Kuczewski, M., Taylor, C., and Walter, J.J. 2009. “Consortium of Jesuit Bioethics Programs.” Commonwealth 136(3): 13–​15. Aquinas, Thomas. Summa Theologiae. http://​www.ccel.org/​ccel/​aquinas/​summa.html. Borrell-​Carrió, F., Suchman, A.L., and Epstein, R.M. 2004. “The Biopsychosocial Model 25 Years Later:  Principles, Practice, and Scientific Inquiry.” Annals of Family Medicine 2(6): 579–​581. Cahill, L.S. 2005. Theological Bioethics:  Participation, Justice, and Change (Washington, DC: Georgetown University Press). Farley, M.A. 2006. Just Love: A Framework for Christian Sexual Ethics (New York: Continuum International Publishing Group). John Paul II. 2004. “Address of John Paul II to the Participants at the International Congress on Life-​Sustaining Treatments and the Vegetative State:  Scientific Advances and Ethical Dilemmas” (Rome:  Vatican). http://​www.vatican.va/​holy_​father/​john_​paul_​ii/​speeches/​ 2004/​march/​documents/​hf_​jp-​ii_​spe_​20040320_​congress-​fiamc_​en.html. Keenan, J.F. 1995. “Proposing Cardinal Virtues.” Theological Studies 56(4): 709. Keenan, J.F. 1996. “Dualism in Medicine, Christian Theology, and the Aging.” Journal of Religion and Health 35(1): 33–​45. Kelly, D.F. 1991. Critical Care Ethics: Treatment Decisions in American Hospitals (Kansas City, MO: Sheed & Ward). Kuczewski, M.G., McCarthy, M. P., Michelfelder, A., Anderson, E.E., and Wasson, K. 2014. “‘I Will Never Let that be OK Again’: Student Reflections on Competent Spiritual Care for Dying Patients.” Academic Medicine 89(1): 54–​59. Lysaught, M.T. 2012. On Moral Medicine:  Theological Perspectives in Medical Ethics (Grand Rapids, MI: W. B. Eerdmans Publishing Co.) Meisel, A. 1992. “The Legal Consensus About Foregoing Life-​Sustaining Treatment: Its Status and Its Prospects.” Kennedy Institute of Ethics Journal 2(4): 309–​345. Meisel, A. 2005. “The Role of Litigation in End of Life Care: A Reappraisal.” The Hastings Center Report 35(6): 47–​51. Moreno, J.D. 2007. “The Triumph of Autonomy in Bioethics and Commercialism in American Healthcare.” Cambridge Quarterly of Healthcare Ethics 16(4): 415–​419. O’Rourke, K., and Boyle, P. 1999. Medical Ethics: Sources of Catholic Teachings (Washington, DC: Georgetown University Press). O’Rourke, K., and Boyle, P. 2005. “The Catholic Tradition on Forgoing Life Support.” The National Catholic Bioethics Quarterly 5(3): 537–​553.

340    Michael McCarthy and Mark Kuczewski O’Rourke, K., and Boyle, P. 2006. “Reflections on the Papal Allocution Concerning Care for Persistent Vegetative State Patients.” Christ Bioethics 12: 83–​97. Paris, J., Keenan, J.F., and Himes, K.R. 2006. “Quaestio Disputata: Did John Paul II’s Allocution on Life-​Sustaining Treatments Revise Tradition?” Theological Studies 67(1): 163–​168. Puchalski, C., Blatt, B., Kogan, M., and Butler, A. 2014. “Spirituality and Health:  The Development of a Field.” Academic Medicine 89(1): 10–​16. Ryan, T. 2010. “Aquinas on Compassion: Has He Something to Offer Today?” Irish Theol Q. 75(2): 157–​174. Sullivan, A.D., Hedberg, K., and Fleming, D.W. 2000. “Legalized Physician-​Assisted Suicide in Oregon-​the Second Year.” The New England Journal of Medicine 342(8): 598–​604. United States Conference of Catholic Bishops (USCCB). 2001. Ethical and Religious Directives for Catholic Health Care Services (Washington, DC: United States Conference of Catholic Bishops).

Section III

P H YSIC IA N -​ A S SI ST E D  DE AT H

Chapter 21

Physician-​A ssi st e d De at h in the Neth e rl a nd s Gerrit Kimsma

The Climate “Dutch euthanasia” as a social phenomenon of the 1960s and 1970s is inseparable from a profound change in publicly shared perceptions and morality based on a new discourse about many areas of life, one that has a focus on emancipation, authenticity, self-​ determination, autonomy, and individualism (Kimsma 2010; Youngner and Kimsma 2012). “Emancipation” became the keyword for workers, students, nurses, and patients striving for change in established, especially hierarchical, asymmetrical relations in the workplace, universities, and hospitals. Society was confronted with demands for more openness in many areas of life, meaning a breakdown of the silence on many taboo subjects, such as power, inequality, sexuality, abortion, and death. New allocations of power were sought to end authoritarian, repressive, and often arbitrary restrictions, allowing individuals to pursue their own goals and distance themselves from conceptions and beliefs that were considered “old fashioned.” This society-​ wide movement originated within the health care system. It first appeared in the area of medicine with a debate on truth-​telling, especially in cases of a poor prognosis, and was followed by calls for patients’ rights in general. The medical custom was to mitigate the details or hide the diagnosis of a malignant disease with few therapeutic options; instead, people wanted an open, in-​depth, bespreekbaar (discussable) dialog about their chances as their right (Kennedy 2012). Brochures and publications appeared, such as the outspoken example written by J. J. Buskes, a Protestant minister, whose book Truth and Lies at the Sickbed was published in 1964 and reprinted many times (Buskes 1975). Elisabeth Kübler-​Ross’s 1969 book On Death and Dying: What the Dying Have to Teach Doctors, Nurses, Clergy and Their Own Families was a major influence for change in the climate of Dutch medicine; it was

344   Gerrit Kimsma subsequently translated into Dutch with the title Lessons for the Living (Kübler-​Ross 1969). Justifications of “respect for patients as persons” were expanded with ideas about “death with dignity.” Even though Kübler-​Ross opposed euthanasia, her ideas about focusing on what patients desired were easily (in hindsight, somewhat too easily) integrated into the formulation of ideas about assisted dying, ideas which were based on conceptions of patient autonomy and conceptions of a “good death.” During the 1960s, the Dutch medical profession, especially the profession’s organization, the Koninklijke Nederlandse Maatschappij tot bevordering der Geneeskunst (KNMG; the Royal Society for the Advancement of Medicine), was debating the issue of euthanasia, both in writing and in the acknowledgment that it existed as a veiled practice. In addition to being part of the general tendency to become “open” about this practice, the organization wanted to protect those physicians who openly professed to have been involved in actively ending lives; officially, this practice was—​and in a way formally still is—​a crime. Although the KNMG tended to come out in favor of euthanasia in early publications, the profession as a whole remained divided. Information from a few surveys undertaken in 1967 about how physicians felt with respect to euthanasia are quite informative: 13% of those interviewed were somewhat positive; 6% were in full favor, if it were socially and legally accepted; and 42% would be in favor of euthanasia (including 26% of Roman Catholic physicians; Medisch Contact 1978). Clearly, a majority opposed allowing euthanasia. This division within the medical profession changed with surprising speed into an almost fully supportive stance. Before the 2002 law went into effect (which allowed only physician-​assisted dying (PAD), thereby creating, in effect, a medical monopoly), the medical profession incorporated the present practice to such a degree that some observers spoke about “self-​regulation” by the profession (Griffiths 2000). There was close cooperation between the legal and medical establishments. Broadly speaking, three successive developments may be distinguished: a phase of confrontation before 1984; acceptance since that year; and a phase of integration, beginning in 1998 and especially after 2002, when the law on euthanasia became effective. Cracks in the apparently harmonious solution to the combustible social issue of euthanasia appeared around 2010.

The Legal Development: From Confrontation to Acceptance It is self-​evident that legal developments are not processes that exist in a vacuum. Given the usual strict or superior position of the law standing above the particular interests of individuals, social parties, or pressure groups, it is interesting to note that a surprising sensitivity to changes in the social climate can be observed with respect to medical behavior that potentially shortens life (MBPSL). This is especially true for the landmark euthanasia case of 1973 (usually known as “the first case,” although it was not the first

Physician-Assisted Death in the Netherlands    345 historically; see Enthoven 1988). This case will be described in some detail because of the shifts in legal angles. The final verdict played a guiding role in what would follow, including the eventual structure and content of the “Euthanasia Law” of 2002. The accused was doctor Postma-​van Boven, a family physician, who ended the life of her paralyzed and incontinent mother in 1971 with a 200 mg morphium injection in response to repeated requests to be helped to die. Her actions were reported to the police by the director of the Catholic nursing home in which her mother was a patient. The director was familiar with the dilemmas of the patient and the family. During Dr. Postma’s trial, the focus of the Leeuwarden (lower) Court at first was on the criminal nature of the act, which was potentially punishable by a maximum jail term of up to twelve years. As witnesses gave testimony during the trial, the focus shifted toward the subject of how the medical profession in general proceeds in helping people who are suffering from an incurable disease to die a good death. The Court took special notice of the statements of a provincial medical inspector, a civil servant who stated that, for most physicians, it was no longer customary to prolong life until the bitter end. He mentioned five conditions in such situations: (1) an incurable disease without prospects for recovery; (2) physical or mental suffering that subjectively is unbearable for a patient; (3) the presence of a clear request, possibly in writing, to have this suffering come to an end; (4) the imminent arrival of the terminal phase; and (5) a decision by a physician, either the physician in charge or by another physician in consultation with the physician in charge. The Court accepted this focus on medical science and medical ethics instead of choosing to focus on the crime of taking a life (e.g., murder). It sentenced Dr. Postma to a suspended jail term of one week because she had not complied with the final “reasonable” procedure of slowly increasing the morphium and thereby acceptably shortening life; instead, she administered one lethal dose of 200 mg. Nationwide, this conviction was hailed as a victory for the proponents of euthanasia. The conditions stipulated by the Leeuwarden Court were recognized in subsequent cases and received a legal status of jurisprudence. However, some observations deserve mention. The Court surprisingly ignored the fact that Dr.  Postma was the daughter of the deceased and not her physician in charge. The Court also diverged from the fourth condition of the provincial inspector by adopting an original position, one stipulating that the terminal phase of a disease was not per se a reason to deny an active end of life to someone with unbearable suffering caused by a nonterminal disease. It is significant that the Court also clearly defined suffering as “subjective for a patient” as a guiding condition, meaning that suffering is defined by how a patient as a person, a subject, experiences his or her suffering. The Leeuwarden Court took a stance against active euthanasia after a patient’s request but focused on allowing passive euthanasia as the alleviation of suffering with an acceptance of shortening life as a consequence. The fact, however, was that Dr. Postma had actively euthanized her mother. The short suspended sentence of one week in jail expressed a change in the legal profession’s appreciation of the severity of this act. Finally, it is still quite surprising and somewhat enigmatic that the public prosecution decided

346   Gerrit Kimsma not to appeal the case even though there are quite a number of reasons that could and can be given for so doing. Undoubtedly, much deliberation must have taken place in the office of the prosecution, including the higher echelons, but that remains outside of public records. Dr. Postma, on the other hand, became a public figure and one of the founders of the remarkably popular Dutch Voluntary Euthanasia Society (NVVE). The NVVE currently has more than 160,000 members, is the largest such society in the world, and is a leading force in the international euthanasia movement.

Subsequent Cases The legal focus of the Leeuwarden Court subsequently was maintained in cases that reached the Dutch Hoge Raad (HR), the Netherlands’ Supreme Court. The chosen criteria for allowing euthanasia were maintained, refined, and extended in each subsequent case. There are four landmark cases. The first is the Schoonheim case of 1984, named after the defendant. The defendant was acquitted by the lower court of Alkmaar following an appeal of a guilty verdict that was handed down by the Amsterdam Court without punishment, resulting in an appeal to the HR. The HR appeal is the first time the court accepted the legality of a claim of force majeure, defined as the legal dilemma of a conflict of duties—​namely, the duty to preserve life versus the duty to alleviate suffering. The case involved the “unbearable suffering” of a ninety-​four-​year-​old woman. Her severe chronic ailments and further deterioration were already unbearable for this woman; the projected progression of symptoms threatened her wish for a death with dignity and left her without options to alleviate her suffering and make her situation comfortable. The final verdict became public several months after the HR session. Some claim that the Raad was waiting for publication of a substantial policy change by the medical profession’s KNMG organization (Pans 2006). The next case before the HR was the 1994 Chabot case, named after the defendant psychiatrist. This case concerned a fifty-​year-​old woman who had many traumatic experiences in life concerning her wedding, her divorce, and the dramatic loss of two adult sons, one by suicide and one due to a malignancy. She underwent psychiatric treatment for many years, made an unsuccessful suicide attempt, and requested help in dying from the NVVE. Chabot, a consultant of the NVVE, concluded after many hours of analysis that her death wish was not directly related to a psychiatric illness, even though her life experiences left her with a mental disturbance. Her objection to even more psychiatric treatments was an important factor. Chabot consulted seven colleagues and one ethicist in writing, all of whom agreed that, in light of her previous treatments and suicide attempts, this patient was not a good candidate for further treatment. The HR found Chabot guilty without punishment. Its main argument for a verdict was the absence of direct, personal contact between the consultants and the person who requested help in dying. There were, however, several additional clarifying positions in the verdict. First, the Raad stipulated that the cause of suffering is not the issue; the issue is its unbearable and hopeless nature. The Raad concluded that suffering from a psychiatric

Physician-Assisted Death in the Netherlands    347 disease can lead to a justified euthanasia, depending on the free and autonomous will of a psychiatric patient. Given the complexity of both the decision-​making capacity and prospects for therapeutic options, the decision stated that consultations should proceed with extreme caution and that a consultant is absolutely required to meet with the patient in person. The Raad stipulated also that the condition of hopeless suffering does not apply when a patient refuses a realistic alternative for the alleviation of that suffering. This verdict was greeted as opening “the doors for euthanasia” for psychiatric patients. The verdict also created the issue of the limits of suffering. By declaring that the cause is not relevant in the assessment, the decision suggested that only a patient could be the judge of unbearable suffering. If self-​determination is the decisive factor, it appears that the doors to acceptable euthanasia are left wide open. The answer to that question came with the next case before the HR. The Brongersma case of 2002, named after the individual whose life was ended by his GP Sutorius, concerned an eighty-​six-​year-​old well-​known former politician. He suffered from general ailments of old age and anxieties: physical limitations, loneliness, and further potential deteriorations in the absence of alleviating treatments or the possibility of improvement. Sutorius consulted with a colleague GP and a psychiatrist, both of whom agreed that the patient faced unbearable and hopeless suffering and was without treatment options. Given the limits of euthanasia practice at the time, his self-​reported case was brought to court. The Haarlem (lower) Court consulted with a psychiatrist and an ethicist, who both testified in favor of the defendant. The ethicist expressed the opinion, which was accepted by the Court, that there is no consensus in medicine on “unbearable suffering.” She argued in favor of a broad definition, one focusing on the acceptance of how an individual experiences suffering, irrespective of its cause, and acceptance of self-​determination as a justifiable criterion. In acquitting the defendant, the Haarlem Court agreed with the ethicist’s conclusion that there is no consensus on suffering in medical ethics and science, and there probably never will be. This verdict is directly opposed to the fundamental verdict of the HR that assessment of unbearable suffering should be based on the norms of medical ethics and medical science. The (theoretical) issue for the Amsterdam Court was the more general question of the absence or presence of consensus in cases of unbearable suffering without a medical classifiable disease, as opposed to the individual merits of the case. Based on witness statements from a professor in medicine and legal counsel for the medical profession (the KNMG), the Amsterdam Court concluded that issues of existential suffering and “suffering on life” (i.e., losing the appetite for living due to loss of abilities and quality of life) are outside “the medical domain” within which physicians have a legitimate interest and are professionally qualified. The Amsterdam Court found the defendant guilty without punishment. After appeal, the HR affirmed the verdict and reasoning of the Amsterdam Court. Even though the final verdict on the Brongersma case seemed to limit the legitimate area of justifiable euthanasia to cases “with a medical classifiable disease or affliction,” that was not the end of the discussion. There remain fundamental questions from a medical point of view, from public sentiments of dissatisfaction, and, last but not least, from issues of consistency within the jurisprudence itself. New social movements appeared that were

348   Gerrit Kimsma aimed at expanding the possibilities for legitimately ending life, and the KNMG responded to the Brongersma case with the 2003 inauguration of the so-​called Dijkhuis Committee, named for its chairman. The Committee’s assignment was to investigate options of acceptable euthanasia for cases of unbearable suffering that did not involve a clearly threatening medical disease or condition. The Committee agreed that assisted dying in cases described as “weary of life,” “having a completed life,” “suffering of life,” or “weary of life” could or should fall within a wider demarcation of the medical domain than the Brongersma case outlined. Even though the KNMG did not formally follow the report’s advice at the time, some of its proposals were followed in 2011. The KNMG now supports this option in the wake of decisions made by the Euthanasia Review Committees concerning acceptable euthanasia in cases of “suffering based on a number of old-​age deteriorations.”

The Medical Profession and Euthanasia: From Acceptance to Integration In the early 1960s and 1970s, the medical profession was divided on the issue of euthanasia. This period was also marked by a slow integration of ethics and law into the ethical debates of the medical profession. Ethicists and lawyers became privy to the medical ethical debates about terminating meaningless medical interventions and turned out to be protagonists of patients’ interests. These internal debates resulted in a rift within the monolithic KNMG and resulted in the secession of the Nederlands Artsen Verbond, the Dutch Physicians Union that included a body of physicians who opposed abortion, euthanasia, and physician-​assisted suicide (PAS). However, the liberal discussions within the KNMG during the 1960s and 1970s remained predominantly in favor of regulating euthanasia. A policy paper published in 1984 was a major step in outlining a prudent policy in favor of euthanasia. The climate favored a liberal position, jurisprudence was focused on existing medical practices in the various court cases, and there was an atmosphere of cooperation between the professions of medicine and law that was unopposed and supported by the institutions of the government in a situation where politics were still divided. The Board of the KNMG, without consulting its members, appeared to justify and regulate existing practices, protecting physicians and ending inequalities for patients. Conscientious objection to helping patients die was clearly recognized in this policy paper by its advice to refer requesting patients to a more “willing” colleague. It underlines a choice for PAS (i.e., actively ending life). The conditions developed in the jurisprudence are repeated, and the paper underscored the position that only physicians should be allowed to end the life of suffering, requesting patients. The paper included the procedural condition that consultations should be conducted by case-​independent physicians before the final act. Further steps were taken by the 1985 inauguration of the Commissie Aanvaardbaarheid Levensbeëindigend Handelen

Physician-Assisted Death in the Netherlands    349 (Committee on the Acceptability of Life-​Ending Treatment; CAL), a multidisciplinary committee. Over the course of twelve years, this committee published reports on the admissibility of euthanasia for both competent and incompetent patients (e.g., neonates with serious defects, comatose patients in vegetative stages, patients with progressive Alzheimer’s disease, patients with chronic psychiatric disorders). Although the ideas and conclusions of these reports did not carry the weight of law, they were clearly intended as professional outlines for future developments in euthanasia policies. Follow-​up policy papers were published in 1995, 2003 (immediately after the law on euthanasia came into force in 2002), and again in 2006. A Guideline on Palliative Sedation was published in 2009. The Guidelines on Euthanasia and Reduced Consciousness were formulated in 2010 after a request from the Department of Justice and are an example of the cooperation between the KNMG and the State. The Guidelines focused on patients who became unconscious after a request for euthanasia and after the professional consultation. A broad position paper on “The Role of the Physician in the Voluntary Termination of Life” was published in 2011 (KNMG 2011). This paper follows-​up on the consequences of the initial supporting position on PAD. It includes an outline of the physician’s role in cases involving previously unknown patients who choose to die, possibly through the cessation of eating and drinking, and also in cases in which patients decide to commit suicide by medication overdose. Physician assistance in cases involving psychiatric diseases or Alzheimer’s disease is outlined as an option within the existing Law on Euthanasia, and a clear line is drawn for patients who cite “a completed life” in their request for help in dying (Rurup, Onwuteaka-​Philpsen, Jansen-​van der Weide, and van der Wal 2005).

Politics and Euthanasia: Acceptance and Regulation Political decisions on euthanasia and its regulation by law trailed behind a well-​ established practice and took a long time to formulate. The practice has become more transparent since 1973 through the support of the medical profession and the courts. The political parties were divided on the issue, and the lines of dissension even ran within some parties. In general, the religion-​based Christian parties were opposed to euthanasia, liberal parties supported regulation, and the socialists remained divided. Many professionals even doubted whether it was necessary to regulate the practice by law. They contended that euthanasia might be considered as fitting within physicians’ responsibilities, based on the framework of “the medical exception” in existing laws. Several events that occurred between 1984 and 1986, all of which pointed to a need for law-​based procedures, ended this impasse. The KNMG came out in full favor of PAD. The HR reached a verdict in the Schoonheim case that was universally viewed as condoning this option. A liberal opposition party in the Dutch Parliament, called Democrats ‘66, introduced a bill allowing euthanasia and

350   Gerrit Kimsma assisted suicide; the Wessel-​Tuinstra bill was named for the member responsible for its introduction. Political divisions and technical legal reasons stopped the parliamentary passage of this bill. The government chose to wait for the 1985 publication of a report by a State Committee on Euthanasia, which had a euthanasia-​supporting majority conclusion. A minority report still reflected deep disagreement on the subject. The State Committee also coined “new” definitions: euthanasia was to be understood as ending life by a physician after a request; PAS was defined as helping a patient die by providing a prescription. The resulting political stalemate was solved in the Dutch fashion, in which governments are formed by coalitions depending on election results. The system works to prevent polarizations along the religious–​secular divide and attempts pacification by delegating further in-​depth studies to government committees. The effect of this procedure often leads to a certain “de-​politization,” which happened with the issues of abortion and euthanasia. In the case of euthanasia, the government inaugurated the Research Committee Medical Practice and Euthanasia in 1990; this committee is better known as the Remmelink Committee, named for its chairman. The Remmelink Committee focused on euthanasia and all medical decisions at the end of life: nontreatment decisions (NTDs), death in the course of alleviating pain and suffering (APS), voluntary active euthanasia and PAS, and life-​ending actions without explicit request (LAWER), a problematic category (Van der Maas, van Delden, and Pijnenborg 1992). This national empirical study and others, repeated in Van der Maas et al. (1995), Onwuteaka-​Philipsen et al. (2003), van der Heide et al. (2007), and Tweede Evaluatie (2012), ended much debate and reduced anxieties and speculation by showing the extent of the various interventions and underscoring the fact that the existing practices needed to be codified into law; later studies verified whether the law functioned well. The figures at first showed rather lower than expected levels of euthanasia and PAS, but there was a marked increase after 2010 (Onwuteaka-​Philipsen et al. 2012). The percentage of voluntary active euthanasia cases varies between 1.7% (1990) and 2.8% (2010) (Table 21.1). These figures show the involvement of a medical decision before death: actively ending life or assisting in suicide, accepting death in the course of alleviation of suffering, and not starting or discontinuing futile treatments. The sum of all the percentages shows an increase since the first study from 1990. The figures vary from around 40% in 1990 to more than 68% in 2010; it is important to note that the numbers for palliative sedation have been included since 2001. Reporting of euthanasia and PAS, which has always been legally required, increased from 18% to 80% over the years. This change reflects the lesser risk for physicians and a tendency to comply with professional standards and legal regulations well before, but even more so after the euthanasia law became effective in 2002. More than ten years of investigations indicate that the reasons for granting a request to end life varied little and include physical, psychic, and/​or social arguments: unbearable and hopeless suffering, prevention of further decay, meaningless suffering, pain, death without dignity, dependence, being weary of life, prevention of suffocation, no longer wishing to be a burden to the family, and prevention of pain (van der Wal and van der Maas 1996). In the beginning, the medical reasons were almost all based on cancer and other serious and often terminal diseases; in recent years, there has been a shift to

Physician-Assisted Death in the Netherlands    351 Table 21.1 Percentage of Medical Decisions at the End of Life Deaths in thousands

128th

135th

140th

138th

136th

% of all deaths 1990

1995

2001

2005

2010

Euthanasia

1.7

2.4

2.6

1.7

2.8

PAS

0.2

0.2

0.2

0.1

0.1

LAWER

0.8

0.7

0.7

0.4

0.2

APS

19

19

21

25

36

NTD

18

20

20

16

18

Palliative sedation Reporting

5.6 18

41

54

7.1 80

11.1 77

th: deaths in the thousands PAS: deaths in the course of physician-​assisted suicide LAWER: deaths after life-​ending without explicit request APS: deaths after alleviating pain and suffering NTD: deaths after nontreatment decisions

include low numbers of dementia patients (4%), psychiatric patients (2%), and those with no serious physical or psychiatric disorders (3%). In general, there are roughly three times more requests than granted life-​ending interventions. The reasons for refusal have been a separate subject for research for their own sake, but also because of public anxieties about inequalities in access to the euthanasia option (Norwood, Kimsma, and Battin 2009; Pasman, Rurup, Willems, and Onwuteaka-​Philipsen 2009).

The Termination of Life on Request and Assisted Suicide (Review Procedures) Act and the Subject of Control: Consultants and Euthanasia Review Committees The euthanasia law was activated on April 1, 2002. Contrary to popular perceptions, euthanasia and assisted suicide remain listed as crimes in the Penal Code and are punishable by, respectively, up to twelve and three years imprisonment. This law is mainly a

352   Gerrit Kimsma codification of the previous jurisprudence and professional regulations. The law defines the conditions for nonprosecution for physicians only, the so-​called requirements of due care. In essence, this is a doctor-​centered law. The conditions in Article 2 Part 1 stipulate that an attending physician: a. holds the conviction that the request by the patient was voluntary and well-​considered; b. holds the conviction that the patient’s suffering was lasting and unbearable; c. has informed the patient about the situation he was in and about the prospects; d. and the patient together hold the conviction that there was no other reasonable solution for the situation of the patient; e. has consulted at least one other, independent physician, who has visited with the patient and has given an opinion in writing with respect to the requirements of due care, referred to in parts a through d; f. has terminated a life or assisted in a suicide with due care. These conditions were considered unproblematic because they reflected the court decisions and verdicts before 2001. Article 2, Part 2, referring to incompetent patients, is new and more problematic. It states that: If a patient aged sixteen years or older is no longer capable of expressing his will, but prior to reaching this condition was deemed to have a reasonable understanding of his interests and has made a written statement containing a request for termination of life, the physician may carry out this request. The requirements of due care, referred to in the first paragraph, apply mutatis mutandis.

This was clearly a politically motivated addition that was especially aimed at making euthanasia possible for Alzheimer’s patients; it is also clearly without legal precedent. Its correct interpretation is still a matter of debate. The next two parts of the law concerning minors are, in effect, an adaptation of the provisions in the euthanasia law to the law dealing with patients’ rights in general and the rights of minors in particular. Article 2, Part 3, concerning minors between sixteen and eighteen years of age, allows the ending of life only after parents or guardians have been involved in the decision process. Article 2, Part 4, concerning minors between twelve and sixteen years of age, allows assistance with dying after parents or guardians agree with the request to terminate life or assist in suicide. Again, the conditions of Article 2, Part 1 apply mutatis mutandis (Vrakking et al. 2005).

Checks (SCEN) and Evaluations (ERCs) The law incorporates two existing elements in the area of quality control: checks and evaluations, one before the intervention to end a life and one afterward. In 1984, the KNMG

Physician-Assisted Death in the Netherlands    353 proposed having independent consultations take place before ending life. This proposal reflects the serious and irreversible nature of such an act and also serves to protect physicians from potentially troubling decisions and emotional transference in situations of terminal care. This plan was realized in 1997, when the training of independent euthanasia consultants was introduced in cooperation with the Amsterdamse Huisartsen Vereniging (the Amsterdam General Physicians Society). With the continued financial support of the government, this program of Support Consultation Euthanasia in the Netherlands (SCEN) has been expanded to cover all of the Netherlands and now includes more than 600 consultants. These consultants are available for information, support, and consultations seven days a week; their independence is assured by the fact that they can be reached only through a central phone number. They visit the patient by the request of a physician who intends to end the life of that patient. If necessary, the visit is made the same day as the request, either at home, in a hospital, or at any other facility, including a hospice. The consultant writes a report that is included with the papers that a physician files after reporting the act to a local medical coroner, who, in the Netherlands is a civil servant. This organization is run by the KNMG and has become very effective and highly appreciated. Even though the euthanasia law does not require that a SCEN physician be consulted, the majority of the consultations are performed by SCEN physicians. The Regionale Toetsings Commissies Euthanasie (Euthanasia Review Committees; ERCs) were the second element incorporated into the euthanasia law. The ERCs were established in 1998 to assure the realization of the goals of an euthanasia practice: transparency, carefulness, a non–​crime-​focused approach. Before 1998, the few reports that were filed resulted in investigations by police and prosecutors; a final decision about prosecution was made at the level of the Justice Department by the Secretary himself. After 1998, it was hoped that transparency would be achieved by increased reporting and an individual assessment of each case. A non–​criminal law approach would hopefully be effected by the interdisciplinary nature of the committees (i.e., “putting the law at a distance”). There are five committees that cover the country’s legal regions. Each committee includes a lawyer-​ chair, a physician, and an ethicist, as well as substitutes for each discipline. Reporting has increased since the inauguration of the ERCs, from 18% in 1990 to 77% in 2010. The annual ERC reports make it clear that PAD is mainly performed by general practitioners (GPs): 3,777, or 95% of the 3,965 cases in 2012, were reported by GPs. Eighty percent of the cases concerned patients with an end-​stage malignant disease. Other causes included cardiovascular disease (4%), diseases of the nervous system (6%), diseases of the lungs (4%), Alzheimer’s disease (1%), psychiatric diseases (0.3%; fourteen cases), other diseases (3.6%), and combinations of diseases (4.3%). The functioning of the ERCs was the subject of a 2010/​2011 national survey. More information was requested about all items in 22% of the cases. Generally speaking, the five regional ERCs have fairly uniform assessment policies. There were some differences in compliance with the requirement for a “careful ending of life.” Committees showed differences in the degree of perseverance exercised to find out details about prescribed pharmaceuticals. Committee members are generally satisfied with the system of evaluation (Tweede Evaluatie 2012, 199, 210).

354   Gerrit Kimsma Research has also been focused on the follow-​up of cases “deemed not-​careful.” These cases are reported to both the public prosecution and the Medical Inspectorate, a government institution. There were thirty-​six such cases between 2007 and 2011; most concerned the items involved in the consultation and the final intervention to end life. None of the thirty-​six cases led to criminal proceedings; all were delegated to the Medical Inspectorate. The main issues of concern were the independence of the consultation and the careful medical procedure used to end life. These issues played a minor role from a criminal legal perspective as compared to the other “requirements of a careful practice,” such as the request, level of suffering, and the presence of alternatives. In some cases, the Medical Inspectorate initiated a follow-​up by communicating with the reporting physician to inform him or her about correct procedures and to prevent recurrences of violations.

Special Groups and Issues Five issues should be mentioned in connection with PAD:  palliative care in the Netherlands (because of its connection with terminal sedation), active ending of life in neonatal care, active ending of life in mentally handicapped incompetent persons, euthanasia and coma patients, and suicide through cessation of eating and drinking.

Palliative Care/​Terminal Sedation Palliative care had a rather late onset in the Netherlands and developed amid international claims that the practice of euthanasia existed because of a poorly developed palliative care system. Since 1996, the Dutch government and the European Community have provided substantial funding for institutions, palliative care teams, research, journals, and the inauguration of university departments. In 2011, the European Association for Palliative Care concluded that “palliative care is well developed in countries with legalized euthanasia/​assisted suicide, or at least no less developed as in other European countries (Palliative Care Development 2011).” Euthanasia is no alternative for palliative care, but it is an option if all palliative care measures have been exhausted. The frequency of terminal sedation (palliative sedation) has been researched in national investigations since 2001 and shows an increase from 5.6% (2001) to 11.1% (2011) of all deaths. The statistics do not indicate a decrease in euthanasia or assisted suicide. There is increased experience with terminal palliative sedation. Research from 2010 indicates that terminal sedation was chosen in 30% of the cases in answer to a request for an active end of life. These figures need clarification. In principle, there is a sharp distinction between the indications for euthanasia or PAS and palliative sedation. Palliative sedation is started when and if there are refractory symptoms that cannot be alleviated in any other way. Sedation with the aim to guide toward death after stopping all intravenous fluids and

Physician-Assisted Death in the Netherlands    355 nutrition is the standard of care if and when life expectancy is less than fourteen days, approximately the time it would take someone in poor condition to die a natural death without any food and water. Palliative sedation is considered “normal medicine,” as opposed to euthanasia and/​or PAS. One reason to choose sedation is that a life expectancy of fourteen days or less falls within the guidelines. Another explanation is that it is a way to avoid the more complex medical and procedural steps (such as consultation or reporting) involved in choosing euthanasia and PAS. Ongoing, fluid, complex situations make a choice for either option more difficult.

Euthanasia and Neonatal Care, Palliative Sedation Procedures of PAD in the Netherlands are based on requests from competent patients over sixteen years of age or from patients who once were legally competent. Competence is not relevant in neonatal care because the parents are the legal guardians. Debates in pediatrics about quality of life and options to end life came somewhat later than the initial debates about adult patients and concerned severely ill neonatal babies with no chance for survival and in dire suffering. In 1992, the Dutch Society of Pediatrics published a report called Doen of Laten (Acting or Letting Go). It advocated the liberal approach of allowing some newborns to die, but lacked legal protections for the physicians. In the aftermath of criminal charges being filed against two pediatricians, Verhagen and Sauer from Groningen University developed the Groningen Protocol. The Groningen Protocol described the steps to be followed in allowing an active end to life in this vulnerable population (Verhagen and Sauer 2005). Despite international criticism, the Dutch Pediatric Society accepted the protocol in 2004, and procedures for case evaluations were established in 2007. Each case has to be reported to a Central Committee on Late Terminations of Pregnancy and Life-​Ending of Neonates, called the Committee Hubben after its chairman. No cases were reported in 2011; in 2012, there were only three cases of late abortions after twenty-​four weeks’ gestation. One explanation for the lack of cases is that ultrasound can be used after eighteen weeks of pregnancy, thereby allowing for late abortions in cases where severe deformities are detected. Another explanation, besides a fear of prosecution, is that physicians and lawyers disagree about whether the use of muscle-​relaxing drugs, such as those used in palliative sedation, after stopping artificial respiration constitutes “normal medicine” and, as such, falls within the limits of accepted procedures. If this is not the case, the procedure should be reported to and evaluated by the Committee Hubben. So far, this “stalemate” has not been resolved because no cases have been reported.

Euthanasia and Severely Handicapped Suffering Patients The Dutch Society of Physicians for the Mentally Handicapped issued a report in 2013 on the subject of “Dealing with Questions to End Life of Incompetent People with a Mental

356   Gerrit Kimsma Deficiency.” The report concludes that actively ending a life would be the only possible option left to end serious and hopeless suffering in rare and exceptional situations—​ particularly since terminal sedation is available for a patient with refractory symptoms and a life expectancy of less than fourteen days. The Report admits that if a physician nevertheless ends a life, he or she may be morally justified but will face legal charges and be prosecuted (Advisory Group NVAVG 2013).

Euthanasia and Coma Patients: Patients in a Vegetative State/​Unresponsive Wakefulness Syndrome (UWS) Coma patients suffer from loss of consciousness due to brain damage. It is a generally established medical opinion that coma patients do not suffer because there is no conscious mind to perceive that suffering. Since there is usually no request, it is legally not possible to end the life of a coma patient. After a prognosis without any chance for recovery has been established, the accepted medical procedure is a nontreatment policy—​ starting with nontreatment of life-​threatening diseases, followed by stopping artificial delivery of fluids and nutrition. The issue of coma patients in relation to euthanasia gained a new dimension after several cases were reported to the ERCs. The cases concerned patients who lost consciousness after a consultation and a procedure to end their lives. Some physicians continued with the intervention, based on the request and their commitment to end life. The central issue was whether patients were suffering, and the ERCs needed a guideline based on a professional consensus. Losing consciousness right before dying is “normal”; this loss may be caused by a disease, treatments, or both. The question then becomes whether to continue with the procedure that was intended to end life. The KNMG stipulated in its guideline “Euthanasia with a Lowered Consciousness” that continuation should depend on the presence of suffering, to be determined using the Glasgow Coma Scale (KNMG-​ Richtlijn 2010). The regulations prescribe the assessment of eye movement (E), muscular movement (M), and verbal expressions (V). A score between 3 and 6 leads to the conclusion that suffering is absent because a patient is not awake and not aware. Scores of 7 through 12 indicate that suffering cannot definitely be excluded because a patient may not be awake but may sometimes be partially aware. Even if a patient has requested help in dying, life will not be ended if the score is between 3 and 6. A physician shall not be prosecuted for ending the life of a patient who has requested help in dying and has a score of 7–​12.

Stopping Eating and Drinking (SED) Dying through the voluntary cessation of eating and drinking has been a recognized and discussed option for many years in the United States. Chabot estimated its prevalence in the Netherlands for the first time in 2007 (Chabot 2007). Two items justify a

Physician-Assisted Death in the Netherlands    357 discussion within the context of PAD. First, roughly 50% of the persons who resorted to this option had requested help in dying but were turned down by their physicians for various reasons. Second, the medical profession has adopted the rule that physicians have a role in helping and supporting patients who have chosen this path. The KNMG and the nursing profession have prepared a detailed list of care items and pharmaceutical support to achieve a comfortable death with dignity (KNMG and V&VN 2014). There is disagreement as to the exact numbers of patients who choose this option (2,800 according to Chabot, or 2.1% of all deaths vs. 600 according to van der Heide, or 0.4%) depending on whether proxies or physicians are the source of information. It is clear, however, that individuals with and without threatening, potentially terminal diseases make this choice. Most are older than sixty years (many older than seventy-​five), and one-​third still have partners. Many suffer from diseases such as cancer, cardiovascular ailments, psychiatric problems, and debilitating disorders associated with aging; some do not suffer from a serious somatic or psychiatric disease. The reasons given are manifold: somatic (weakness, pain, shortness of breath), increased dependence (being a burden to others, being handicapped, loss of dignity in general), demoralization (no goal in life, loneliness, depression/​somberness), and to exercise control over dying (having a completed life, preventing loss of control). Chabot recently published a comprehensive guide on “Taking Control of Your Death by Stopping Eating and Drinking” (Chabot 2014).

A Changing Landscape After 2010:  New Tensions and (New) Players For the first few years after the 2002 law was passed, the practice of euthanasia seemed to develop without much ado. Although the number of reported cases did vary a little, it remained relatively stable and low. The ERCs, which were inaugurated by the state in 1998 and integrated into law in 2002 with a more definitive status of having the final say in reviewing, refined their approach to acceptable procedures and slight deviations in some cases. They submitted yearly reports to the government. As stated earlier, extensive government-​subsidized evaluations of the functioning of the law were undertaken in both 2005 and 2010. The conclusions from 2005 are mainly positive. The numbers of euthanasia and PASs were lower than projected, the physicians’ organizations were satisfied, and the goals of the law seemed to have been met (Evaluatie 2007). The conclusions from the 2010 evaluations are almost identical, but with some more or less important differences, depending on how one looks at the practice. One conclusion is that there is no slippery slope. However, there is an undeniable increase in the percentages of calculated euthanasia and PASs in the national investigations. There were 2,325 euthanasia cases (1.7% of all deaths) in 1990 and 3,850 (2.8% of all deaths) in 2010. The increase is significant and has no clear explanation because the types of underlying diseases are generally the same. This increase is even more obvious if one looks at the absolute numbers of ERC-​reported cases: 3,136 cases in 2010; 3,695 cases in 2011; 3,995 cases in 2012,

358   Gerrit Kimsma 4829 cases in 2013 and 5306 cases in 2014. This is an increase of approximately 10 to 16% each year in the number of euthanasia/​PAS cases. Over the past few years, frictions and polarizations have appeared in the relatively quiet landscape of PAD that existed before 2010. The participants in these changes are, respectively, the NVVE, a new public initiative called Uit Vrije Wil (translated literally as “based on one’s free will”), and, last but not least, the government’s ERCs, with their ongoing shifts or nuances in justifications of “acceptable euthanasia” in the areas of psychiatric diseases, Alzheimer’s patients, and persons without serious diseases. The response of the KNMG, the fourth participant, is both accommodating to the “new challenges” but also an attempt to set limits, especially in the area of assisted dying for Alzheimer’s patients. The overall harmony among the important participants in shaping the Dutch practice of PAD has come to an end. The ERCs adopt practices that the KNMG finds unacceptable, there are growing frictions between the liberal NVVE and the now seemingly conservative KNMG, and the agreement between the state and the KNMG is jeopardized by the ERCs’ very liberal interpretation of the “interpretive space” of the euthanasia law.

The NVVE The NVVE is unhappy with a practice in which physicians have a dominant position, especially since there is a large gap between those patients who want and those who finally receive an active ending of life. Over the years, there have been three times more requests than actual reported cases, indicating that a high number of physicians refuse requests for different reasons. The NVVE was and became even more of a pressure group after the euthanasia law was enacted, aiming to promote more liberal approaches to ending life for groups that seemed to be outside the limits of the law. Its view on the euthanasia law is that physicians interpret its limits far too narrowly. In addition to the “incorrectly refused” patients, the NVVE supports patients with chronic psychiatric diseases, patients with Alzheimer’s disease, and a group of elderly people who have no serious diseases but feel that they have “a completed life.” In order to service these groups of potential euthanasia subjects, the NVVE inaugurated the so-​called Life Ending Clinic (LEC) early in 2012. It was part of the NVVE at first but later became a legally separate and independent affiliate. The LEC operates using mobile physician–​nurse teams. Although the apparent intent is to realize euthanasia within the limits of the law, the LEC’s basic conviction is that the existing medical professional procedures are too narrow in their interpretations of the law. The teams mainly visit requesting persons at their homes, develop a relationship, assess the conditions of the law, ask for a professional SCEN consultation, and then end life according to accepted careful procedures. So far, there have been 1,094 applicants between March 2012 and August 2013, a number that increases at a rate of sixty persons a month. Between March 2012 and August 2013, 101 lives were ended. All cases are reported to the ERCs, and so far none has been assessed as having been handled “without due care.”

Physician-Assisted Death in the Netherlands    359 The diseases of the applicants differ significantly in comparison with the cases reported to the ERCs. Almost 60% have a somatic disease, almost 40% have psychiatric diseases, almost 10% have “a completed life,” and 8% suffer from Alzheimer’s disease (the total of more than 100% reflects the presence of comorbidities in one person). The LEC seems to be “responsible” for an increase in the category of psychiatric patients in 2013. The LEC is a new phenomenon in the area of ending life, and its functioning is met with apprehension, criticism, and also relief. Its appearance allowed the NVVE to realize several aims consistent with its policy of broadening the availability of PAD, but with quite a difference in appreciation by the general public and the medical profession. The LEC is a functional answer for the group of “incorrectly refused patients” and corrects a certain “wrong.” However, the LEC also intends to open up the option of PAD for “refused patients,” possibly including those with a chronic psychiatric disease and Alzheimer’s disease. Taking a closer look leads to some observations. The Clinic is one further step in a process of the “functionalization” of PAD that signals a change in the doctor–​patient relationship. The relationship is moving from an attending physician involved in a long process of care and palliation to a physician whose only goal is ending life (i.e., a “death-​ doctor”), clearly a new phenomenon. A further sign of this process of the LEC’s increasing social acceptance is its attempt to receive funding from health insurance companies for thirty additional teams, thus underlining a process of the “normalization” of euthanasia that, from the start, was recognized as “not normal medicine.” As part of its overall policy to make euthanasia more widely available, the NVVE organized public support in an attempt to repeal Article 294 in the Penal Code concerning the crime of assisted suicide. The intent is to change the law to decriminalize this assistance, as is the policy in several European countries. The NVVE used the so-​called Heringa case, a recent court case in which an elderly son assisted in the suicide of his ninety-​nine-​year-​old stepmother, to gain momentum for this drive.

Uit Vrije Wil The second and “new player” in the euthanasia field is the Uit Vrije Wil group. It is an example of a public drive to expand the options of PAD to a self-​chosen death that is parallel to the options of the euthanasia law. This is a public initiative led by well-​known Dutch citizens; it is supported by more than 140,000 signatures and has gained the right to a hearing in the Dutch Parliament. The group introduced a proposed law that would allow euthanasia practically outside the scope of physicians for anyone older than seventy years of age. According to the proposal, trained lay personnel, called “death councilors,” will take the place of physicians. The assessment of the criterion of unbearable suffering is intended to be determined solely by the requesting person, without an outsider’s assessment. Mandatory reporting is proposed. A review by a committee, similar to that required by the current euthanasia law, would be mandatory but the physician would be

360   Gerrit Kimsma replaced by a death councilor. Physicians would be involved in writing out prescriptions because they are the only ones who have a legal mandate to do so; they would, however, be writing prescriptions without any knowledge about the case details. It is probably unnecessary to note that the KNMG did not support this proposal. No further steps were taken after the proposal was heard in a Parliament Committee. The group formally ended at the end of 2012 and delegated its actions to the NVVE. The public drive made it obvious that a fairly large percentage of the Dutch public would like more options for ending life than are available under the current law.

Euthanasia Review Committees The ERCs have been in existence since 1998. For the first few years, practically all the cases concerned somatic diseases, mainly terminal cancer or chronic heart or neurological diseases with no hope for recovery. Recent years, however, have shown an increase in cases with psychiatric patients, Alzheimer’s disease patients, and patients who are “weary of life.” The last category includes elderly people who no longer wish to continue living: they have lost the sense of being meaningful and are semantically described as “having a completed life,” “being tired of living,” or, as the Dutch refer to it, “being ready with life.” The central question is whether these newly assessed cases are within the limits of the law or constitute a stretching of the law beyond its intended limits. One needs to realize that the ERCs only respond to cases that are reported. They seem to be lenient in their evaluations, but they respond to what physicians apparently seem to consider justified euthanasia (i.e., opinions that have been seconded by independent medical consultants). There are some explanations for this development. One is an intended vagueness in the wording of the euthanasia law. Conditions such as unbearable suffering are defined in general terms. The law is intentionally “open,” sensitive, or susceptible to changing views in society, as the government stated in its response to the first evaluation of the euthanasia law (Kamerstukken). Another reason is the intended allowance of euthanasia or PAS as options for psychiatric patients, even though the emphasis has always been on an extremely careful approach, with long periods of discussions, evaluations, and reserve. Early cases were described in detail in the ERC yearly reports, and no cases have been found to have been treated without “due care.” The general approach to psychiatric patients who request to die used to be a refusal, reflecting a professional consensus that suicidal ideas are proof of a need for treatment and that these patients are, by definition, not competent. Psychiatrists in general professed to feeling extremely uncomfortable with a request to be helped to die. Two issues are cardinal:  competency and suffering without prospects for improvement. Despite the presence of a psychiatric disease, competency became procedurally accepted in many patients with rational underpinnings. A refusal to accept that there were no more options for treatment was gradually replaced by a more lenient approach if patients cooperated in following some generally applied available treatments.

Physician-Assisted Death in the Netherlands    361 Until 2010, there were roughly two cases involving psychiatric patients each year. In 2011, there were thirteen cases, in 2012 there were fourteen, and in 2013, forty-​two cases. This change in numbers reflects a change in approach and has led to a public and medical-​professional debate. Because many psychiatric patients were deemed competent, the general conclusion was that psychiatrists were too reserved in their approach. The influence of psychiatric cases on the LEC undoubtedly contributes to the increase in numbers. The Dutch Society of Psychiatry has published a guideline for dealing with psychiatric patients who request to end life. There is a similar increase with respect to help in dying for Alzheimer’s patients: twelve cases in 2009, twenty-​five in 2010, forty-​nine in 2011, forty-​two in 2012 and fourty-​one in 2014. The moral acceptance of the issue hinges on the determination of competency, suffering, and communication. The issue largely concerns cases of early-​onset Alzheimer’s patients who had adequate competency and full awareness at the time but expected deterioration of the mind and personality. Anxiety over future suffering is considered a real part of the condition of “unbearable suffering.” The euthanasia law does not require a written request for help with dying from competent patients; however, there is an important difference with Alzheimer’s disease patients or, in general, when patients have lost competence. The law requires a written request or at least such convincing documentation of persistent requests in the patient files that any and all uncertainty about its substance is removed (Rurup, Onwuteaka-​Philipsen, van der Heide, van der Wal, and van der Maas 2005). The ERC assessments of past years show a shift in patients’ necessary levels of abilities and forms of communication. The ERC accepts cases as being handled “with due care” when patients in the final stages were only able to communicate on a rudimentary basis, either with body language or symbolically. In one case that was portrayed in detail in the media, the patient was barely able to show that she wanted to die. In fact, she did not realize what the physician who ended her life intended. The ERC approved the case on the basis of the medical information, a concurring report from a second independent consultant, and interviews with both the physician and the consultant. The case was played out in the media with additional information. Proponents and opponents differ in stressing the contractual nature of a written request versus the ability to communicate and understand the nature of the request, even with limited competence and communication. The issue of acceptable PAD for Alzheimer’s disease patients has created a conflict between the ERCs and the KNMG. A 2012 case caused an irreconcilable confrontation. The case concerned a patient who did not realize that the physician had come to end her life and was drugged into unconsciousness before the euthanasia was performed. This confrontation between the KNMG and the ERCs also created a conflict between the KNMG and the government, which is responsible for the functioning of the ERCs. The KNMG’s position is based on the conclusion that the active ending of the life of an Alzheimer’s disease patient does not fall within the limits of the euthanasia law if there is no competence and communication. It should be possible to review a requesting patient’s ability to communicate in order for the attending physician and a consultant to verify the conditions of the euthanasia law. This conflict still exists in 2016.

362   Gerrit Kimsma Both the public debate and the ERCs are now including a different group of people. The public debate focused on a group of mainly elderly people who have “lost the appetite for living”; these individuals see no sense in continued living, and life no longer has any meaning for them—​they are “weary of life.” In the ERC’s evaluations, this group appeared in increasing numbers with “ailments of old age,” all of which added up to a conclusion that unbearable suffering was indeed an appropriate assessment. The public debate focused on self-​determination in a choice to die; the ERCs’ focus became the ailments of old age, which was evident in symptoms, loss of functions, and loss of independence. In reality, both reasons for requesting an end to life apply to the same population. According to the ERC case descriptions associated with this category, the suffering of the patients was related in a major or minor extent to “medical classifiable somatic or psychiatric disease or affliction.” The case descriptions testify to the existence of a much more complex reality, which is for some a “gray area.” There are hardly any old people asking for an end to their lives who do not have a medical condition associated with old age that is partly or completely responsible for a low quality of life and feelings of isolation and meaninglessness. Some cases are characterized by a “piling up” of several conditions, each of which may be less serious but, when considered together, constitute a situation that is experienced as unbearable. The ERCs have accepted these cases and the KNMG has tended to side with the ERCs at this point because it was agreed that all conditions of the law had been fulfilled.

Reflections and Conclusion There is a general agreement that the goals of euthanasia practice and law have been realized: societal control of a life-​ending practice performed after a patient’s request and transparency and legal security for physicians. There is, however, some discontent with the law because its requirement that only physicians can terminate life is perceived as being too narrow. This discontent is also more fundamental in that the law itself is perceived as not being flexible enough. If and when physicians have applied the limits of the law “correctly,” many, especially the elderly, perceive its limits as being too narrow. Many feel that a person should have the option to end life if he or she experiences “unbearable suffering” in the absence of medical ailments. Having a “completed life,” “experiencing life as meaningless,” and “wishing to die a dignified death” are fundamental reasons that many people believe should justify ending lives, with or without the involvement of physicians. It is unexpected and surprising that the law-​supported “self-​regulation” of the medical profession is criticized at three different levels by different groups. The KNMG restricts the options of PAD to situations in which there is a “medical classifiable disease or ailment” or even “a medical basis,” and it opposes the NVVE’s drive to expand options. The KNMG follows the euthanasia law in accepting a written request in cases of incompetence, but restricts the option of PAD for Alzheimer’s disease patients to those

Physician-Assisted Death in the Netherlands    363 who are capable of (some form of) communication right up to the very intervention to end life; this policy differs from the ERC’s decisions and the NVVE’s demands. Last, a hitherto unexpected finding has arisen secondary to the actions of the LEC. This is related to the fact that there are three times as many requests for help as there are final interventions to end life. This difference was previously viewed as proof of the medical profession’s careful approach. Outsiders, such as the NVVE, however, have concluded that physicians have incorrectly refused to help patients die. A “hidden” psychological factor also seems to play a role in the acceptance or rejection of a request to help a patient to die—​that is, a certain mental saturation or exhaustion of physicians that is inherent in the often long process of euthanasia (van Marwijk, Haverkate, van Royen, and The AM 2007). Some physicians tend to welcome the appearance of the LEC in order to avoid the taxing confrontations inherent in requests for help in dying, especially in the “gray areas” presented by those who claim to have “a completed life,” psychiatric patients, and patients with Alzheimer’s disease (personal communication). Clearly, the subject of PAD has not been exhausted.

References Advisory Group NVAVG. August 2013. Omgaan met vragen om levenbeëindiging bij wilsonbekwame mensen met een verstandelijke beperking:  een handreiking [Managing questions regarding ending-​of-​life of incompetent people with a mental handicap:  An inventory for support]. Buskes, J. J. 1975. Waarheid en Leugen aan het Ziekbed [Truth and lies at the sickbed]. Baarn: Ten Have. Chabot, B. 2007. Auto-​Euthanasie. Verborgen stervenswegen in gesprek met naasten [Auto-​ euthanasia: Hidden dying trajectories in conversation with proxies]. Amsterdam: Bert Bakker. Chabot, B. 2014. Taking control of our death by stopping eating and drinking. Amsterdam. www. dyingathome.nl. Enthoven, L. 1988. Het Recht op leven en Dood [The right to life and death]. Deventer: Kluwer. Evaluatie. May 2007. Wet toetsing levensbeëindiging op verzoek en hulp bij zelfdodin [Evaluation: Termination of Life on Request and Assisted Suicide (Review Procedures) Act]. ZonMw, Den Haag: Summary: 13–​23. Griffiths, J. 2000. Self-​regulation by the Dutch medical profession of medical behavior that potentially shortens life. In Regulating morality. A comparison of the role of the state in mastering the mores in the Netherlands and the United States, edited by H. Krabbendam and H. -​M. ten Napel. Antwerpen-​Apeldoorn: Maklu, 173. Kamerstukken II (Parliamentary Annals), 31700 XVI, nr. 114, p. 15. Kennedy, J. C. 2012. The lateness of the Dutch Euthanasia debate and its consequences. In Physician-​assisted death in perspective. Assessing the Dutch experience, edited by S. J. Youngner and G. K. Kimsma. Cambridge: Cambridge University Press, 3–​21. Kimsma, G. K. 2010. “Death by request in the Netherlands: Facts, the legal context and effects on physicians, patients and families.” Medicine, Health Care and Philosophy 13: 355–​361. KNMG. 2011. The role of the physician in the voluntary termination of life. Position paper. Utrecht. www.knmg.nl/​voluntary-​termination-​of-​life.

364   Gerrit Kimsma KNMG-​Richtlijn. April 2010. Euthanasie bij een verlaagd bewustzijn [KNMG-​Guideline: Euthanasia and lowered consciousness]. Utrecht. https://​www.google.com/​search?sourceid= navclient&ie=UTF-​8&q=KNMG-​Richtlijn.+April+2010&gws_​rd=ssl KNMG and V&VN. 2014. Zorg voor mensen die bewust afzien van eten en drinken om het levenseinde te bespoedigen [Caring for people who consciously refuse food and drinking to hasten the end-​of-​life]. Position paper. https://​www.google.com/​search?sourceid=navclient&ie=UTF-​ 8&q=KNMG+and+V%26VN+2014&gws_​rd=ssl Kübler-​Ross, E. 1969. On death and dying. New York: Macmillan. Medisch Contact 1978. 33: 749. Norwood, F., G. Kimsma, and M. P. Battin. 2009. “Vulnerability and the ‘slippery slope’ at the end-​of-​life: A qualitative study of euthanasia, general practice and home death in the Netherlands.” Family Practice. doi:10.1093/​fampra/​cmp065. Onwuteaka-​Philipsen, B. D., A. Brink-​Stoppelenburg, C. Penning, G. W.  F. de Jong-​Krul, J. J. M. van Delden, and A. van der Heide. September 2012. “Trends in end-​of-​life practices before and after the enactment of the euthanasia law in the Netherlands from 1990 to 2010: A repeated cross-​sectional survey.” Lancet 380(9845): 908–​915. Onwuteaka-​Philipsen, B. D., A. van der Heide, D. Koper, et al. August 2, 2003. “Euthanasia and other end-​ of-​ life decisions in the Netherlands in 1990, 1995 and 2001.” Lancet 363(381): 395–​399. Palliative Care Development in Countries with a Euthanasia Law. October 2011. Report for the Commission on Assisted Dying. Briefing papers. European Association for Palliative Care. Pans, E. 2006. De normatieve grondslagen van het Nederlands euthanasierecht [The normative foundation of the Dutch euthanasia law]. Dissertation, Vrije Universiteit, 24. Pasman, H. R. W., M. L. Rurup, D. L. Willems, and B. D. Onwuteaka-​Philipsen. 2009. “Concepts of unbearable suffering in context of ungranted requests for euthanasia: Qualitative interviews with patients and physicians.” British Medical Journal 339: b4362. Rurup, M. L., B. D. Onwuteaka-​Philipsen, M. C. Jansen-​van der Weide, and G. van der Wal. October 2005. “When being ‘tired of living’ plays an important role in a request for euthanasia or physician-​assisted suicide: Patient characteristics and the physician’s decision.” Health Policy 74(2): 157–​166. Rurup, M. L., B. D. Onwuteaka-​Philipsen, A. van der Heide, G. van der Wal, and P. J. van der Maas. 2005. “Physicians’ experiences with demented patients with advance directives in the Netherlands.” Journal of the American Geriatrics Society 53: 1138–​1144. Tweede Evaluatie. 2012. Wet toetsing levensbeëindiging op verzoek en hulp bij zelfdoding [Second evaluation Law Termination of Life on Request and Assisted Suicide Act]. ZonMW, Den Haag. www.zonmw.nl/​publicaties. van der Heide, A., B. D. Onwuteaka-​Philipsen, M. L. Rurup, et al. May 10, 2007. “End-​of-​life practices in the Netherlands under the Euthanasia Act.” New England Journal of Medicine 356(19): 1957–​1965. Van der Maas, P. J., J. J. M. van Delden, and L. Pijnenborg. 1992. Euthanasia and other medical decisions concerning the end of life. Health Policy Monographs. Amsterdam: Elsevier. Van der Maas, P., G. van der Wal, I. Haverkate, et al. 1995 “Euthanasia, physician-​assisted suicide and other medical practices involving the end of life in the Netherlands, 1990–​1995.” New England Journal of Medicine 335(22): 1699–​1705. van der Wal, G., and P. J. van der Maas. 1996. Euthanasie en andere medische beslissingnen rond het levenseinde [Euthanasia and other medical decisions at the end-​of-​life]. Den Haag: Sdu uitgevers, 57.

Physician-Assisted Death in the Netherlands    365 van Marwijk, H., I. Haverkate, P. van Royen, and The AM. 2007. “The impact of euthanasia on primary care physicians in the Netherlands.” Palliative Medicine 21: 609–​614. Verhagen, E., and P. J. Sauer. 2005. “The Groningen Protocol: Euthanasia in severely ill newborns.” New England Journal of Medicine 352: 959–​962. Vrakking, A. M., A. van der Heide, W. F. Arts, et al. September 2005. “Medical end-​of-​life decisions for children in the Netherlands.” Archives of Pediatric and Adolescent Medicine 159(9): 802–​809. Youngner, S. J., and G. K. Kimsma (eds.). 2012. Physician-​assisted death in perspective. Assessing the Dutch experience. Cambridge: Cambridge University Press.

Chapter 22

The Case Ag a i nst Physician-​As si st e d Suic ide and E u t ha nasia Ira Byock

Doctors must not kill patients. This statement is neither an assertion nor a conclusion. It is a statement of principle and an appropriate starting point from which to discuss physician-​assisted suicide and euthanasia. The question of whether doctors should be legally allowed to hasten a patient’s death evokes passionate feelings among those who answer yes and those who say no. I argue that for many people these disagreements are rooted in alternative assumptions about the fundamental roles, obligations, and responsibilities, as well as the limitations, of the medical profession within society. Those disagreements will not be resolved here. Instead, by delineating divergent suppositions, beliefs, and differential weighting of values that underlie passionate opinions on these issues, I hope to foster mutual understanding, tolerance, and respect among people with strongly held opposing views. Although my task in this chapter is to present a case against legal physician-​ assisted suicide and euthanasia (PASE), it is not my intention to change readers’ minds. Rather, my intention is to present a worldview and ethical framework built on core social values and principles within which doctors can responsibly serve society and each individual patient, alleviating suffering without intentionally causing patients’ deaths. In dialogue, people who will never agree about physician-​assisted suicide may nevertheless recognize values held in common and, at times, find opportunities to collaborate on initiatives that advance common goals.

The Case Against Physician-Assisted Suicide and Euthanasia    367

Fundamental Assumptions About the Nature of Society and Role of Professions If civilization can be thought of as a physical edifice, its moral values and ethical principles represent the foundation and weight-​bearing walls of the social superstructure. Moral values define what is right and wrong. A society’s values and principles protect and support individual and communal life by providing broad guidelines, responsibilities, and boundaries for human behavior. Protection and preservation of life is a core value of civilization and undergirds fundamental functions of any society. It is the organizing principle that is the rationale for expending common resources for sanitation, vaccinations, building codes, food inspection, road design and repair, hazardous material units, and search and rescue teams. Correspondingly, a strict prohibition against killing human beings pervades laws, regulations, and guidelines of virtually all societies. Specific exemptions exist only in carefully delineated circumstances. Killing in self-​defense—​to protect one’s own life—​is recognized as a legal defense against the charge of murder. Societies give statutory authority to police to kill aggressors in protection of the lives and safety of others. Some societies (a diminishing few in the contemporary world) reserve the right to capital punishment, executing convicted criminals ostensibly to deter future violent crimes. Soldiers are accorded authority to use lethal force against enemy armies and individual combatants. Even here, in principle the military profession exists for the protection of the lives of countrymen. The professions of ministry, medicine, law, and education arose in antiquity as loci of specialized knowledge and expertise in service, first to individuals and second to society as a whole. Within the social architecture, the professions perform essential services and are accorded special authority and privilege. To become a professional, an individual must acquire specialized knowledge and expertise and either swear an oath or otherwise formally promise to act in a manner that benefits each client, patient, student, or congregant. Additionally, each profession imposes strict prohibitions on its members’ actions, creating boundaries around the profession’s authority and privilege in protection of vulnerable clients, patients, students, or congregants. The profession of medicine developed to bring specialized knowledge and skills in response to injury and illness. The fundamental responsibilities of physicians are to cure when possible, alleviate suffering, and promote healing and well-​being (Bernat 2012; Bulger and McGoveren 2001). The medical profession has also traditionally advised society, including governments, on matters related to prevention of illness and injury and promotion of health. Consistent with the nature of professions within society, for the protection of vulnerable people, the profession of medicine imposes certain specific limitations on the behavior of its physician members. For instance, doctors are prohibited from engaging

368   Ira Byock in sexual relations with patients they are treating. Because of the heightened power and authority physicians acquire within society and the potential for patients to be unduly influenced, manipulated, or even coerced by the physician, the proscription against having sex with patients applies even when the relations are consensual (Bulger and McGoveren 2001; Miles 2004). Doctors are prohibited from participating in executions of condemned prisoners or assisting in forcible interrogation or torture, even in jurisdictions in which doing so may be legal (Miles 2006; Miles, Alencar and Crock 2010). The proscription that physicians must not kill patients is among the core values and founding principles of the profession rooted in the earliest writings of medicine (Kass 2002). It is, by extension, fundamental to the architecture of civilized society. Across centuries, countries, and cultures, the medical profession has asserted sovereignty in matters related to the proper roles and boundaries of its members and has defended the prohibition against killing against contemporary exigencies and pressures from civil authorities. Thus most medical societies maintain the proscription against physician-​ assisted suicide even where it has been made legal by democratic processes or court decisions (Bernat 2012). Similarly, the profession continues to prohibit doctors from participating in torture or forcible interrogations, even when ordered to do so by police or military authorities (Miles 2006; Miles, Alencar, and Crock 2010). Doctors are prohibited from carrying out capital punishment, even when the death sentence results from proper judicial procedures and is entirely legal (Gawande 2006; Gawande, Denno, Truog, and Waisel 2008; Sikora, and Fleischman 1999). In the latter part of the twentieth century, as the discipline of hospice and palliative medicine was being developed to best serve seriously ill and dying people and their families, the profession extended this basic protection of vulnerable people within the founding principles of the new specialty (Cassel and Foley 1999; Byock 2009).

Stipulating to Good Intentions and Points of Agreement This issue frequently evokes passionate responses among many people who support and those who oppose legalizing PASE. It is possible for people of good faith to come to divergent conclusions and disagree vehemently about this issue without impugning the motives or characters of one another. Those who support and those who oppose PASE are hardly monolithic. They share sufficient commonality of attitudes and beliefs on these issues to make discussion of their differences worthwhile. However, important diversity of opinion exists within these broad opinion camps, and any representation of the stance of one side or the other is a generalization. I am keenly aware that my efforts to identify commonly held assumptions of those in favor of and those opposed to PASE might misrepresent the thoughts of some people on both sides.

The Case Against Physician-Assisted Suicide and Euthanasia    369 Opposing sides come to this issue with important agreements. In general, people who feel strongly—​one way or the other—​about PASE tend to believe that care for dying people is unacceptably deficient in important ways and that, as a result, many people suffer to needless extents as they die. Both sides agree that current systems of clinical and social support are commonly poorly coordinated, focused on disease rather than people, and not infrequently disserve those they are intended to help. Most people engaged in these issues believe that doctors may not adequately treat physical discomfort and frequently do not skillfully communicate and counsel people who they cannot cure. Agreement also exists on key ethical points: Autonomous persons have a right to refuse medical treatment, including treatments required to sustain life, such as mechanical ventilation, renal dialysis, antibiotics, and medically administered nutrition and hydration. Those who oppose PASE draw a critical distinction between allowing a person to die peacefully and acting with the intention of ending a person’s life. While acknowledging differences of opinion on this distinction, there is near-​universal agreement that palliative sedation is ethically acceptable when suffering persists due to physical pain, dyspnea, or agitation in a dying person despite best available treatments (Quill and Byock 2000; Rousseau 2000).

The Nature of Disagreements Over Physician-​Assisted Suicide The degree of vehemence evoked by disagreements over this issue is itself instructive. In panel discussions or debates over PASE, opponents often infuriate one another. The concept of PASE is abhorrent to opponents. Correspondingly, opponents can appear insular and obstinate in seemingly failing to substantively respond to carefully reasoned arguments in support of PASE by proponents. In fact, principle-​based opposition greatly narrows the potential for well-​reasoned arguments and empiricism to change minds or identify middle ground. Such seeming intransigence is not close-​mindedness but a considered affirmation of core values and basic social architecture. Within the worldview presented, the principle that physicians must not kill patients is embedded in the social fabric; it can be explained but need not be defended. From the moral and social perspective of those opposed to PASE for principle-​based reasons, the act of prescribing or injecting a medication with lethal intention is not medical care. The definition of the verb care is “to look after, to provide for” (Webster’s New Universal Unabridged Dictionary 1979). Care for a person who is dying extends into an open-​ended future in which death is acknowledged as inevitable but not intended. It could be argued that in some instances a physician’s decision to end an individual’s life represents an abrogation of the responsibility to provide care. The dictum that physicians must not kill patients does not diminish a principle-​based obligation to care, including a responsibility to respond to suffering. However, there is a meaningful distinction to be made between alleviating suffering and ending the life of the sufferer.

370   Ira Byock It is not obstinacy that causes principle-​based opponents to dismiss empiric data from Oregon, Washington, the Netherlands, or Belgium that purport to show either a benefit or a lack of harm from the practices of PASE (Battin et al. 2007; Smith et al. 2011; Loggers et al. 2013; Tucker 2012). Instead, when correctly understood and applied, principle-​based opposition to medically hastened deaths renders such data irrelevant to the question of the acceptability of the practices. To understand the nature of this disagreement, it may help to consider hypothetical violations of unrelated core values or professional boundaries. No amount of reasoning or empirical data purporting to show lack of harm could overturn society’s proscriptions against slavery, child pornography, and forcible sex with children, or the involuntary harvesting of organs from living donors. This seems obvious; such basic prohibitions can be explained but need not be defended. Indeed, these analogies might seem shocking or inappropriate to proponents of PASE. If so, their reaction is instructive in conveying the corresponding extent of abhorrence that is evoked in many opponents of PASE by the idea of physicians intentionally killing patients. To further understand the limitation of empiricism in this application, consider the value of honesty in physician–​patient communications and the prohibition against physicians having sexual relations with patients. If a series of hypothetical clinical studies and public surveys showed that most people do not mind doctors misleading them, would the findings make lying to patients acceptable? If another series of hypothetical studies found that a majority of patients who had sex with their doctors did not feel coerced or harmed, would that make sexual encounters with patients acceptable? For reasons of principle, consistent with the social architecture described, the answers to both questions are no. Studies from jurisdictions in which hastened death is legal have value for other reasons. It is important to track the frequency of these acts and to understand reasons and circumstances in which they occur. The findings may enhance understanding of prevailing public attitudes and advance our understanding of suffering associated with advanced illness and dying, thereby highlighting important foci for clinical and health service research.

Tension Between Principles and Pragmatism In addition to differences in basic assumptions and understanding about the nature of the medical profession in relation to society, tension between principles and pragmatism underlie disagreements that form the debate over physician-​assisted suicide. Ethicists might characterize this as tension between principle-​based and utilitarian ethics. People who support legal PASE may agree wholly or in part with my description of social and professional architecture yet believe that at times the needs of society, and the suffering of individual members of society, properly outweigh this historic prohibition.

The Case Against Physician-Assisted Suicide and Euthanasia    371 In the view of many proponents, the knowledge and expertise held by the medical profession and its members primarily serves a utilitarian function. Medical services can be applied when indicated in response to informed requests and consent by individuals with capacity to make decisions. The choice to hasten one’s own death rests with the individual. The ethical principle of autonomy must not be abridged by constraints on physician practice. Since this physician-​assisted suicide would be restricted to people who are formally determined to be dying, the prohibition against ending life is weakened. If a dying person is of sound mind and wants to die, he or she has the right to seek help from a willing physician. Guidelines and regulations must ensure that people making requests are fully informed, not made in haste, and not coerced. Within this perspective, proponents of PASE see the current systemic problems and deficiencies as adding to the urgency for extending the practice of hastening death to sick people. Looking forward, the extended human lifespan and subsequent swelling of older, sometimes frail adults and the burgeoning numbers of people living with serious chronic illnesses will strain our health and social systems in unprecedented ways and extents. The deficiencies of care and social support and the suffering people face are likely to worsen. It is only a matter of time before limitations of our resources force us to face the fact that we cannot meet all people’s needs. If we cannot otherwise prevent or alleviate people’s suffering, offering the choice of a hastened death seems only humane. Principle-​based opponents of PASE see the situation differently. Any society’s core values and principles are predictably tested during times of social upheaval. The very purpose of principles is to provide guidance on wise and proper actions that can withstand intense contemporary pressures. This function of principles pertains individually and collectively: It is easier to be kind when there is no reason to be angry, easier to be faithful when there is no temptation, easier to be honest when the truth will cause no distress, and easier to be fair when there is little at stake. Rather than being irrelevant or out of touch with current realities, the importance of principles rises during times of war, economic depression, and political and social strife. Indeed, the prohibition against physicians taking the lives of patients has endured through two millennia of wars, plagues, depressions, and natural disasters. The Geneva Conventions, which outlined protections and basic treatments of prisoners of war, injured or sick soldiers, and noncombatants, the establishment of war crimes, and subsequent trials, such as the Nuremberg Trials, collectively comprise evidence that even in the worst of times, some principles of human behavior remain fundamental. That includes some things that physicians are specifically forbidden from doing because of their special knowledge and skills. Today’s economic and demographic challenges are unprecedented but are, thankfully, less severe than many other periods in human history. Although there is far less economic and social disruption now than during the Great Depression or World Wars I and II, within the United States and throughout much of the world, these are not times of peace and prosperity. This chapter is being written as America’s war in Afghanistan—​ the longest war in the nation’s history—​has yet to end, and while the nation fights ongoing, smoldering wars against terror around the globe, during a slow recovery from an

372   Ira Byock historically deep recession with persistently high unemployment, and at a time when health-​care costs absorb 18% of the gross domestic product. This social context impacts individuals and families in poignant ways. Health-​care costs regularly conscript most or all of an ill person’s savings. Threadbare social services and inadequate staffing levels of nurses and aides in long-​term care commonly contribute to frail elders and chronically ill people’s feeling undignified, simply because of their frailty and physical dependences. In addition to the financial strain that caregiving imposes, family caregivers commonly experience financial strain and physical and emotional distress that diminishes their own health and longevity. At times like these, principles are often in tension with pragmatism and expedience. At present, in addition to the debate over PASE, challenges have been mounted to ethical guidelines disallowing physicians from participating in executions (Gawande 2006; Gawande et al. 2008) and forcible interrogations (Miles 2006; Miles et al. 2010) because, from a utilitarian perspective, both could be carried out in a more humane fashion if physicians performed or oversaw the procedures. These social and philosophical issues become tangible in meetings of hospital ethics committees—​as well as in analyses of real cases during academic forums. Typically the discussion involves a seriously ill individual who is suffering and requests a physician’s assistance in committing suicide. The discussion and case analysis entails identifying and balancing key principles of autonomy of the individual and the responsibility of physicians to act beneficently to alleviate suffering while avoiding harms, including abandonment of patients (Bernat 2008). For proponents of PASE, when suffering or the fear of future suffering cannot otherwise be alleviated, it seems logical to conclude that access to a lethal prescription or injection is ethically, and therefore clinically, appropriate. In contrast, opponents contend that in order to be valid, any case analysis and subsequent treatment recommendations must conform to an ethical framework built on core values and fundamental principles, which includes refraining from intentionally ending the life of a patient.

Flawed, Damaging, and Dangerous While acknowledging the sincerity and best intentions of proponents of physician-​ assisted suicide, because their reasoning is based on faulty assumptions and misweighted values, the conclusions they reach are flawed, and their actions and policy recommendations are, unfortunately, damaging and dangerous. The acts of PASE weaken the moral grounding and structural integrity of the medical profession in service to society. This damage has current, real-​world consequences that are particularly acute for the discipline of hospice and palliative medicine, which has earned formal status as the locus of special expertise in care of incurably ill and dying people and specifically affirms nonhastening of death in its founding principles (Cassel and Foley 1999; Byock 2009).

The Case Against Physician-Assisted Suicide and Euthanasia    373

Public Confusion A significant portion of the public conflates hospice and palliative medicine with assisted suicide and euthanasia (Goldstein et al. 2012). It is not at all rare in palliative care practice to encounter patients or family members who explicitly ask how palliative care differs from hastening death. The public’s confusion is understandable. The Chinese sage Confucius taught that the beginning of wisdom is to call things by their proper name. Correspondingly, calling things by improper names can sow confusion Advertisers and propagandists understand that what something is called influences how people feel about it. Organized efforts to promote PASE have skillfully conflated the meanings of important terms. “Right to die” may be an effective social marketing slogan; however, the concept does not withstand ethical or legal scrutiny. No civil right to suicide is found in social compacts such as the Magna Charta, the Declaration of Independence, or the U.S. Constitution. The United States was founded on certain unalienable rights, “that among these are Life, Liberty and the pursuit of Happiness.” The Hemlock Society, which for years was the lead organization advocating for legalizing physician-​assisted suicide morphed to become Compassion and Choices—​terms that resonate with liberal values. The organization objects to the use of the phrase “physician-​assisted suicide,” preferring terms such as “self-​deliverance” or “hastenings.” “Aid in dying” is preferred by proponents of PASE because it conflates palliative care with acts that intentionally end a patient’s life. In its brochures, newsletters, and published documents, Compassion and Choices has repeated asserted that physician-​ assisted suicide is merely part of the continuum of good medical care, including palliative care. More than a few physicians agree. Dr. Marcia Angell, a pathologist and past editor-​ in-​chief of the New England Journal of Medicine, one of the most influential medical journals in the country, has traveled extensively giving testimony in favor of assisted suicide in multiple states as well as provinces in Canada. Dr. Angell frequently refers to the need for physician-​assisted suicide because of the limitations of palliative care and contrasts lethal prescriptions with “death hastened by palliative measures” (Angell 2012). (Of note, the best available data shows that palliative measures, including sedation for otherwise intractable suffering, do not shorten survival.) (Maltoni, Scarpi, Rosati et al. 2012; Claessenset al. 2011) Dr. Timothy Quill, a prominent palliative care physician and a past president of the American Academy of Hospice and Palliative Medicine, is also a leading academic advocate for legalizing physician-​assisted suicide. Dr. Quill was the plaintiff in Vacco v. Quill, a seminal case through which proponents petitioned the U.S. Supreme Court to find a Constitutional protection for physician-​assisted suicide (Vacco v.  Quill. 1997). In a number of articles he has delineated ethical and clinical criteria for writing prescriptions for lethal medications as an option of last resort when no other therapeutic options have been effective in alleviating suffering. (Quill et al. 1992; Quill et al 2000)

374   Ira Byock Voluntary physician-​assisted death … becomes a permissible option when comfort care ceases to be effective for the terminally or incurably ill (Miller et al. 1994). We regard physician-​assisted death as a nonstandard medical practice reserved for extraordinary circumstances, when it is requested voluntarily by a patient whose suffering has become intolerable and who has no other satisfactory options (Miller et al. 1994).

Without mentioning or justifying discrepancies with those criteria, Dr. Quill published a case in which he wrote a prescription for lethal medications for a thirty-​two-​ year-​old man with a potentially incurable but treatable cancer who was not currently suffering. Instead, the patient, who was a physician-​in-​training, “requests access to the means for a hastened death from his physician as a way of reassuring himself that if it becomes clear that his treatment is failing and his suffering becomes unbearable, he will be empowered to make the choice about whether to continue living or end his life … Clearly, this patient needed to settle end-​of-​life issues at the outset of his disease, before he could move on and fully engage in treatment.” (Quill and Priddy 2002) While hospice and palliative medicine physician proponents assert that the availability of such lethal medications can reassure patients and build a therapeutic relationship, in my experience as a palliative care consultant, it is also true that some ill patients, or their family members, resist or refuse a hospice and palliative care referral out of their perception and concern that their (or their loved one’s) life will be intentionally shortened.

Slippery Slopes Proponents of legalizing PASE tend to dismiss concerns by opponents of a slippery slope as being abstract, future-​based, and unsupported by data from Oregon, the Netherlands, and Belgium. However, from the perspective of those opposed to these practices, the slippery slope has already arrived (Finlay and George 2011). The so-​called slippery slope was long assumed to refer to a dark future in which people would be coerced into PASE, a slide from voluntary to involuntary deaths (Meier 1992). Overt coercion has not occurred. However, persistent, egregious deficiencies of care undeniably contribute to some people feeling undignified, a burden to others, and that their lives are not worth living. The phrase “death with dignity” sends a cultural message to elderly or ill people that, in order to remain dignified, they must die before becoming senile or physically dependent. Socially, with well over 1 million bankruptcies attributable to medical expenses each year and the costs of home care often over $3,000/​ month and nursing home care over $6,000/​month, it is not surprising that many ill people express feeling they are a financial burden to their families. Clinically, hospice care has long been considered reliable care for people who are dying. However, no provision in Oregon’s or Washington state’s Death with Dignity Acts protects dying patients who obtain lethal prescriptions from losing hospice services if they live longer than expected. In 2010, 14% of hospice patients in Oregon were eventually

The Case Against Physician-Assisted Suicide and Euthanasia    375 discharged from hospice care. These were people with cancer, emphysema, heart failure, or dementia whose conditions improved slightly, often as a result of the meticulous care provided by hospice. They were still dying, just not swiftly enough to warrant hospice under Medicare regulations; however, they retained their lethal prescriptions. One of the safeguards designed into Oregon’s Death with Dignity Act was a requirement for physicians to refer any patient who may be depressed or have a psychological disorder influencing their judgment for psychiatric evaluation. In practice, very few people (6.2%) given lethal prescriptions have been referred for evaluations (Division OPH 2013). There is reason for concern that some people end their own lives without adequate attention to treatable depression (Ganzini et al. 2008). The slope has slipped in additional ways. While proponents were careful to acknowledge that losses of dignity or control can contribute to a wish to die, in the 1990s the case for legalizing physician-​assisted suicide relied heavily on the premise that some dying people were doomed to unrelieved pain or physical agony (Quill et al. 1992; Quill 1994; Quill et al. 2000). Those who have witnessed difficult deaths in hospice programs are not reassured by the glib assertion that we can always make death tolerable, and patients fear that physicians will abandon them if their course becomes difficult or overwhelming in the face of comfort care. In fact, there is no empirical evidence that all physical suffering associated with incurable illness can be effectively relieved (Quill et al. 1992).

In reality, since Oregon’s Death with Dignity Act took effect, the large majority of lethal drugs have been ingested in response to end nonphysical and existential suffering, (Ganzini et al. 2009) such as losing autonomy (91.2%), inability to engage in pleasurable activities (88.8%), loss of dignity (82%), and feeling like a burden on family, friends, or caregivers (38.6%); and less than 25% of completed suicides were attributed to current or the fear of future uncontrolled pain (Division OPH 2013). Proponents of legalizing physician-​assisted suicide acknowledge that pain and physical distress are not the main reasons people seek to hasten death, yet they are continuing to promote and testify in support of bills modeled after Oregon’s Death with Dignity Act as socially responsible methods of ending suffering. Whether one agrees or disagrees, it is clear that there has been a shift in rationale. A second face of the slippery slope is the extension of assistance with suicide to people who are not terminally ill. In the Netherlands in 1994, Hilly Bosscher, a fifty-​ year-​old woman who requested lethal prescription because of treatment-​resistant long-​term depression, was assisted in suicide by Dr. Boudewijn Chabot, a psychiatrist who had met her only days before. The Dutch Supreme Court found him guilty of violating legal guidelines for mercy killings but said he should not face criminal punishment. “Dr. Chabot’s lawyer, Eugene Sutorius, said the ruling established the principle that mercy killings were allowed in cases of mental suffering” (Doctor unpunished for Dutch suicide 1994). More recently, in Belgium, Marc and Eddy Verbessem, middle-​ aged twin brothers who were congenitally deaf, requested euthanasia because of near certainty that a genetic condition would cause them both to lose their eyesight and, thus,

376   Ira Byock the ability to easily communicate with one another. Authorities in Belgium debated and ultimately agreed to the brothers’ request. They died by lethal injection in December of 2012 at Brussels University Hospital (Tomlinson 2013). This case fulfills the darkest fears of many disabled people that suicide will become a socially acceptable way of dealing with disability (Coleman 2010; Coleman 1999). In the United States, an organization is publically promoting and actively providing assistance with suicide to people of sound mind who voluntarily request aid in dying without requiring a doctor to be involved or the precondition of suffering or even being terminally ill. The group, Final Exit Network, is affiliated with the more mainstream Compassion and Choices (Navasky and O’Connor 2012).

The Possibility of Narrowing Disagreements The disagreements examined here are not fully resolvable. However, a thorough understanding of principle-​based opposition does provide for a theoretical path forward. A society’s fundamental principles are durable. However, not all principles are immutable. Social compacts can evolve over time. In the United States, the clearest way for a fundamental social principle to change is to formally amend the U.S. Constitution. It has been accomplished many times in the nation’s history, first to establish a Bill of Rights but also subsequently to abolish slavery, extend citizenship to former slaves, and extend voting rights to women. In 1997, in two “right to die” cases, Washington v. Glucksberg and Vacco v. Quill, the U.S. Supreme Court found that there was no Constitutional right to physician-​assistance in suicide. However, the Court concluded that states could still enact laws permitting physician-​assisted suicide. Subsequently, three states have enacted statutes that legalize physician-​assisted suicide in defined circumstances, and the Montana Supreme Court determined that privacy and dignity provisions in the state’s constitution did not give the state authority to prohibit physicians from aiding patients in committing suicide (Baxter v. State of Montana 2009). While these actions do not obliterate the principle that doctors must not kill patients, it is undeniable that they collectively erode respect for that principle within civil society. In Roe v. Wade the Supreme Court found that a right to privacy within the due process clause of the Fourteenth Amendment to the Constitution limited a state’s authority to restrict abortions until the point of independent viability of the unborn being (Washington v Glucksberg. 1997). (It is noteworthy that this case is among the most controversial and contentious decisions in the Court’s history and is an indication of how strongly the prohibition against taking life is held by many people.) The stance of principle-​based opposition described in this chapter maintains that no right to suicide or euthanasia has historically existed, nor does one exist now. It does not foreclose the possibility that a civil right to suicide could be recognized as a moral and legal principle in the future.

The Case Against Physician-Assisted Suicide and Euthanasia    377 The distinctions between suicide, assisted suicide, and physician-​assisted suicide will be critical to exploring strategies for narrowing the disagreements discussed. Here again, proper terminology matters. Suicide can be construed as a personal and private act (albeit one that unavoidably impacts all who know the person who dies). A person who commits suicide renders mute questions about a socially recognized right to die. In contrast, by definition, assisted suicide and euthanasia entail one or more other persons and, therefore, are not solely personal or private acts. Physician-​assisted suicide by definition entails a member of the medical profession, someone who has been trained, licensed, and given prescriptive authority by formal components of society. Therefore, although it is undeniably personal for the individual who dies, legally sanctioned physician-​assisted suicide is rightly considered a social act. Despite the strenuous objections of many principle-​based opponents, it is conceivable that a future U.S. Supreme Court decision could find that a civil right of individuals to commit suicide under specified circumstances exists within the due process clause of the Fourteenth Amendment or, alternately, that such a new civil right could established by an amendment to U.S. Constitution. Were either to occur, authority to end one’s life would originate in the social compact and flow directly to autonomous individuals. Under specified circumstances, people of sound mind and qualifying medical conditions could seek legal means of ending their lives peacefully, presumably by the use of medications or gases (such as helium.) In order to mitigate the damage to the social architecture, it would be important for the laws and regulations governing such acts to specifically limit the role of physicians. A physician’s patient-​specific knowledge would be needed only for the discrete tasks of certifying an individual’s capacity to make decisions and that the individual qualifies by having the requisite medical conditions, such as an incurable diagnoses or terminal prognoses. It would be critically important for procedures governing the dispensing of lethal medications to exclude a patient’s treating physician. The doses of drugs and gases that reliably and painlessly cause death are well established. Because death is the intended outcome, overdoses are irrelevant, and rarely, if ever, is it necessary to adjust the doses of lethal agents. If adjustments of dosages or components of legally acceptable protocols proved necessary, pharmacologist, physiologists, and other nonphysicians could provide the requisite technical advice. The rationale for minimizing the role of physicians in this theoretical process is to keep each treating physician focused on improving care for people who are suffering and dying. Collectively, this must be the medical profession’s focus as well.

Expanding Our Capacity to Alleviate Suffering Dr. Robert Twycross, one of the founders of the specialty of palliative medicine, observed, “A doctor who has never been tempted to kill a patient probably has had limited clinical experience or is not able to empathize with those who suffer” (Twycross 1996)

378   Ira Byock Alleviation of suffering is a core duty of the medical profession and its practitioners. Physician-​assisted suicide and euthanasia appear compassionate—​and may seem the only compassionate responses—​when a physician cannot imagine what else he or she might do to alleviate a person’s suffering. The limitation is not fixed in the human condition; instead, it exists in the individual physician’s imagination. Collectively as well, our current inability to imagine viable alternatives inhibits us from preventing and alleviating suffering (Byock 2012). The question of what can we do when suffering persists is the proper focus for society. The medical profession is duty-​bound to focus on alleviating suffering that persists despite application of evidence-​based treatments and current best-​practice guidelines. The challenge of improving and expanding treatments to alleviate patients’ suffering should define a research agenda for the profession of medicine, particularly the discipline of hospice and palliative medicine. While our society and professions continue to devote considerable time, energy, and space in public forums, interviews, newspapers, textbooks, and journals to PASE, relatively little resources are devoted to creative responses to persistent suffering (Byock 1994). Both sides in the debate over PASE have been negligent in allowing our disagreements to divert public attention and sap political will that might lead to effective social and clinical strategies for reliably alleviating suffering and improving the quality of ill persons’ and their families’ lives. As mentioned, current practice guidelines reflect general agreement that palliative sedation employing anesthetic medications is a recommended treatment for a dying person who is experiencing persistent physical suffering (Quill and Byock 2000; Quill et al. 2009). However, there is no clear agreement on best practices for alleviating suffering of emotional, spiritual, or existential nature, which predominate among ill people who seek to end their lives. When suffering is severe and proves intractable, many palliative medicine specialists are challenged to imagine anything more to offer (Cassell and Rich 2010). This challenge should properly define a high-​priority strategic research agenda. To date it has not, and, consequently, PASE continues to be discussed as the only conceivable option for responding to suffering other than sedation. New therapeutic approaches urgently need to be developed. A few novel treatments exist yet have been all but neglected by the medical profession and relevant specialties. In the 1950s and 1960s, research into pharmaco-​assisted psychotherapy using hallucinogenic drugs showed substantial effects in alleviating severe distress among dying people (Grof et al. 1973; Pahnke et al. 1970). The studies were conducted by respected academic psychiatrists and reported in mainstream journals. Research into this treatment modality was, however, abruptly halted after several high-​profile deaths were associated with nonmedical use of LSD and related agents, which were then classified as drugs of abuse with no therapeutic use. For over three decades, while support for legally hastening death has risen among clinicians, remarkably few references to these published studies can be found in the palliative medicine literature. Research into the use of these drugs remains severely inhibited by the medications being classified as Schedule I, drugs of abuse.

The Case Against Physician-Assisted Suicide and Euthanasia    379 Recently, thanks to the persistence of a few researchers, including Dr. Charles Grob, Dr. Roland Griffiths, and Dr. John Halpern, renewed studies have been conducted. Results suggest that therapy with psilocybin is well tolerated by seriously ill patients and, for some, remarkably effective in alleviating nonphysical distress. Treatment usually entails one or two sessions, yet therapeutic benefits often persist long after any pharmacologic effect of the drug (Griffths and Grob 2010; Grob, Danforth, Chopra, et al. 2011; Slater 2012). Psilocybin has no known tissue toxicity and few reported adverse reactions. Uncommon side effects can be managed using standard doses of anxiolytics and neuroleptics. Pharmaco-​assisted therapy with psilocybin, or related medications, may or may not ultimately prove useful. However, clearly further research into this modality is warranted. Today, in Oregon, Washington, Montana, and Vermont, a physician can legally write a prescription for lethal medications for a terminally ill person but cannot prescribe medications or make a referral for pharmaco-​assisted psychotherapy that might alleviate the distress that is making the person want to end his or her life. At very least, relaxing the prevailing legal and regulatory restrictions on such research merits a place within the policy agendas of the medical profession, particularly the specialties of psychiatry and hospice and palliative medicine.

Conclusion Substantive disagreements over physician-​assisted suicide exist and are likely unresolvable. In large part, disagreements turn on different assumptions and beliefs about the nature of society, including the roles, responsibilities, and limits of the professions within society. Physician-​assisted suicide weakens the moral grounding and structural integrity of the medical profession in service to society. This damage is particularly acute for the specialty of hospice and palliative medicine in sowing confusion among the public and in diverting the discipline’s research and policy foci away from alleviating persistent suffering. In acknowledging good intentions and maintaining mutual respect, people who vehemently disagree about PASE may be able to work together to correct social and clinical deficiencies that exacerbate suffering among seriously ill and dying people. We must not allow our disagreements to hobble our collective efforts to provide the best care possible to each and every member of our society through the very end of life.

References Angell, M. 2012. May doctors help you to die? The New York Review of Books, October 11. Battin, M. P., van der Heide, A, Ganzini, L., van der Wal, G., and Onwuteaka-​Philipsen, B. D. 2007. Legal physician-​assisted dying in Oregon and the Netherlands: evidence concerning the impact on patients in “vulnerable” groups. The Journal of Medical Ethics 33(10):591–​597.

380   Ira Byock Baxter v. State of Montana. 2009.

Bernat, J. L. 2008. Ethical Issues in Neurology, 3rd ed. Philadelphia, PA: Lippincott, Williams & Wilkins. Bernat, J. L. 2012. Restoring medical professionalism. Neurology 79(8):820–​827. Bulger, R. J., and McGoveren, J. P. 2001. Physician Philosopher: The Philosophical Foundation of Medicine: Essays by Dr. Edmund Pellegrino. Charlottesville, VA: Carden Jennings Publishing. Byock, I. R. 1994. When suffering persists. Journal of Palliative Care 10(2):8–​13. Byock, I. 2009. Principles of palliative medicine. In:  T. D. Walsh, ed. Palliative Medicine. Philadelphia: Saunders Elsevier: 33–​41. Byock, I. 2012. The Best Care Possible: A Physician’s Quest to Transform Care Through the End of Life. New York: Avery. Care Consensus Panel. American College of Physicians–​ American Society of Internal Medicine. Annals of Internal Medicine 132(5):408–​414. Cassel, C. K., and Foley, K. M. 1999. Principles for Care of Patients at the End of Life: An Emerging Consensus among the Specialties of Medicine. New York: Milbank Memorial Fund. Cassell, E. J., and Rich, B. A. 2010. Intractable end-​of-​life suffering and the ethics of palliative sedation. Pain Medicine 11(3):435–​438. Claessens, P., Menten, J., Schotsmans, P., and Broeckaert, B. 2011. Palliative sedation, not slow euthanasia: a prospective, longitudinal study of sedation in Flemish palliative care units. The Journal of Pain and Symptom Management 41(1):14–​24. Coleman, D. 2010. Assisted suicide laws create discriminatory double standard for who gets suicide prevention and who gets suicide assistance: Not Dead Yet responds to Autonomy, Inc. Disability and Health Journal 3(1):39–​50. Coleman, D. 1999. Home services or euthanasia:  at the heart of the debate. Caring 18(7):16–​18, 20–​11. Division OPH. 2013. Oregon’s Death with Dignity Act—​2012. Portland. Doctor unpunished for Dutch suicide. 1994. The New York Times. June 22: 10A. Finlay, I. G., and George, R. 2011. Legal physician-​assisted suicide in Oregon and the Netherlands: evidence concerning the impact on patients in vulnerable groups—​another perspective on Oregon’s data. Journal of Medical Ethics 37(3):171–​174. Ganzini, L., Goy, E. R., and Dobscha, S. K. 2008. Prevalence of depression and anxiety in patients requesting physicians’ aid in dying: cross sectional survey. British Medical Journal 337:a1682. Ganzini, L., Goy, E. R., and Dobscha, S. K. 2009. Oregonians’ reasons for requesting physician aid in dying. Archives of Internal Medicine 169(5):489–​492. Gawande, A. 2006. When law and ethics collide—​why physicians participate in executions. The New England Journal of Medicine 354(12):1221–​1229. Gawande, A., Denno, D. W., Truog, R. D., and Waisel, D. 2008. Physicians and execution—​ highlights from a discussion of lethal injection. The New England Journal of Medicine 358(5):448–​451. Goldstein, N. E., Cohen, L. M., Arnold, R. M., Goy, E., Arons, S., and Ganzini, L. 2012. Prevalence of formal accusations of murder and euthanasia against physicians. Journal of Palliative Medicine 15(3):334–​339. Griffths, R. R., and Grob, C. S. 2010. Hallucinogens as medicine. Scientific American 303(6):76–​79. Grob, C. S., Danforth, A. L., Chopra, G. S., et al. 2011. Pilot study of psilocybin treatment for anxiety in patients with advanced-​stage cancer. Archives of General Psychiatry 68(1):71–​78.

The Case Against Physician-Assisted Suicide and Euthanasia    381 Grof, S., Goodman, L. E., Richards, W. A., and Kurland, A. A. 1973. LSD-​assisted psychotherapy in patients with terminal cancer. International Pharmacopsychiatry 8(3):129–​144. Kass, L. 2002. “I will give no deadly drug”: why doctors must not kill. In: K. Foley and H3H, eds., The Case Against Assisted Suicide: For the Right to End-​of-​Life Care. Baltimore, MD: Johns Hopkins University Press: 17–​40. Loggers, E. T, Starks, H., Shannon-​Dudley, M., Back, A. L., Appelbaum, F. R., and Stewart, F. M. 2013. Implementing a Death with Dignity program at a comprehensive cancer center. The New England Journal of Medicine 368(15):1417–​1424. Maltoni, M., Scarpi, E., Rosati, M., et al. 2012. Palliative sedation in end-​of-​life care and survival: a systematic review. Journal of Clinical Oncology 30(12):1378–​1383. Meier, D. E. 1992. Physician-​assisted dying: theory and reality. The Journal of Clinical Ethics 3(1):35–​37. Miles, S. H. 2004. The Hippocratic Oath and the Ethics of Medicine. Oxford:  Oxford University Press. Miles, S. H. 2006. Oath Betrayed: Torture, Medical Complicity, and the War on Terror. 1st ed. New York: Random House. Miles, S. H., Alencar, T., and Crock, B. N. 2010. Punishing physicians who torture: a work in progress. Torture 20(1):23–​31. Miller, F. G., Quill, T. E., Brody, H., Fletcher, J. C., Gostin, L. O., and Meier, D. E. 1994. Regulating physician-​assisted death. The New England Journal of Medicine 331(2):119–​123. Navasky, M. and O’Connor, K. 2012. The suicide plan. Public Broadcasting System. November 13. Quill, T. E., Cassel, C. K., and Meier, D. E. 1992. Care of the hopelessly ill: proposed clinical criteria for physician-​assisted suicide. The New England Journal of Medicine 327(19):1380–​1384. Quill, T. E., Coombs, L. B., and Nunn, S. 2000. Palliative treatment of last resort and assisted suicide. Annals Internal Medicine 133(7):563. Quill, T. E., and Byock, I. R. 2000. Responding to intractable terminal suffering: the role of terminal sedation and voluntary refusal of food and fluids. ACP-​ASIM End-​of-​Life Rousseau, P. 2000. The ethical validity and clinical experience of palliative sedation. Mayo Clinic Proceedings 75(10):1064–​1069. Quill, T. E., and Priddy, B. 2002. Patient 18. In: B. Priddy and T. E. Quill, eds., Palliative and End-​of-​Life Pearls. Philadelphia, PA: Hanley & Belfus: 57–​59. Quill, T. E. 1994. The care of last resort. The New York Times. July 23. Quill, T. E, Coombs Lee, B., and Nunn, S. 2000. Palliative treatment of last resort and assisted suicide. Annals of Internal Medicine 133(7):563. Quill, T. E., Lo, B., Brock, D. W., and Meisel, A. 2009. Last-​resort options for palliative sedation. Annals of Internal Medicine 151(6):421–​424. Pahnke, W. N., Kurland, A. A., Unger, S., Savage, C., and Grof, S. 1970. The experimental use of psychedelic (LSD) psychotherapy. JAMA 212(11):1856–​1863. Sikora, A., and Fleischman, A. R. 1999. Physician participation in capital punishment: a question of professional integrity. The Journal of Urban Health 76(4):400–​408. Slater, L. 2012. How psychedelic drugs can help patients face death. The New  York Times. April 20. Smith, K. A., Goy, E. R., Harvath, T. A., and Ganzini, L. 2011. Quality of death and dying in patients who request physician-​assisted death. Journal of Palliative Medicine 14(4):445–​450. Tomlinson, S. 2013. Deaf Belgian twins bought new suits and shoes before killing themselves, reveals brother who was with them when they died … but couldn’t talk them out of it. Daily Mail. January 15.

382   Ira Byock Tucker, K. L. 2012. Aid in dying:  guidance for an emerging end-​of-​life practice. Chest 142(1):218–​224. Twycross, R. G. 1996. Euthanasia: going Dutch? J R Soc Med. 89(2):61–​63.

Vacco v. Quill. 1997. 117 SCt 2293. (U.S. Supreme Court). Washington v Glucksberg. 1997. 117 SCt 2258.

Webster’s New Universal Unabridged Dictionary. 1979. New York: Simon & Schuster.

Chapter 23

G o odby e, Th omas  The Case for Physician-​Assisted Dying Margaret P. Battin

Whether Physician Aid-​in-​D ying Is Morally Permissible In the current political tempests over how we die that are brewing almost everywhere in the developed world, one side claims that legalization, tacit recognition, or even mere social acceptance of physician-​assisted suicide (as it is called by opponents) is not morally permissible. The case against physician-​assisted suicide, argued voluminously in the contemporary literature, has three core elements. If we are to participate in a world capable of reflection about fundamental social issues, we must take these claims seriously: The case against physician-​assisted suicide: 1.  Suicide is killing and thus violates universal human moral standards against killing. To be sure, these are often violated by aggressors in war, civil violence, and domestic treachery, and in many other ways, but this hardly changes the clear moral fact that killing is wrong. Thus, because physician-​assisted dying is suicide, and suicide involves killing, whether the physician directly administers the lethal potion (i.e., euthanasia) or merely supplies it by means of prescription (labeled physician-​assisted suicide or aid-​in-​dying), it is still involvement in killing and hence wrong. 2.  Were physician-​assisted suicide to become legal, it could corrupt the integrity of the physician. The physician is healer, not killer, and any swerve from this central purpose of medicine can invite corruption. Where the physician has incentives for acting other than as healer—​for instance, if the patient is difficult or abusive, if institutional and financial incentives make another course more profitable, or if the physician needs to cloak mistakes in the treatment so far—​the corruption of

384   Margaret P. Battin medical integrity is close at hand. Where the physician has an incentive to cease to serve as healer by cooperating in the self-​administered death of the patient or by directly ending the patient’s life, as recognition of physician-​assisted suicide or euthanasia would do, the risk to his or her integrity is even greater. 3.  Most fearsome of all is the risk of abuse. This is an issue of life and death, and where the procuring of death becomes legitimate medical practice, the great risk is that of the “slippery slope” leading from sometimes sympathetic acts downhill to widespread moral holocaust. Not just patients who ask for help, not just patients whose illnesses cannot be allayed, not just patients who face real suffering and pain will be helped to die, but patients in vulnerable groups like the disabled or the poor or minorities will be nudged, pressured, even forced, where no other alternatives are available, into dying in this way. Negative messages from all sources—​from immediate personal contacts to the ambient media and entrenched cultural prejudices about illness and aging—​can push in this direction. That’s the core of the argument against physician-​assistance in dying. All three of these are strong points, all to be taken seriously. But on the contrary, I respond, none of them is strong enough to defeat the central case for legalization, tacit recognition, and social acceptance of physician aid in dying. This opposite side of the case we must take equally seriously. Physician aid-​in-​dying, I hold, is morally permissible for two basic reasons. The case for physician-​assisted suicide (more properly called aid-​in-​dying): 1.  First, it is an unconscionable abridgment of one’s liberty in a society that celebrates freedom that, although you may have the right to shape your life as best you can throughout its course, when your life is coming to its end, you may lose choice over how to live the last part of your life and hence be forced to endure dying in ways that seem repugnant to you, not at all in accord with your own values. The case for aid-​in-​dying is founded on the basic principle of liberty, also called freedom and self-​determination, limited, of course, by the requirement of restricting harms to others, but nevertheless a central tenet of a free society. 2.  Second, a society that asserts a right to the pursuit of happiness, as this one does, must thus also defend the right to try to avoid the opposite of happiness: unhappiness. In its more extreme and medicalized forms, “unhappiness” includes suffering and pain. Where means for relieving or avoiding suffering and pain are available, one has the right to use them. This entailment of a fundamental principle—​the right to pursue happiness—​so fundamental to this society that it is enshrined in the Declaration of Independence—​must also make us recognize how fundamental must be the unspoken but correlative principle of a right to try to avoid unhappiness; in this case, suffering and pain. There are circumstances in which one might reasonably and thoughtfully choose to die rather than endure continuing suffering or pain. Hence, I say, these two considerations together form the overriding argument for legalization, recognition, and acceptance of physician aid-​in-​dying. A person must be free

The Case for Physician-Assisted Dying    385 to whatever extent possible to choose the character, timing, and setting of his or her own death, and a person must also be able to try to avoid suffering and pain. Where both of these considerations are in play—​it is the person’s choice, and it is made in order to avoid what the dying person considers a worse death—​they make a claim on the physician for help. It is a negative claim, in the sense that it does not oblige any specific physician to help, but it does mean that the requesting person and a willing physician are not to be interfered with or prevented from this action to which they are both willing parties. It’s as simple as that, at least from a theoretical point of view: if dying now, with means provided by the physician to guarantee an easy, painless dying process is what the patient genuinely wishes in order to avoid a harder death, and if this is something the physician is willing to do in order to ease the patient’s dying (provided, of course, that it does not harm third parties), then there can be no objection. Sadness may be part of it, and grief, perhaps an overwhelming sense of loss, even horror at the underlying ineluctability of death, yes—​but objection, no. But what about the three points listed earlier as objections to permitting physicians to provide easier, direct deaths to their dying patients—​considerations that I grant are strong points, to be taken seriously? Rebuttal to the case against physician-​assisted suicide: 1.  To the first argument, that because suicide is killing and killing is morally wrong, and that therefore physician-​assisted suicide is wrong, it may be pointed out that killing is morally accepted in some contexts: in war, in self-​defense, in defense of innocent others, and, more controversially, in capital punishment. These are all cases in which the person killed does not wish to die and is not benefited by being killed. In physician aid-​in-​dying, however, the act of life termination is the person’s genuine wish: inasmuch as he faces death in any case, the person may reasonably wish that it come sooner and in an easier way than the suffering and pain that likely otherwise awaits him. Indeed, the very terminology of the controversy over end-​of-​life measures reflects the conceptual issues here:  opponents call it “physician-​assisted suicide,” capitalizing on that term’s severely negative connotations, whereas proponents reject this term for more favorably nuanced language, “physician aid-​in-​dying,” “death with dignity,” and so on. Severely negative connotations also attach to “killing.” To assert that suicide is killing and hence wrong is simply to exploit the negative connotations of both terms. It is not an unbiased, factual, cogent argument; it is mere rhetorical suasion. 2.  To the second argument, “Were physician-​assisted suicide to become legal, it could corrupt the integrity of the physician; the physician is healer, not killer … ” it may be objected that this misunderstands the role of the physician and what it would be to corrupt it. Some physicians are already corrupt, true enough. Many are swayed by institutional incentives and disincentives of various sorts, true enough. But to assert that allowing physicians to aid their patients in dying would corrupt them is a hypothetical conjecture, not an evidence-​based claim of future fact, and it overlooks the obvious objection: not to be able to aid one’s patient in dying is

386   Margaret P. Battin itself corrosive of the physician’s integrity. In fact, physicians do sometimes find ways around the legal restrictions on overt aid in dying, but this forces them to lie, dissemble, misrepresent, collude with the family in secrecy, and, perhaps most dangerous, not be able to be fully frank with the patient and family in what the patient wishes. To the patient’s earnest request for help, does an answer like “I won’t let you suffer” mean “I’ll give you means to end your life without suffering,” or “I’ll end your life for you,” or “I’ll put you into a coma and withhold food and fluid until you die,” or what? Patient/​physician honesty is the biggest casualty here. Yes, the physician is first and foremost healer, but not every patient can be healed; on the contrary, every patient, every human being, eventually dies. It is this reality that the naïve claim that the physician is (only) healer overlooks. 3.  To the third argument, that 800-​pound gorilla about the risks of abuse, also known as the slippery-​slope argument, it may be objected that this is an exercise in conjecture, but one without firm empirical support. It warns that if physician-​assisted dying were legalized, the brunt of the policy would fall most heavily on patients in vulnerable groups. Because it is a future-​oriented prediction, there is no way to confirm or disconfirm it with certainty. But it is possible to show that there is no current or retrospective empirical evidence to support such claims. A 2007 study examining data from the two principal jurisdictions (Oregon and the Netherlands) in which physician-​assisted dying had become legal and in which data had been collected over a substantial period of time found no evidence of heightened risk compared to background populations for people in nine of ten groups identified in the literature as “vulnerable”: elderly, women, the uninsured (inapplicable in the Netherlands, where all are insured), people with low educational status, the poor, the physically disabled and chronically ill, minors, persons with psychiatric illnesses including depression, or racial minorities. The only exception was people with AIDS; this finding was based primarily on data before the development of highly active antiretroviral drugs. Extralegal cases were not the focus of this study, yet none had been uncovered in Oregon; among extralegal cases in the Netherlands, there is no evidence of higher rates in vulnerable groups. Thus, this study showed that there is no current evidence, where assisted dying is already legal, for the slippery slope claim that legalized physician-​assisted dying will have disproportionate impact on patients in vulnerable groups. Indeed, this study concluded that those who receive physician-​assisted dying in the jurisdictions in which assisted dying is legal and has been studied over time appear to enjoy comparative social, economic, educational, professional, and other privileges.1 Slippery-​slope claims, including those in medicine, are among the most difficult to address. Sometimes antecedent policies appear to have contributed directly to what is clearly moral holocaust, like Hitler’s T4 program and the Nazi euthanasia and concentration camp programs. On the other hand, the use of anesthesia in childbirth does not appear to have caused great libertinism, and the administration of smallpox vaccine did not substantially disrupt business activity in Boston when argument in Jacobson v. Massachusetts (1905) advanced this

The Case for Physician-Assisted Dying    387 as a reason for rejecting mandatory immunization, any more than making contraception available to adolescents increases their rate of sexual activity. In short, slippery-​slope claims about future outcomes, although the most powerful political argument (at least with respect to this issue) and aimed to instill fear, cannot be rooted in fact. Slippery-​slope predictions may indeed serve as a useful claim for vigilance in the future, but they cannot be used to override something for which there is a compelling case now. The upshot:  There is no sound argument against legalizing, recognizing, or accepting physician aid-​in dying. Laws that prohibit it should be replaced by those that facilitate its careful, patient-​centered, patient-​respecting and also physician-​ respecting and -​protecting practice. A  person must be acknowledged to have the right to determine, as much as possible, the way his or her life goes at its end, just as he or she is acknowledged to have the right to self-​determination and the liberty to make choices about how his or her life should go up to the point where it begins to be lost. The physician also needs respect and protection in order to be able to openly assist the patient. To come to death with the assistance of a physician is a choice made by a comparatively small number of people—​in the Netherlands, about 3% of people who die, and in Oregon about 0.3%—​a tiny fraction of deaths in either place. But where it is legal, it is a choice widely recognized to be an important, reassuring, comforting option.

Goodbye, Thomas Goodbye, Thomas. The reader will no doubt have noted that the format of this argument follows that of Thomas Aquinas, Summa Theologiae, 2a2ae, q. 64, a. 5, “Whether it is licit to kill oneself ”—​but, of course, to the opposite conclusion. Thomas held that suicide was wrong. His claim has remained central in the current controversy, not just in contemporary Thomist thinking but in a much broader range of social concern. Thomas was writing in the thirteenth century. Why should we, almost a millennium later, pay any attention to his few little paragraphs-​-​in an English translation, a mere 1,128 words long? We should attend to Thomas in part because Thomas’s argument should be analyzed for its central concerns, but perhaps even more because of the influence these words have had. They are still controlling in much of contemporary thought, even where their ancestry is not recognized. Responding to earlier turmoil in the new Christian faith, where, in the second and third centuries, Church fathers had argued over whether a virgin might kill herself in order to avoid rape by Roman soldiers or whether a believer might do so to avoid being forced in apostasy. They also recognized the lure for the ardent Christian believer of reaching a beatific afterlife, in the minds of some to be achieved by first confessing sin and repenting, then committing suicide before any new sin could be committed.

388   Margaret P. Battin Ultimately Augustine and then Thomas developed the position that suicide is wrong, a sin more grave than any that could be avoided by it. Their arguments against suicide held the day. Thomas considered issues about injustice, public authority for killing criminals, amputating infected limbs, killing oneself to avoid sin, natural law, charity, the person as part of a community, and life as a gift from God. Almost a thousand years’ worth of response to these arguments has been offered—​but, until very recently, not in contexts like the conditions of modern medical dying. An issue that had been troubling had seemed to be settled, at least once if not for all. It is now the twenty-​first century. There have been unimaginable advances made in medicine since Thomas’s time, and, in any case, Thomas was not addressing what we recognize as medical end-​of-​life issues. It is, I think, the moment to reconsider this issue, to free physicians and their patients from the constraints of a position articulated almost a thousand years ago, when belief systems were different but, more importantly, most people died at comparatively young ages of infectious and parasitic diseases, subject to epidemics, endemic infections, overwhelming plagues, and influenced by malnutrition. With some few exceptions, they did not die of lingering cancers, they did not die of the exacerbation-​and-​remission patterns of organ failure, and they did not live for decades with dementias like Alzheimer’s disease. They did not die in intensive care units, they did not become bedridden for sustained periods (pneumonia set in quickly, the “old man’s friend”); they did not die hooked up to the kinds of life-​prolonging technologies like dialysis and ventilators popularly referred to as “tubes and machines.” But that is how we die now, as many as three-​quarters of us in the developed world with access to advanced health care systems. We are grateful when advanced technologies and expert care give us continuing life that we still want, but not when the policies surrounding it mean that we cannot die when we want, when we think it is time to go, when we face more suffering than we can bear, when we see the effacement of our very selves in the name of not taking any action to cause death—​a view not so much antiquated as directed to a quite different set of problems. It may be that most people will never want aid in dying, but some will, and we must respect these choices too.2

Notes 1. Margaret P. Battin, Agnes van der Heide, Linda Ganzini, Gerrit van der Wal, Bregje D. Onwuteaka-​ Philipsen. 2007. “Legal physician-​ assisted dying in Oregon and the Netherlands: evidence concerning the impact on patients in ‘vulnerable” groups,’ Journal of Medical Ethics 33: 591–​597. 2. I dimly (very dimly) remember having seen somewhere, maybe thirty or forty years ago, an attempt to do the same thing, to utilize Thomas’s argument against suicide but backward, but I have been unable to locate such a piece. If it’s out there somewhere, my compliments to it, and, of course, I’d like to see it. In any case, the argument advanced here, even if presented in Thomas’s format, is my own.

Chapter 24

Depression a nd t h e De sire to Die Ne a r t h e End of L i fe Nathan Fairman and Scott A. Irwin

Introduction With little effort, one can call to mind a common image of “the good death.” Picture, for example, an elderly woman, lying in her comfortable bed. A warm and gentle light fills the quiet room. Family members sit nearby; friends have gathered just outside. There are tears, but also laughter, as they grieve a deep loss and celebrate a life near its end. From her bed, the woman watches with a serene smile, perhaps drifting gently in and out of consciousness, until, with one final breath, she peacefully slips away. Contrast this with the picture of death in the intensive care unit (ICU): an elderly man lying in a hospital bed, his wrists lashed to chrome railings to prevent him from pulling out the tubes that bind him to the machines that fill the space. One can hear the beeps and alarms from the monitors, the mechanical hiss of a ventilator, groans of discomfort. Family members pace in the hallway outside. Under a fluorescent glare and sealed behind gowns and gloves, nurses and doctors observe the machines, keeping watch over the body, until it draws a final breath. These images are caricatures, to be sure. In reality, the scenes of dying are likely as varied as notions of “the good death.” Nonetheless, these archetypes are reinforced by clinical experience and depicted frequently in documentary accounts of the dying process. For example, in June 2012, Time magazine’s cover article—​provocatively titled “How To Die”—​described the last months of the author’s parents’ lives, a period marked by aggressive interventions and a kind of powerlessness to script for them a more peaceful final chapter (Klein, 2012). Similarly, the 2010 Frontline documentary, Facing Death, chronicled the experience of patients, families, and physicians in an ICU, each struggling with the possibility that their actions, intended to prolong life, instead prolong

390    Nathan Fairman and Scott A. Irwin a terrible kind of suffering before a certain death (Facing Death, 2010). And, a decade earlier, Bill Moyers’s acclaimed PBS series On Our Own Terms examined the challenges that “emerge from balancing medical intervention with care and comfort at the end of life”—​neatly framing the two images of death described in the preceding vignettes (On Our Own Terms, 2000). There is an implicit moral claim that runs through these examples, one captured obliquely in the title of Moyers’s series. The claim is rooted in ideas about autonomy and ownership, and, arguably, it is one of the central moral tenets of the modern hospice movement: that death belongs to the dying and not to the doctors. After all, facilitating a “self-​determined life closure” is one of the core aims of hospice care, along with comprehensively addressing symptoms and enabling effective grieving (Ferris et al., 2002). Hospice has been positioned to offer the possibility that dying patients and their families, with the help of a multidisciplinary team of experts, might be able to orchestrate a peaceful final act. This is often cast as an antidote to death in the ICU. Hence, it is not difficult to see broad moral undertones in the shift from “medicalized” dying in the ICU to the person-​centered end-​of-​life care that is the goal of hospice. That is, hospice and, more broadly, palliative medicine aim to restore the individual patient (better: the individual person) as the pilot in this final phase of life. In practice, though, whether one is ultimately able to have “a death of one’s own choosing” (a phrase from Moyers’s series [On Our Own Terms, 2000]) depends on far more than just the presence of a capable hospice team. Diseases evolve along unpredictable paths, family and friends may not be able or willing to participate in the ways expected, personal financial constraints and system-​level resource issues can limit options for care, sometimes even the hospice team may not be fully prepared to honor the wishes of a dying patient—​particularly when there are requests to hasten death. In sum, success in achieving a so-​called good death can be undermined by a process that is unpredictable and not easily controlled. This chapter focuses on depression near the end of life, how it can undermine a patient’s ability to achieve a “self-​determined life closure,” and how it can undo a clinician’s willingness to support the patient’s choices. Consider the terminally ill man who chooses to stop a life-​sustaining intervention. In general, there is broad agreement within bioethics and the law that an informed decision to stop treatment should be supported, even when it will hasten death. Certainly, one of the aims (arguably, the main purpose) of hospice care would be to honor this choice by helping to guide decision-​ making, by anticipating and addressing distressful symptoms, by providing emotional and spiritual assistance to the patient and family, and more. But suppose the man’s physician suspects he has developed depression and that the depression is influencing his desire to stop treatment. Now the physician hesitates. The appropriate stance, with respect to supporting the patient’s preferences (which is to say, respecting his autonomy) is far less certain. Before going further, its worth considering what gives depression this authority. How is it that the suspicion of depression can make the physician pause, when otherwise the clear path would be to support the patient’s decision? Within psychiatry, professional

Depression and the Desire to Die    391 norms have traditionally viewed any statement of desire for death as a sign of psychopathology, evidence of incapacity, and, often, justification for involuntary hospitalization and/​or treatment (Leeman, 1999; Parker, 2004; Sullivan & Younger, 1994). But tradition is not the only guide—​and, as others have argued (Parker, 2004; Bostwick & Cohen, 2009), it is not at all the best one here. Still, sadness, hopelessness, thoughts of death, and despair are among the defining features of depression. By its nature, depression can create a certain kind of emotional and cognitive experience with the power to make death a desirable state. As well, several studies have established compelling links among depression, hopelessness, suicidal ideation, and desire for hastened death (Chochinov, Wilson, Enns, & Lander, 1998; Breitbart et al., 2000). One would reasonably hesitate when irreversible decisions appear to satisfy the desires borne of an illness. In this context, decisions to limit treatment with the aim of hastening death ought to be examined with deliberate care. So, what should the physician do? Can the depressed patient make the decision to limit treatment, even if it brings about his death—​even if that is his aim? Are there clear lines that separate capacity from incapacity or useful tools to assist the physician in making these determinations? Is there an obligation to treat the depression? To what limit? Does it matter how long the patient has to live? Does depression near the end of life, coupled with a desire for hastened death, represent a wish for suicide, or is this some other, ethically distinct, life-​ending act? Could it be that death belongs to the dying, but only if they are not depressed? In this chapter, we ask how depression may influence one’s capacity to make life-​ shortening decisions in the context of end-​of life care. Since the first step for the clinician is to characterize the suspected depression, we begin with a review of depression and similar clinical conditions seen in the setting of care near the end of life. Second, we briefly examine the construct of capacity and the elements of its assessment. Third, we review data about the relationship between depression and decisional capacity, followed by a discussion of considerations unique to this issue near the very end of life. Along the way, some of these questions will be resolved, but we will find that others lack clear answers. For this reason, we will close by providing some guidance from our own experience in situations like this by describing a posture that we think one should adopt and a process to help guide decision-​making.

Vignette An example may help to anchor the discussion: Mr. P. is an eighty-​five-​year-​old man with recurrent acute myeloid leukemia. Once an accomplished academic as well as an avid outdoorsman, he has become increasingly weak and debilitated with the progression of his disease. He counts his sister, his niece, and a former student among his closest friends and family. Despite their help and the

392    Nathan Fairman and Scott A. Irwin care he receives from his hospice team, his quality of life has suffered. A bothersome tremor and poor concentration have left him unable to write, and he no longer has the stamina to get outdoors. The social worker on his hospice team has become concerned that he may be depressed. She is alarmed, in particular, by some oblique references to wishing his life were over. She arranges a consultation with the hospice psychiatrist. The psychiatrist is struck by how cogent and thoughtful Mr. P. is about his situation. He is deeply touched by the love his friends and family have shown and grateful for the care he’s received. He acknowledges a deep loss of meaning, and he is profoundly hopeless about the possibility of finding any purpose in the time he has left to live. The hospice team believes he is in the last few months of life—​a prognosis that seems unbearably long to him. “All I look forward to is my final breath,” he says. He mourns the loss of his independence, and he dreads becoming a further burden on others. “My fond wish is to die; to take my last exit; the sooner, the better; and preferably in my sleep.” He is not thinking of any particular way to end his life, although he has begun to wonder what would happen if he stopped taking his medication or if he stopped eating. He invites advice from the psychiatrist about what the hospice team might be willing to do to help.

Depression (and Its Look-​Alikes) Psychiatrists involved in care near the end of life are familiar with requests to evaluate patients like Mr. P. Such consults come in a variety of disguises—​as requests to evaluate for depression, to assess capacity, to determine suicidality, and the like. (On one occasion, one author was even asked to assess “end-​of-​life ideation”—​as though a patient near the end of life ought not to think about it!) In responding to consult requests like this, psychiatrists frequently hear some version of the following:  “of course he’s depressed; he’s dying of cancer!” Such statements can come from patients, families, or fellow-​clinicians, and they carry a certain kind of logic in that they normalize the emotional experience of coping with an advanced illness. There is also a warning embedded within such statements, one that cautions against the danger of pathologizing normal human experience. At the same time, such statements can trivialize or mask more profound varieties of suffering. Hence, we begin with a very small but important distinction: “feeling depressed” is not the same as “having depression.” In the broadest view, depression (in the sense of “feeling depressed,” or “sad,” or “down,” etc.) is a part of the normal range of human experience—​and perhaps even a universal experience as life draws to a close. This is to be distinguished, though, from depression in the sense of “having depression”—​that is, a serious disturbance of mood, marked by either profound sadness or the inability to experience pleasure and associated with cognitive and somatic symptoms such as feelings of hopelessness, worthlessness, thoughts of death, and changes in sleep and appetite—​all contributing to significant impairments in function. Mental health clinicians will recognize this cluster of symptoms as the core condition of Major Depressive Disorder, as described in the Diagnostic and Statistical Manual (DSM) (American Psychiatric Association, 2013). However, the DSM’s neat description of clinical depression (Table 24.1) belies the fact that, in practice,

Table 24.1 Major Depressive Disorder and Related Conditions Condition

Characteristics

Major Depressive Disorder (MDD) (Breitbart et al., 2000)

A. Five or more of the following present, over at least two weeks, and at least one of the symptoms is either depressed mood or anhedonia: 1. Depressed mood (or, in children/​adolescents, irritable mood) 2. Anhedonia: markedly reduced interest/​pleasure in most activities 3. Changes in weight or appetite 4. Insomnia or hypersomnia 5. Psychomotor agitation or retardation 6. Fatigue or diminished energy 7. Feelings of worthlessness or excessive guilt 8. Poor concentration or indecisiveness 9. Recurrent thoughts of death, suicidal ideation, or suicidal behavior B. Symptoms cause clinically significant distress or functional impairment. C. Symptoms are not the result of substances or a medical condition. D. Symptoms are not better explained by one of the psychotic disorders. E. The patient has never experienced mania or hypomania.

Adjustment disorder with depressed mood (Breitbart et al., 2000)

Emotional/​behavioral symptoms that develop within three months of an identifiable stressor. Symptoms are disproportionate to the severity or intensity of the stressor. May occur with features of depression, anxiety, behavior, or any combination. Note: If criteria are met for MDD, then MDD should be diagnosed and not adjustment disorder.

(continued)

Table 24.1  Continued Condition

Characteristics

Grief (Breitbart et al., 2000)

In grief, the predominant emotional state is characterized by emptiness and loss; in MDD it is depressed mood and/​or inability to experience pleasure. In grief, dysphoria often occurs in waves, generally triggered by thoughts or memories of the deceased; in MDD dysphoria is unrelenting, and cognitions center on worthlessness/​hopelessness. In grief, the mood state is reactive (i.e., individuals can have periods of happiness, laughter, etc. in relation to pleasant or humorous experiences); in MDD, the mood state can be pervasive or intractable. In grief, self-​esteem may be preserved, and, if feelings of guilt are present, they are usually constrained to the relationship with the deceased. In grief, thoughts of death often concern “joining” the deceased; in MDD, they are aimed at ending one’s own life and rooted in feelings of hopelessness and worthlessness. Note: Bereavement is no longer an exclusion criteria for MDD. Even in the setting of bereavement, if criteria are met for MDD, then MDD should be diagnosed and appropriate treatment initiated.

Demoralization (Jacobsen et al., 2010)

Core features: hopelessness, helplessness, meaninglessness, existential distress. “Subjective incompetence” In demoralization, improvement is undermined by profound sense of incompetence and hopelessness; in MDD, improvement is hindered by problems with motivation and drive. In demoralization, perceived (im)possibility of future improvement is experienced with distress; in MDD, it is experienced with apathy or indifference. In demoralization, reactivity of mood is preserved; in MDD, it is frequently lost. There is disagreement about whether there is sufficient evidence for demoralization as a separate diagnostic category.

Depression and the Desire to Die    395 it is often difficult to distinguish this condition from nonpathologic states of sadness, as well as from related clinical conditions such as demoralization and grief. Indeed, for the psychiatry consultant asked to evaluate a dying patient with “depression,” the initial task frequently involves distinguishing among these different states. In the setting of an advanced, serious illness, differentiating major depression from normal states of sadness can be quite challenging, even for experienced clinicians. Patients with serious illness often experience intense sadness; they may endure periods of anhedonia, low motivation, or even hopelessness; and it should be expected that patients facing the end of life will also contemplate death. Yet none of these phenomena, taken alone, indicates the presence of pathological depression. Similarly, the somatic dimensions of major depression (e.g., changes in sleep, low energy, changes in weight and appetite) are precisely the kinds of physical symptoms that patients experience in the advanced stages of a serious illness, and so these are not reliable indicators of depression either. Instead, experts in palliative care psychiatry give greater weight to symptoms in the emotional and cognitive domains, as well as to changes in mood from baseline and the intensity and time-​course of symptoms (Block, 2000). Thus, intense feelings of worthlessness, hopelessness, guilt, or thoughts of suicide (distinct from simply thinking about death) suggest the presence of major depression. Similarly, true anhedonia, in which the patient has lost the ability to experience pleasure (as distinct from being unable to engage in pleasurable activities due to physical limitations), also helps to identify major depression. A further complication is that major depression is not the only clinical condition marked by sadness:  several other important entities may overlap with, and may be mistaken for, major depression (Table 24.1). Adjustment disorder with depressed mood occurs in the context of an identifiable stressor, in which the patient experiences marked distress to a degree in excess of the intensity of the stressor. In theory, the approach to “treatment” in adjustment disorder is nonpharmacologic, aimed at bolstering coping strategies or resolving/​removing the stressor. Grief, the emotional experience associated with a significant loss, is also distinct from major depression, although the two conditions have in common the experience of a depressed mood (Jacobsen, Zhang, Block, Maciejewski, & Prigerson, 2010). Of note, in the most recent iteration of the DSM, the “bereavement exception” was removed from the diagnosis of major depression, so that even in the setting of bereavement, major depression should be diagnosed (and treatment considered) if criteria are met (American Psychiatric Association, 2013). This distinction is important because the general approach to addressing grief in the absence of major depression involves supportive therapeutic interventions and not drug therapy, although there can be a limited role for medication to target specific symptoms such as insomnia. Two common neurocognitive disorders, delirium and dementia, are also frequently mistaken for depression, particularly when marked by social withdrawal, psychomotor retardation, and diminished motivation. In both conditions, however, the predominant symptom is a significant cognitive disturbance, with an onset that is either insidious (in the case of dementia) or acute/​subacute (in the case of delirium). Although cognitive

396    Nathan Fairman and Scott A. Irwin deficits can be seen in major depression, these tend to arise only after the emergence of changes in mood or the development of anhedonia. Of course, identifying a neurocognitive disorder (mislabeled as “depression”) presents its own set of concerns with regard to autonomy in the setting of end-​of-​life care, outside the focus of this chapter. A final condition, demoralization syndrome, warrants particularly careful consideration from the standpoint of concerns about capacity and decision-​making at the end of life. Demoralization syndrome captures a suite of psychological phenomena commonly seen in patients with advanced, serious illnesses, many of which overlap with major depression. The core features of demoralization include hopelessness, helplessness, meaninglessness, and existential distress (Kissane, Clarke, & Street, 2001). A  cornerstone of demoralization, distinguishing it from depression, is the individual’s experience of “subjective incompetence,” the sense of ineptitude with respect to being able to make improvements arising from the loss of purpose and meaning (de Figueiredo & Frank, 1982). In major depression, problems with motivation and drive hinder progress toward improvement. By contrast, in demoralization, improvement is undermined by this profound sense of incompetence and hopelessness. Faced with the perceived (im)possibility of future improvement, the depressed patient experiences apathy or indifference, whereas the demoralized individual experiences distress and despair. A further distinction can be made with respect to happiness: the demoralized individual tends to retain reactivity of mood (i.e., he or she can experience happiness in relation to positive events), whereas that capacity is frequently lost in depression. There is considerable debate as to whether or not demoralization syndrome can be reliably distinguished from major depression. Furthermore, although some contend that demoralization de-​pathologizes emotional distress in serious illness (by correcting for misdiagnosed major depression or adjustment disorder; Slaveny, 1999), others argue that it medicalizes normal experience (and that, in doing so, it justifies a particular moral claim about autonomy in end-​of-​life decision-​making; Parker, 2004). Before moving on, a final comment concerning whether distinguishing among these depressive conditions is important from the standpoint of assessing capacity: none of these conditions necessarily precludes a patient’s ability to make informed decisions, even ones about withdrawing or withholding interventions. This is discussed in greater detail later. Nonetheless, a careful assessment may reveal the severity of disordered thinking (which could undermine capacity) and may also elucidate which avenues are most likely to alleviate distress and (if relevant) help to restore capacity.

Capacity A full discussion of capacity is beyond the scope of this chapter, and the interested reader can find detailed treatments elsewhere (Grisso & Appelbaum, 1998; President’s Commission, 1982). However, several comments warrant mention before moving on to examine the relationship between capacity and depression. First, “capacity” refers to an

Depression and the Desire to Die    397 individual’s ability to make decisions about his or her own health, and it relates to the principle of respect for autonomy: adults are presumed to have decision-​making capacity as an extension of respect for their intrinsic worth and their right to control their own lives. In turn, capacity relates to the doctrine of informed consent, in that individuals with decisional capacity who have been provided the relevant medical information, in the absence of coercive influence, can exercise their autonomy in making health care decisions regardless of whether those decisions accord with the physician’s own recommendations or preferences. Questions about capacity are also connected, although often in ways that are less apparent, to the principle of nonabandonment in the sense that to follow the “preferences” of a patient who lacks capacity is to risk abandoning the individual and disrespecting his or her autonomy. Second, any physician is qualified to assess capacity, although in practice many clinicians have poor understanding of what capacity is, how it relates to the ethical care of patients, and what elements are entailed in an assessment. A psychiatrist’s expertise is frequently sought in challenging cases or when a known or suspected mental illness (particularly one that undermines cognitive function or impairs reality testing) provides the basis for questioning the patient’s capacity. In such cases, the psychiatric assessment is often positioned as a second opinion. Third, the elements of capacity generally include understanding, appreciation, reasoning, and choice. The patient must demonstrate the ability (1) to understand relevant information, (2) to appreciate the particular medical situation and its consequences, (3) to reason about treatment options, and (4) to communicate a choice (Appelbaum, 2007). In addition, the clinician should ensure that the risks, benefits, and alternatives have been discussed with the patient, and the basis for questioning capacity should also be acknowledged and documented. Fourth, questions of capacity should relate to specific interventions (a proposed amputation, a trial of dialysis, etc.). Relatedly, most experts accept the notion of a “sliding scale” threshold, one based on the likelihood and significance of risks and potential gains (President’s Commission, 1982). Thus, the threshold for demonstrating capacity to decline a high-​gain/​low-​risk intervention would be higher than the threshold to decline a low-​gain/​high-​risk intervention. Several important conclusions flow from these two observations. In some situations, a patient may have the capacity to accept, but not to decline, a particular intervention—​when there are major differences in the magnitude of risks and benefits when it comes to accepting versus declining the intervention. Also, it follows that capacity relates to particular qualities of reasoning about particular interventions, and so the presence of a comorbid psychiatric condition (major depression, Alzheimer’s dementia, schizophrenia, etc.) cannot be used as a simple substitute to determine the question of capacity. Rather, more discrete abilities of cognition and reasoning, which may be impaired in different degrees in any particular patient, influence the capacity to provide informed consent. In short, as alluded to in the previous section, the presence of any particular psychiatric diagnosis alone does not provide a priori justification for determining that the patient lacks capacity.

398    Nathan Fairman and Scott A. Irwin Finally, we note that published reviews and educational resources about the assessment of capacity tend to leave the impression that such assessment is a mere matter of applying certain criteria in the context of a thoroughgoing conversation with the patient. In practice, however, capacity assessments in complicated cases are rarely straightforward, and they often involve an uncomfortable degree of subjectivity and judgment (Brody, 2009).

Depression and Capacity A considerable amount of research has explored the relationship between psychiatric illness and decisional capacity. These studies have examined either psychiatric inpatients or those seen by psychiatric consult-​liaison specialists in acute-​care hospitals, and they focus on the general capacity to provide consent to treatment (Candi & Barba, 2011). Of note, in both settings, incapacity is the exception, not the rule; most individuals retain capacity to provide consent. Among patients in acute-​care hospitals, incapacity is often the direct result of impaired cognition due to a neurocognitive disorder (e.g., delirium and dementia). By contrast, among psychiatric inpatients, incapacity is most commonly associated with a psychotic disorder (e.g., schizophrenia), which can undermine capacity as a result of delusional thoughts, ideas of reference, or other impairments in “reality testing.” Mania frequently undermines capacity as well, also through delusions (often grandiose in nature) or from impaired insight and judgment as to the need for treatment (Cairns et al., 2005). With respect to depression, roughly one-​fifth of patients with depression lack the capacity to provide general informed consent, with the likelihood of incapacity related to the severity of depression (Grisso & Appelbaum, 1995; Vollmann, Bauer, Danker-​Hopfe, & Helmchen, 2003). These data, however, are of limited usefulness in several respects. For example, they pertain to the general construct of capacity (i.e., the ability to provide informed consent to treatment in general), as opposed to the ability to provide consent for specific interventions—​which most experts agree is the appropriate focus for questions of capacity in any particular case (Appelbaum, 2007; Sessums, Zembrzuska, & Jackson, 2011). More importantly, these data fail to shed light on what it is about depression that can undermine capacity. Separate from data about prevalence, the relevant question here, from the standpoint of ethics, is “what is it about depression that can undermine capacity and thereby ‘taint’ the process of consent?” In fact, there is some empirical data relevant to this question, although it does not pertain specifically to the setting of end-​of-​life care. For example, it is in the nature of depression to impair a patient’s ability to recognize that depression is present and that treatment can help. In fact, a relationship between poor insight and diminished capacity has been observed empirically: in their study of psychiatric inpatients, Cairns et al. (2005) reported a strong association between lower scores of insight and decisional incapacity. Similarly, cognitive deficits, which commonly accompany depression, may impair capacity in the areas of

Depression and the Desire to Die    399 understanding, appreciation, or reasoning. Again, some recent empirical data support this claim: Ghormley et al. (Ghormley, Basso, Candlis, & Combs, 2011) found that in depressed adults discrete neurocognitive deficits in new learning were linked to decisional incapacity in the specific area of understanding treatment disclosures. Taken together, these two dimensions of the experience of depression—​impaired insight and neurocognitive deficits—​may in some cases be so significant as to undermine capacity. Beyond measurable impairments in insight and cognition, there are other dimensions of the depressive conditions that are likely to influence capacity, although in these areas there is an absence of empirical data. For example, Sullivan and Youngner observe that the dimension of capacity most vulnerable to distortion in depressed patients is the ability to appreciate their unique clinical situation and its consequences, for a variety of reasons: Distorted assessments of self, world, and future … typify depressive thinking. Depressive helplessness produces an underestimation of one’s possible effectiveness in the face of serious illness. Guilt and worthlessness may make one believe that suffering and death are deserved and should not be forestalled. Anhedonia may make it impossible to imagine that life will offer any pleasures for which it is worth enduring the discomforts and indignities of medical illness. Depressive hopelessness can make it impossible to imagine that life will ever offer a better balance of pleasure and pain than it does at present. If some of these cognitive symptoms are present, competence to make decisions concerning life-​sustaining medical treatment may be impaired. (Sullivan & Younger, 1994, p. 976)

Kissane makes a strikingly similar observation: “the key issue in affectively disordered patients is usually their capacity to appreciate the significance of any information for their personal circumstances” (Kissane, 2004, p. 25, emphasis added). Demoralization and other affective disturbances fundamentally influence the patient’s “assumptive world”—​a set of beliefs and attitudes about life and one’s place in the world and an important filter through which individuals interpret factual information about their disease and its import to their life. “Depression and demoralization can disrupt this assumptive world, causing a person to have a disordered relationship to his future” (Kissane, 2004, p. 26). In this manner, Kissane argues, demoralization can constitute a “morbid mental state” that, in its most pathologic form, “may have the capacity to interfere with informed consent” (Kissane, 2004, p. 24). Despite clear disagreements in the literature about the merits of parsing out different affective states—​major depression, grief, demoralization, even “fractured dignity” (Chochinov et al., 2002)—​there seems to be consensus that, as a practical matter, getting the correct diagnosis does not solve the dilemma of determining capacity. Each condition exists along a continuum of severity, with impairments in different domains—​only some of which may influence capacity in any particular case. Ultimately, it seems, the clinician must rely on conducting a very careful assessment, paying close attention in particular to the ways in which the affective state influences the patient’s appreciation of the medical situation and its consequences.

400    Nathan Fairman and Scott A. Irwin

Depression and the Desire to Die Near the End of Life Returning for a moment to the vignette, the broadest questions concern to how the physician and her hospice team ought to respond to Mr. P. in this particular context. Addressing concerns about capacity would be an essential element of the physician’s response. However, there are also other important, ethically relevant factors at play. Given that there are not distinct lines to separate capacity from incapacity, these other important contextual matters (which seem to get short-​shrift in the literature about capacity near the end of life) should also receive careful consideration. First, it bears mention that there are effective treatments for depression, even near the end of life (Block, 2000; Irwin, Fairman, & Montross, 2012). A variety of medications and psychotherapies have been shown to be beneficial, and, for this reason, if depression is suspected, a time-​limited therapeutic trial should be recommended. At the same time, the clinician should acknowledge the limits of therapy because a sizeable portion of treated patients may not respond to therapy or may not respond quickly enough. In our experience, however, the more common error is to underestimate the chances of improvement with treatment. Second, effective treatment also appears to influence end-​of-​life preferences and desire to die. Ganzini showed that treatment of depression increased patient preferences for life-​sustaining therapies among severely depressed adults in an inpatient psychiatric unit (Ganzini, Lee, Heintz, Bloom, & Fern, 1994). Similarly, Breitbart showed that depression and hopelessness are the strongest predictors of desire for hastened death among patients with advanced cancer (Breitbart et al., 2000), and, in a second study of patients with advanced AIDS, effective treatment of depression was significantly associated with reductions in desire for hastened death (Breitbart et al., 2010). Taken together, these data argue for the importance of a careful assessment of depression, coupled with aggressive treatment when indicated, in patients who seek hastened death in the context of an advanced medical illness—​separate from considerations about capacity to make decisions that will hasten death. These interventions may have the dual effect of ameliorating depression and reducing the desire to die—​each of which can improve quality of life near the end of life for patients and their families. Third, clinicians should adopt a measured view of “the stakes” involved in decisions to limit or stop treatments near the end of life, and they should help other stakeholders to see the stakes in the same light. The language used in discussions about decision-​ making near the end of life frequently belies an assumption that such decisions are “high-​stakes”—​using language, for example, that seems to frame decisions as life-​or-​ death choices. Even in the academic literature about desire for hastened death near the end of life, there is often an overemphasis, in our view, on the finality and irreversibility of a decision to hasten death, without giving due weight to the context: that is, that the alternative is generally some other kind of death, equally final and irreversible, some short period later. In our experience, patients who desire to die sooner rather than later seem less apt make this mistake. They are, in many cases, very deliberately trying to choose one kind of death because doing so will enable them to avoid some other kind,

Depression and the Desire to Die    401 which in their view would be far worse. We suspect that maintaining a more measured view of “the stakes” may help to reduce distress around such decision-​making near the end of life, for patients, family, and clinicians. It follows from this that a thorough understanding of the patient’s perspective and preferences, even concerning the very difficult topic of how the final moments of life will go, is essential to this process. In the same vein, when patients nearing the end of life begin to give voice to a desire to die, clinicians considering psychiatric intervention would do well to consider that death in a psychiatric hospital (after failed compulsory psychiatric treatment) might be a far worse fate than a comfortable, hastened death (even if influenced by the presence of depression). Fourth, and related, others have argued that proximity to death, in providing justification for decisions that hasten death, may not carry as much moral weight as it seems (Cellarius, 2008; Jansen, 2003). In judging the risks and benefits of pursing a path of hastened death, though, there is an important sense in which proximity to death does matter: as one’s life draws closer and closer to its end, the likelihood of some improvement in quality of life becomes diminishingly small. As a consequence, less and less is likely to be lost (in terms of a missed opportunity for improvement) if death comes a little bit sooner. A fifth consideration relates to “the stakes” as well, but in a different way: the final period of a loved one’s life will have a long-​lasting impact on grief and adaptation for the bereaved, particularly for those who act as surrogate decision-​makers. For this reason, family members and other loved are often major stakeholders to the decisions about care near the end of life, and their views should be carefully considered.

What to Do: Posture and Process We close with a summary and suggestions, based on our own clinical experience, for how to proceed. We have found that the posture and the process are of critical importance in navigating requests for hastened death from depressed patients in the palliative care setting (Table 24.2). By “posture,” we mean that the clinician’s stance and manner of engagement with the patient will help to shape the decisions and improve the quality of what is often a challenging experience for the patient, family, and providers. When patients give voice to a desire to die, they are often met with well-​intentioned responses that leave them feeling alone, misunderstood, or unheard. Instead, a more effective posture is one that conveys curiosity and concern about the patient’s suffering, compassion for their situation, commitment to try to be of help, willingness to collaborate with other stakeholders, humility about the limits of expertise, and confidence in the process to help address their suffering. In terms of “process,” a first step should be to ensure a thorough assessment of the patient’s distress in all dimensions, using the expertise of a multidisciplinary palliative care team. Interventions aimed at relieving suffering should be structured in time-​ limited therapeutic trials, with clear aims and time course. These are basic elements of palliative care interventions, and they should be emphasized and explicitly discussed

402    Nathan Fairman and Scott A. Irwin Table 24.2 Posture and Process for Addressing Depression and Desire to Die Near the End of Life Posture

Clinicians should cultivate a posture that conveys: Curiosity and concern about the patient’s suffering Compassion for the patient’s situation Commitment to try to be of help Collaboration with other stakeholders Humility about limits of expertise (given diagnostic and therapeutic uncertainty) Confidence in the process

Process

Clinicians should adopt a process that ensures: Expert, multidisciplinary assessment and management of all dimensions of distress (“Total Pain”) Clear recommendations for therapy, when appropriate Time-​limited, therapeutic trials—​with clear aims and time course Clarification of prognosis (seek second opinion if appropriate) Expert assessment of capacity: • Attention to ability to appreciate situation and consequences • Attention to authenticity of choice Expert assessment of depression (preferably by palliative care psychiatrist): • Distinguish major depression from related conditions • Close attention to hopelessness • Firm recommendation for appropriate therapy, taking into account prognosis and goals of care (may ameliorate depression and reduce desire to die) Consideration of other ethically relevant issues: • Measured assessment of “the stakes” (most often, different kinds of death) • Impact on other stakeholders (especially family and surrogates) Open, direct communication among stakeholders (patient, family, clinicians)

with the patient and loved ones. We have seen that capacity needs to be assessed very carefully and with particular attention to the patient’s ability to appreciate his or her particular medical situation and its consequences because this is the domain of capacity most vulnerable to depression and related affective states. Also, careful diagnostic clarification of depression is important, but it will not directly influence the capacity assessment because the presence of any particular depressive condition cannot be used as a proxy for incapacity. Rather, diagnostic clarification may help to elucidate potential therapeutic avenues that might have the benefit of ameliorating distress related to depression and reducing the patient’s desire to die. Last, requests to hasten death near the end of life occur in a context laden with many ethically relevant considerations, although often attention is drawn to the question of capacity as though it is the only thing that matters. Awareness and careful consideration of these other factors, alongside clarification of capacity, may help the clinician to maintain solid footing when faced with a request from a depressed patient to hasten death near the end of life.

Depression and the Desire to Die    403

Vignette There remain, as expected, some unanswered questions (e.g., “under what circumstance would it be appropriate to compel psychiatric treatment, and to what limit?”), but we argue that, in most cases, adopting the posture and process can help to achieve an acceptable solution even in the most complicated cases. The vignette of Mr. P. exemplifies this: The psychiatrist asks Mr. P. about other troublesome symptoms. Low energy and weakness are his main physical complaints. In fact, he shares that he collapsed in the bathroom two weeks ago. Lacking the strength to get up, he lay there for hours until his sister happened to visit. He is deeply shaken by the idea of dying in the bathroom, his sister discovering his body. That event led to his transfer to the hospice inpatient facility. Mr. P. shares that he is not troubled spiritually. He feels he has not left any “unfinished business” in relationships or other areas of his life. The psychiatrist asks Mr. P.  if it’s possible he’s become depressed. He seems to have trouble relating to this concept, but after considering it he shrugs and allows that “my condition, it leads to a pessimism, a depression—​if you like.” He is more clear about anhedonia: even the pleasurable activities that are still physically possible for him—​reading a book, spending time with his family—​have become tedious and unenjoyable. A short conversation with his sister and niece confirms this report. They add that nothing seems to lift his spirits, which is out of the ordinary for him. They are aware that he’s been wishing his life would end—​he’s shared the idea with them, vaguely. Though this leaves them sad, they quickly add “it’s understandable; he’s just waiting to die.” The psychiatrist acknowledges to Mr. P. that life has become rotten: “this illness seems to have stolen everything important from you.” She affirms that she will do her best to help him, and she encourages Mr. P. to allow her to facilitate a conversation with his family and the hospice team about his distress, including his desire to die. She also shares her suspicion that he’s become depressed, and she asks if he would consider allowing her to suggest a medication, if she thought it might help quickly. They discuss a short trial of methylphenidate, aimed at boosting his mood and energy. He relents, after being assured that the medication will not prolong his life—​though he adds, “it won’t change my mind.” This is framed as a five-​day trial. By the second day of treatment, Mr. P shows noticeable improvements: he is markedly more engaged in conversation, and his mood appears more reactive to good humor. After “putting up with” being moved to a chair outdoors, he acknowledges a rediscovered pleasure in simply sitting in the sun. His sister and niece comment on the improvements, sharing their own relief at seeing him in better spirits. He himself is more reserved: he acknowledges feeling better, but he adds that these changes hold little deep value to him given the context of his life: they do not change his readiness for life to end. The psychiatrist discusses Mr. P.’s case with his multidisciplinary team. The other clinicians feel his distress has been addressed in most areas, save for the sense of meaninglessness in waiting to die. They discuss how to respond to his requests to end his life. Although there is disagreement about this, the consensus is that active

404    Nathan Fairman and Scott A. Irwin means of hastening his death (palliative sedation, for example) would be difficult to support. There is agreement around supporting Mr. P. in stopping eating and drinking. The psychiatrist shares her sense that he has the capacity to engage in decision-​ making about this, and others agree. The psychiatrist discusses with Mr. P. the option of stopping eating and drinking. This appeals to him, although he is concerned about what symptoms he might experience. She reassures him that the team would support his decision and that they will continue to ensure he is comfortable. His sister and niece are brought into the conversation, and they give their support to this plan. He dies peacefully four days later.

References American Psychiatric Association. 2013. Diagnostic and statistical manual of mental disorders, 5th edition. Arlington, VA: Author. Appelbaum, P.S. 2007. Assessment of patients’ competence to consent to treatment. New England Journal of Medicine 357: 1834–​1840. Block, S. D. 2000, February 1. Assessing and managing depression in the terminally ill patient. ACP-​ASIM E-​of-​Life Care Consensus Panel. American College of Physicians—​American Society of Internal Medicine. Annals of Internal Medicine 132(3): 209–​218. Bostwick, J. M., & Cohen, L. M. 2009. Differentiating suicide from life-​ending acts and end-​of-​ life decisions: A model based on chronic kidney disease and dialysis. Psychosomatics 50: 1–​7. Breitbart, W., Rosenfeld, B., Gibson, C., Kramer, M., Li, Y, Tomarken, A., … Schuster, M. 2010. Impact of treatment for depression on desire for hastened death in patients with advanced AIDS. Psychosomatics 51: 98–​105. Breitbart W., Rosenfeld B., Pessin H., Kaim M., Funesti-​Esch J., Galietta M., Nelson C. J., Brescia R. 2000. Depression, hopelessness, and desire for hastened death in terminally ill patients with cancer. Journal of the American Medical Association 284(22): 2907–​2911. Brody, B. 2009. Who has capacity? New England Journal of Medicine 361(3): 232–​233. Cairns, R., Maddock, C., Buchanan, A., David, A. S., Hayward, P., Richardson, G., Szmulker, G., & Hotopf, M. 2005. Prevalence and predictors of mental incapacity in psychiatric inpatients. British Journal of Psychiatry 187: 379–​385. Candia, P. C., & Barba, A. C. 2011. Mental capacity and consent to treatment in psychiatric patients: The state of the research. Current Opinion in Psychiatry 24: 422–​426. Cellarius, V. 2008. Terminal sedation and the “imminence condition.” Journal of Medical Ethics 34: 69–​72. Chochinov, H. M., Hack, T., Hassard, T., Kristjanson, L. J., McClement, S., & Harlos, M. 2002. Dignity in the terminally ill: A cross-​sectional, cohort study. Lancet 360: 2026–​2030. Chochinov, H. M., Wilson, K. G., Enns, M., & Lander, S. 1998. Depression, hopelessness, and suicidal ideation in the terminally ill. Psychosomatics 39: 366–​370. de Figueiredo, J. M., & Frank, J. D. 1982. Subjective incompetence, the clinical hallmark of demoralization. Comprehensive Psychiatry 23(4): 353–​363. Facing Death [television broadcast]. 2010, November 10. Frontline. Boston, MA:  Public Broadcasting Service. Ferris, F. D., Balfour, H. M., Bowen, K., Farley, J., Hardwick, M., Lamontagne, C., … West, P. 2002. A model to guide hospice palliative care. Ottawa: Canadian Hospice Palliative Care Association.

Depression and the Desire to Die    405 Ganzini, L., Lee, M. A., Heintz, R. T., Bloom, J. D., & Fern, D. S. 1994. The effect of depression treatment on elderly patients’ preferences for life-​sustaining medical therapy. American Journal of Psychiatry 151: 1631–​1636. Ghormley, C., Basso, M., Candlis, P., & Combs, D. 2011. Neuropsychological impairment corresponds with poor understanding of informed consent disclosures in persons diagnosed with major depression. Psychiatry Research 187: 106–​112. Grisso, T., & Appelbaum, P. S. 1995. The MacArthur Treatment Competence Study: III. Abilities of patients to consent to psychiatric and medical treatments. Law and Human Behavior 19: 149–​174. Grisso, T., & Appelbaum, P.S. 1998. Assessing competence to consent to treatment: a guide for physicians and other health professionals. New York: Oxford University Press. Irwin, S. A., Fairman, N., & Montross, L. 2012. UNIPAC 2: Alleviating psychological and spiritual pain. In C. P. Storey (Ed.), UNIPAC, 4th Ed. Glenville, IL: American Academy of Hospice and Palliative Medicine. Jacobsen, J. C., Zhang, B., Block, S. D., Maciejewski, P. K., & Prigerson H. G. 2010. Distinguishing symptoms of grief and depression in a cohort of advanced cancer patients. Death Studies 34(3): 257–​273. Jansen, L. A. 2003. The moral irrelevance of proximity to death. Journal of Clinical Ethics 14: 49–​58. Kissane, D. W. 2004. The contribution of demoralization to end-​of-​life decision-​making. Hastings Center Report 34(4): 21–​31. Kissane, D. W., Clarke, D. M., & Street, A. F. 2001. Demoralization syndrome—​a relevant psychiatric diagnosis for palliative care. Journal of Palliative Care 17(1): 12–​21. Klein, J. 2012. The long goodbye. Time 11: 18–​25. Leeman, C. P. 1999. Depression and the right to die. General Hospital Psychiatry 21: 112–​115. On Our Own Terms [television broadcast]. 2000, September 10. Boston, MA:  Public Broadcasting Service. Parker, M. 2004. Medicalizing meaning:  Demoralization syndrome and the desire to die. Australia and New Zealand Journal of Psychiatry 38: 765–​773. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1982. Making health care decisions: A report on the ethical and legal implications of informed consent in the patient-​practitioner relationship. Vol. 1. Washington, DC: Government Printing Office. Sessums, L. L., Zembrzuska, H., & Jackson, J. L. 2011. Does this patient have medical decision-​ making capacity? Journal of the American Medical Association 306(4): 420–​427. Slavney, P. R. 1999. Diagnosing demoralization in consultation psychiatry. Psychosomatics 40: 325–​329. Sullivan, M. D., & Youngner, S. J. 1994. Depression, competence, and the right to refuse lifesaving medical treatment. American Journal of Psychiatry 151: 971–​978. Vollmann, J., Bauer, A., Danker-​Hopfe, H., & Helmchen, H. 2003. Competence of mentally ill patients: a comparative empirical study. Psychological Medicine 33: 1463–​1471.

Section IV

T H E E M E RG E N C E OF PA L L IAT I V E C A R E A N D HO SP IC E

Chapter 25

Hospice a nd Palliativ e  C a re  Developments, Differences, and Challenges David Clark

Modern interest in organized arrangements for care at the end of life began in the late 19th century, shaped by the epidemiological transition to chronic illness, the growth of hospitals, the rising incidence of cancer, and the transformation of dying into a “process.” Hospice care has deep historical roots, but in the 20th century it came to signify something quite specific—​a reaching out to those in the terminal phase of illness, often in a manner influenced by religious or wider social commitments and with a goal to engage with the fundamentals of human mortality—​making the departure from life dignified, meaningful, and free from suffering. Viewed from the perspective of the public health system, however, hospice was seen as a limited approach, focused mainly on malignant and end-​stage disease, in some places with low coverage and often unlikely to be capable of meeting population need. Palliative care, once defined, soon developed an uneasy relationship to hospice. The new field had a “model” of care; it sought to transcend diagnostic groups and to engage earlier in the trajectory of illness. It seemed to have the potential for wider impact at the system level and became associated with attempts at “mainstreaming” beyond specialist services. Palliative care espoused holistic and multidisciplinary principles but sometimes ran into difficulties in implementing them “at scale.” In addition, palliative medicine emerged as a recognized specialty and highlighted the need for a close relationship to the profession of medicine and the importance of a scientific basis for interventions, particularly those focused on pain and symptom management. The two approaches of “hospice” and “palliative” care were adopted and adapted in various ways around the world and in some countries existed in quite explicit juxtaposition. While the former enjoys fairly high public recognition in some societies, the latter remains poorly understood beyond those who practice it. In a sense, one remains a social movement while the other is a health-​care specialty. The overlap between them

410   David Clark can be complementary but can also create a certain tension. If hospice is death embracing, palliative care emphasizes quality of life; whereas hospice frequently draws on a social model, for palliative care the model is increasingly medical; if hospice promotes community engagement, palliative care supports professional development; and so on. Bridging the two areas is the more recently styled endeavor known as “end-​of-​life care.” This term continues to lack definitional clarity but has gained significant interest from policymakers and wider publics. As these interrelated fields grow and mature, they generate contestations, not only between themselves but also in relation to other models focused on end-​of-​life issues—​the calls for legalizing assisted dying, for promoting community “resilience” in the face of death, and for promoting choice in end-​of-​life decision making. The 21st century will bring many challenges to these endeavors, not the least due to population growth and aging, which will see unprecedented numbers of people facing death and bereavement. More work is needed on how interventions in the face of serious illness and death should be framed, what we can learn from these that might translate across cultural and resource settings, and how a robust strategy can be developed to respond to the challenges we will face in the coming decades, as the world faces dying and death on a scale that has never before been witnessed.

Enter the Dying By the late 19th century, the people of Europe and North America were living longer and in societies of rapidly increasing size. A transformation of unprecedented proportions had brought industrialization, urbanization, geographic and social mobility, the rise of scientific rationalities, political and ideological upheaval, and a growing questioning of religious values. Mortality rates were falling, and the prospects of living into old age began to increase for many people. But with the benefits of longevity and the diminished threat of early death came other consequences. The predominant causes of death had started to shift —​from sudden demise brought on by infection, disaster, and plague to more protracted dying associated with the emerging chronic diseases of the modern era, not the least cancer and tuberculosis. Popular culture presented idealized images of a slow and controlled farewell to the world, with family members gathered around and confident in a passage to another and better place—​good death as a sign of coming salvation. But by the later years of the 19th century, concerns were emerging about the precise manner of dying—​now coming to be seen not only as a social and cultural event but also as a medical process. The changing personnel around the deathbed, a new secrecy about the imminence of death, as well as the desire to quell the threat of pain and suffering all reveal a growing anxiety about the dying process. That in turn opened up a space for medical intervention that first took hold toward the end of the 19th century but was to have repercussions for more than 100 years thereafter, indeed right up to the present day.

Hospice and Palliative Care    411 Against this background, people referred to as “the dying” began to emerge more clearly as a social category, and, over time, the most common place for life to end began to shift from the domestic home to some form of institution. For the first time in history, special institutions were established, often the work of religious orders or religiously motivated philanthropists who were uniquely concerned with the care of dying people. At first the influence of these “hospices” and “homes for the dying” was quite limited, and they appeared scarcely capable of changing the wider attitudes and practices that were in existence at the time. They emerged in several countries through the later 19th century and into the early decades of the 20th century but seemed to make little impact on the wider environment of care for dying people. In the United States and Europe, they also overlapped in mission and activity with homes for “incurables”—​those suffering from chronic conditions for which the emerging science of modern medicine as yet had little to offer. Nevertheless, these places signified an important transition in which “the dying themselves were repositioned behind the opaque veneer of hospitals, nursing homes and mortuaries” (Wood and Williamson 2003). In the United States, after the Civil War, people increasingly died in hospitals (Corr, Nabe, and Corr 2009), and by the end of the 19th century, “moribund patients were systematically transferred to special rooms” (Rosenberg 1987, 292). The dying were being recognized but were quickly seen as persons to be hidden away and sequestered from public space and in particular from the day-​to-​day organization of the rapidly expanding public hospitals. In 1887 the London physician William Munk published his groundbreaking and well-​ received work on “easeful death,” setting out a case for the skilled and sympathetic care of the dying patient (Munk 1887). Over the next half century, other prominent physicians and surgeons, including Herbert Snow, William Osler, and the American physician Alfred Worcester, wrote eloquently on the same subject. In a series of famous lectures, Worcester (1935) could note “many doctors nowadays, when the death of their patients becomes imminent, seem to believe that it is quite proper to leave the dying in the care of the nurses and the sorrowing relatives. This shifting of responsibility is un-​pardonable. And one of its results is that as less professional interest is taken in such service less is known about it” (33). In the aftermath of World War One, which had seen death and bereavement on an unprecedented quasi-​industrial scale and in a context where modern medicine was seeking to make growing claims for its power to cure and rehabilitate, interest in the old, the terminally ill, and those with chronic conditions was not at the forefront of innovative thinking. A modern culture of dying and bereavement as “taboo” aspects of social life further bolstered this, and the first half of the 20th century showed little progress in knowledge, practice, or policy relating to the care of those at the end of life. In most places the influence of pioneer thinkers was limited, and the care of the dying remained a neglected field of medicine. A mere handful of terminal care homes and hospices, isolated and undeveloped, showed a dogged but largely noninfluential interest in the problem of suffering at the end of life. The dying were in the wings but still remote from center stage.

412   David Clark

The Idea of Hospice From the middle of the 19th century it is possible to identify certain important developments in the hospice care of dying people, several of these led by women. The young widow and bereaved mother Jeanne Garnier, together with others in similar circumstances, formed L’Association des Dames du Calvaire in Lyon, France, in 1842 and opened a home for the dying the following year, characterized by “a respectful familiarity, an attitude of prayer and calm in the face of death” (Clark 2000). Jeanne Garnier died in 1853, but her influence led to the foundation of six other establishments for the care of the dying between 1874 in Paris and 1899 in New York. In both of these cities, modern-​day palliative care services now exist that originate directly from the work of L’Association des Dames du Calvaire. Mary Aikenhead was born in Cork, Ireland, in 1787. At age twenty-​five she became Sister Mary Augustine and was established almost immediately as Superior of a new Order, known as the Irish Sisters of Charity, the first of its kind in Ireland to be uncloistered. The Order made plans to establish a hospital. Three of the sisters went to Paris to learn the work of the Notre Dame de la Pitié Hospital. In Ireland they opened St Vincent’s Hospital, Dublin, in 1834. Following many years of chronic illness, Mary Aikenhead died at nearby Harold’s Cross in 1858. Fulfilling an ambition that she had long held, the convent where Mary Aikenhead spent her final years became Our Lady’s Hospice for the Dying in 1879. The Sisters of Charity followed it with others in Australia, England, and Scotland, all of which still exist today and are run by the Order as modern palliative care units. In the United States, Rose Hawthorne had experienced the death of a child and watched her friend, the poet Emma Lazarus, die of cancer. During the late 1890s she organized a group of women known as the Servants of Relief of Incurable Cancer, and when her husband died she took religious orders in 1900 under the title Mother Alphonsa and formed an order known as the Dominican Sisters of Hawthorne. They established St Rose’s Hospice in Lower Manhattan and then another in New York, followed by others in Philadelphia, Fall River, Atlanta, St. Paul, and Cleveland. These early hospices and homes for the dying reveal three sets of concerns: religious, philanthropic, and moral. They placed a strong emphasis on the cure of the soul, even when the life of the body was diminishing. They drew on charitable endeavors and were often concerned to give succor to the poor and disadvantaged. They were not, however, places in which the medical or nursing care of the dying was of any real sophistication. Rooted in religious and philanthropic concerns, which would diminish as the 20th century advanced, the early homes for the dying represent simply the prologue to a period of subsequent development, which got underway in the decades after World War Two (Humphreys 2001). It was this hospice tradition, however, that was to prove so influential on the thinking of Cicely Saunders, as she trained in medicine in the 1950s in London, having already gained experience as a nurse and a social worker. When she founded St. Christopher’s

Hospice and Palliative Care    413 Hospice in South London in 1967, it quickly became a source of inspiration to others. As the first “modern” hospice, it sought to combine three key principles: excellent clinical care, education, and research. It therefore differed significantly from the other homes for the dying that had preceded it and sought to establish itself as a center of excellence in a new field of care. Its success was phenomenal, and it soon became the stimulus for an expansive phase of hospice development not only in Britain, but also around the world. Modern hospice developments occurred first in affluent countries, but in time they also took hold in poorer settings, often supported by mentoring and partnership arrangements with more established hospices in the West. In the United Kingdom there was an initial emphasis on voluntary inpatient hospices, many of which had Christian origins. Echoes of this model can also been seen elsewhere but were almost always nuanced to the local context and conditions. For example, in Germany, the influences of religion and civil society were strong, and the early hospice services were more likely to be community based with limited medical involvement. In India, some hospices were opened but seemed ill suited to local conditions, where huge population need required a different, more public health oriented approach—​such as that subsequently developed by the Neighbourhood Networks in Palliative Care in Kerala. In Romania the first hospice had many of the attributes of its British equivalents but was adapted to a culture where voluntary engagement was relatively unfamiliar and where religious traditions had been suppressed under a communist dictatorship. In the United States, hospices focused on home care programs and made significant gains from 1982 with new legislation that provided reimbursement through Medicare. In many settings, hospice developments were closely allied to civil society, notions of service, social reform, and even political change. UK developments built on a tradition of local involvement in delivering services with voluntary input. In the United States, hospice activists rode a wave of social ferment associated with civil rights, feminism, and the politics of the personal. In Eastern Europe hospices quickly emerged after the collapse of communist systems and as part of a new enthusiasm for social action and public engagement; indeed, in Poland, hospice activists were among those who were instrumental in bringing the about the collapse of the old regime through the Solidarity movement. By the late 1980s, new services were appearing in many countries, and, in some, such as the United Kingdom and the United States, hospices were becoming an established part of a mixed economy of care. Significantly, the idea of hospice—​with its emphasis on holistic approaches that included physical, spiritual, psychological, and social care and its attention to the needs of patient and family—​was attracting growing interest well beyond a core circle of activists and enthusiasts. Teaching programs emerged in colleges and universities, tentative steps were taken toward establishing research-​based evidence in support of further development, specialist journals as well as journalistic and more popular writings found a growing readership, and policymakers started to factor “hospice” into their thinking and planning.

414   David Clark

Defining Palliative Care The development of hospices, thus described, masks some other important factors. By the mid-​20th century, specialization in medicine and health care was advancing rapidly. From oncology to cardiology, geriatrics, psychiatry, and paediatrics—​as well as many other areas—​new fields of specialization were opening up with defined boundaries, internal status hierarchies, esoteric claims to knowledge, and specific skills and technologies. It would be important that the growing interest in hospice should articulate with these developments. To survive and thrive, hospice would need to speak the language of the modern medical establishment—​and that would include fluency in assessing outcomes, modelling interventions against best evidence, and some acumen in assessing the costs and benefits of particular approaches. While “hospice” often captured public interest and attention, it played less well with medical culture, could be seen as overly attached to charitable or religious perspectives, and ran the risk of being dismissed or marginalized as the preserve of an enthusiastic but ultimately slightly anachronistic few. In particular there was the criticism that it was “too good to be true and too small to be useful” (Douglas 1992, 479). There was a need for an approach and way of thinking that would translate better into health system contexts and that would engage more effectively with professional agendas and public health orientations. Inspired by the thinking of Canadian surgeon Balfour Mount (1997), who embraced the St. Christopher’s philosophy but sought to adapt it to the environment of the Royal Victoria Hospital in Montreal, the term palliative care came into more common use. Here he was taking a cue from the established use of the word “palliative” in medical practice, meaning to treat for goals other than cure, as well as the Latin palliare, meaning to cloak or disguise. This was not merely semantic. The new term brought some important innovations, not the least of which was wider settings for practice and an expanded clinical ambition. Within a decade of the opening of St. Christopher’s Hospice, some were seeking to establish that the principles of hospice care could be practiced in many contexts, such as specialist inpatient units as well as in home-​care and day-​care services. Likewise, hospital units and support teams were established that engaged with the challenge of bringing the new thinking about dying into the very heartlands of acute medicine—​across the various settings of the acute teaching hospitals. As these services expanded, the focus of intervention also widened from care of the actively dying to those earlier in their disease progression with symptom problems, with an increasing focus on care management, quality of life, better coordination of care, and a desire to match patients’ and families’ needs and preference to actual experiences. As hospice and palliative care developed, there was growing interest in establishing it as a specialty. In the United Kingdom in 1987, palliative medicine became a subspecialty of general medicine, initially on a seven-​year “novitiate,” which, once successfully concluded, led to the creation of a specialty in its own right (Doyle 2005). Over the next twenty-​five years some twenty more countries would variously recognize palliative medicine as a full or (more commonly)

Hospice and Palliative Care    415 subspecialty, complete with recognized training programs and centers. As the field emerged in this way—​indeed as a prerequisite for its emergence—​there was a need for greater clarity of definition and scope for this new form of activity, as well as greater precision about what it could achieve and where its boundaries might begin and end. The World Health Organization (WHO; 1990) first defined the term “palliative care” in 1989. It emphasized the active, total care of the patient, focusing on the control of pain and other symptoms as well as providing psychological, social, and spiritual support to improve the quality of life and to affirm life, while regarding dying as a “normal process” and seeking neither to hasten nor postpone death. The most recent definition of palliative care to appear from the WHO (Sepúlveda et al. 2002), states that palliative care is an approach that improves the quality of life of patients and their families facing the problems associated with life-​threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial, and spiritual. Palliative care has become more expansive in its goals and has sought to move its influence “upstream” to earlier stages in the trajectory of illness. In doing so, it encounters definitional and boundary problems in relation to “supportive care” provided by oncologists and (in the United Kingdom) the concept of acute oncology, with its emphasis on chemotherapy-​related side effects, as well as such problems as hypercalcaemia, spinal cord compression, and superior venocaval obstruction. At the same time there is growing interest in end-​of-​ life care, viewed more broadly in relation to people with a wide range of conditions and diagnoses. In 2004, a National Institutes of Health State-​of-​the-​Science Conference Statement noted that “There is no exact definition of end of life; however, research supports the following components: 1) the presence of a chronic disease(s) or symptoms or functional impairments that persist but may also fluctuate; and 2)  the symptoms or impairments resulting from the underlying irreversible disease that require formal either paid, professional or informal unpaid or volunteer care and can lead to death” (National Institutes of Health 2004). There are similar problems associated with the definitions of assisted dying; some use the term to describe procedures that might otherwise be described as (physician)-​assisted suicide or voluntary euthanasia, arguing that “although mechanically they differ—​at the heart of each of them is the positive choice for death” (McLean 2007). End-​of-​life care may or may not include palliative care or assisted dying. More provocatively, a further issue is whether or not (as in Belgium) palliative care and legalized euthanasia can sit side by side within the same model of care or service delivery in an “integral” fashion. It is within this nexus of issues that considerable debate is now taking place. It has particularly challenging dimensions in the United States where a so-​called reframing from end-​of-​life to palliative care means focusing on the concern of patients and families that physicians will abandon them if hospice care is recommended. Thus it is observed: “The notion of a ‘good death,’ which hospice effectively enables for many people, has greatest salience after a loved one has died. Most people and families facing a serious illness do not want a good death; they want a cure. The frame of palliative care allow[s]‌patients, families and physicians to accept care and amelioration of pain while they pursue active

416   David Clark treatment” (Patrizi, Thompson, and Spector 2011, 27). Palliative care, seen in this way, is not something that commences when all else has failed but is rather an integrated component of medical mainstream medical practice, to be pursued legitimately alongside other clinical goals. Such an approach has in turn led to a new definition of palliative care, which, strikingly, makes no reference to dying, death, or bereavement: Palliative care is specialized medical care for people with serious illnesses. This type of care is focused on providing patients with relief from the symptoms, pain, and stress of a serious illness—​whatever the diagnosis. The goal is to improve quality of life for both the patient and the family. Palliative care is provided by a team of doctors, nurses, and other specialists who work with a patient’s other doctors to provide an extra layer of support. Palliative care is appropriate at any age and at any stage in a serious illness, and can be provided together with curative treatment. (Center to Advance Palliative Care. 2011)

This latter approach may be described as a more consumerist orientation to the delivery of palliative care. It is based on the assertion that the model of care must follow the demands and orientations of the patients and users. It argues that delivering palliative care to achieve better symptom control, improved communication, and greater alignment with patient and family wishes are primary goals and that only once these are achieved is it appropriate to engage with questions of meaning, mortality, and the reality of death. This seems a departure from the Saunders’ model that has been widely adopted by hospices and in which a central focus is on finding personal and/​or religious closure at the end of life and in which a wider social goal is also being pursued, in addition to the specific task of providing care to patients and families: that of making society better. Inspired by charismatic leadership, hospice was a movement that condemned the neglect of the dying in society; called for high-​quality pain and symptom management for all who needed it; sought to reconstruct death as a natural phenomenon, rather than a clinical failure; and marshalled practical and moral argument to oppose those in favor of euthanasia. Indeed, for Cicely Saunders (1962) and her followers, such work served as a measure of the very worth of our culture: “A society which shuns the dying must have an incomplete philosophy.”

International Patterns Whether seen through the trope of hospice or palliative acre, end-​of-​life issues will have major importance in the 21st century as the population ages and as individual patterns of disease, symptoms, and disability in later life become more complex. Global projections indicate that the number of people aged over 80 worldwide will increase from 102 million in 2009 to 395 million in 2050 (United Nations Department of Economic and Social Affairs/​Population Division 2009). Recent estimates suggest that 91 million

Hospice and Palliative Care    417 people will die worldwide in 2050, compared to 56 million in 2009 (Gomes et al. 2011). It is also estimated that for each death another five persons will be affected by care-​giving and grieving (Stjernsward and Clark 2003), so by the second half of the century some half a billion people each year will experience the death of a person close to them. How will death and dying on this scale be supported, and what recognition is being given to the challenges ahead? Around the world, various propositions are emerging in relation to this major societal and public health challenge. It is important to recognize, however, that palliative care is only one such response. Since 2000, a number of key developments have been taking place in end-​of-​life care globally, including a series of important summit meetings on international palliative care development and the creation of the Worldwide Palliative Care Alliance (http://​www.thewpca.org/​). In autumn 2011, two separate declarations emphasized the importance of palliative and end-​of-​life care. In the first, the United Nations made reference to the need for palliative care provision in its statement about the care and treatment of people with noncommunicable disease (General Assembly of the United Nations 2011). In the second, the World Medical Association (2011) made its case for end-​of-​life care improvement, stating that receiving appropriate end-​of-​life medical care must not be considered a privilege but a true right, independent of age or any other associated factors. At the same time, calls for the legalization of assisted dying and euthanasia have been growing in several jurisdictions, and public debate for and against continues to flourish, often stimulated by high-​profile examples of key cases and endorsement by leading opinion formers. Also situated in this space are advocates for “natural death,” “green burial,” and those who recognize that dying on a scale that the world now faces will have significant environmental consequences. How we should die—​and what should happen to our bodily remains—​is therefore becoming a topic of massive public interest as population trends point to a huge increase in the quantum of death and bereavement that will occur in society and as alternative and sometimes conflicting solutions are canvassed and implemented. A series of international studies conducted by myself and other colleagues led in 2006 to a “world map” showing levels of palliative care development for every country in the world, using a four-​part typology (Wright et al. 2008). It showed that only half of the world’s countries had some form of designated palliative care service. In 2011, the analysis was updated with a refined, six-​point typology (Lynch, Connor, and Clark 2013). At that point 136 of the world’s 234 countries (58%) had one or more hospice-​palliative care services established—​an increase of 21 countries (+9%) from 2006. Although by 2011 there were indications of growing interest in palliative care on the part of national governments and policymakers, advanced integration of palliative care with wider health services (the highest category in the typology) had been achieved in only 20 countries globally (8.5%). A 2010 study by the Economist Intelligence Unit (2010) attempted a ranking of palliative care development in 40 countries of the world with a more complex set of indicators. The Quality of Death Index scored on 24 indicators in four categories, each with a separate weighting: (a) basic end-​of-​life health-​care environment (20%) (b) availability of end-​of-​life care (25%) (c) cost of end-​of-​life care (15%) and (d) quality of end of life

418   David Clark (40%). The top five countries in the ranking were the United Kingdom, Australia, New Zealand, Ireland, and Belgium. The five countries at the bottom of the ranking were Mexico, China, Brazil, Uganda, and India, a list that includes three of the top five countries in the world by population (in italics). There is no doubt that these mapping and ranking studies have made a contribution to our understanding of how palliative care and end-​of-​life care are developing in the international context. At the same time, the studies have struggled to attain comparability between settings and to generate analyses of the relative quality of services. They have also failed adequately to take account of variations between health-​care systems in different jurisdictions and have viewed palliative care development as an independent variable in relation to the underlying policy and service delivery context, nor have they documented conflicting debates on end of life issues—​thus they reveal little about the influence of voluntary assisted dying, where it has been legalized or where it is proposed. Most significantly, these mapping studies have had nothing to say about the cultures of death, dying, and bereavement in which they are located. Floating free from such sociological and anthropological concerns, they have produced a one-​dimensional analysis of specific palliative care service development and used this as a tool to measure the progress of the specialism in the global context—​thereby suggesting the questionable view that improved palliative care is both the necessary and the sufficient response to the challenge of end-​of-​life care around the world. It is not uncommon to come across exhortations to promote care at the end of life that is “culturally appropriate” or even “culturally sensitive” and that draws on “indigenous knowledge” (Crawley 2002). If we understand culture to consist of the beliefs, values, behaviors, and dispositions that exist among groups in society and also to include material dimensions—​in objects, manufactured things, and practical inventions—​then the relationship to end-​of-​life care becomes clearer. Although palliative care often makes universal and global cultural claims on behalf of all people with life-​limiting conditions, there has long been a recognition that it must be rooted in specific cultural contexts if it is to thrive and be effective. There are two dimensions to this. First, there is the particular recognition that patients and families “have a culture.” Working with and understanding this culture is seen as key to successful care giving. Recently there have been efforts to suggest that our understanding of culture, seen in this way, should be nuanced and sensitive—​and alert to internal heterogeneities. Members of a cultural group may not share the same beliefs about a given phenomenon—​for example, as to whether a dying person should be told of his or her condition (Cherny 2012) or whether the desire for freedom from pain is a universal goal (Chen, Tang, and Chen 2012). This seems to mark a step away from a previous approach, developed in some countries, that involved the use of “checklists” to guide practitioners encountering dying patients and their families from specific religious, ethnic, or cultural groups—​an approach that led to criticisms from some social scientists (Gunaratnum 2007). But what does it mean to say that cultural dimensions of end-​of-​life care will be dealt with in a “nuanced” way? Who will hold such cultural competence, and what possibility is there of it being enacted?

Hospice and Palliative Care    419 Second, there is the realization that palliative and end-​of-​life care systems and services and those who deliver them are also possessed of “culture.” Such care delivers a culture as part of its activity—​it perpetuates and maintains the values and beliefs of those who work within it, those who support it, and those who contribute to its development. Seen in this way, palliative care teaching, research, advocacy, fundraising, publications, presentations, conferences, and seminars are all cultural manifestations. In the future as resources shrink and as demand for end-​of-​life care increases, it seems likely that cultural factors are going to play an even more significant part in shaping provision around the world—​just as significant as policies, professional ideologies, and service configurations. There is a need to take culture extremely seriously as an aspect of understanding contemporary issues in end-​of-​life care delivery, and yet cultural matters are too often marginal to palliative care thinking, research, and planning. Despite the progress made by advocates of palliative and end-​of-​life care services, the barriers to development remain obdurate and are widely articulated in the literature (Graham, Kumar, and Clark 2010). A lack of public recognition and understanding, professional indifference, restricted third-​party funding, and weakly developed training programs are all widely cited. Of special note is the limited evidence base about the efficacy of palliative care and its costs and benefits, and this continues to be seen as an impediment to expansion as well as an explanation for initiatives that fail; the case of the Liverpool Care Pathway is a striking example of the latter (Sleeman 2013). Meanwhile end-​of-​life care remains poorly framed within the architecture of national and international policymaking. It is not one of the Millennium Development Goals; it is hard to find in the priorities of the United Nations, UNESCO, and other global organizations; it has only the most tentative recognition in the frameworks of the World Bank, the Global Fund, and the concerns of the world’s largest philanthropic donors and foundations. Indeed, it is not mentioned at all in the 2012 strategic plan for WHO Europe—​even though the 53 countries in the European region have the highest median age in the world and include 9 of the 10 countries with the longest life expectancy (WHO Regional Office for Europe 2012). This policy issue needs to be better understood. Why do other international health priorities gain more attention? Is end-​of-​life care perceived to be a low priority when resources are scarce? Are demands for the legalization of voluntary assisted dying and euthanasia having more influence than the call for increased palliative care? Should the two perspectives work together more closely to gain greater leverage? What are the implications of the “contested” status of dying in the modern world—​for policy, for practice, and for research? These are important questions for further study and go well beyond our current body of knowledge. There is a sense, however, in which the wrong questions are being asked within much of the palliative and end-​of-​life care research endeavor. In some areas of such research, there is a discernible and problematic trend toward studies that are dominated by individual and clinical perspectives and shaped by the perspectives of biomedical and health services research. This is producing a growing body of small-​scale descriptive studies, clinical trials, systematic reviews, and “evidence-​based” guidelines that frame the problem of end-​ of-​life care in particular ways and concentrate on demonstrating the efficacy of specific

420   David Clark interventions. It has led to the recent claim that “there is now acceptable evidence of the effectiveness of specialist palliative and hospice care in improving pain and symptom control and psychological well-​being for patients and families” (Higginson 2011). Typical dimensions of this approach are an emphasis on problems for medical management, the characterization of different syndromes and clusters of pain and other symptoms, aspirations of clinical benefit arising from knowledge about genetic origins of response to available drug treatments, as well as measurement of the relative impact of differing therapeutic regimes on “outcomes” of care. A smaller body of work is focused on the economic costs and benefits of palliative care—​for the health-​care system as well as for patients and families. Overall, there is a somewhat simplistic view that once clinical and economic “evidence” has been uncovered in this way, then policies for end-​of-​life care can be generated by experts and “rolled out” unproblematically across entire jurisdictions. More positively, recent years have seen growing interest in the twin ideas that palliative care is both a public health and a human rights issue. One assumes the insertion of palliative care into the public health system, thereby positioning it within a framework of “need,” “supply,” and “resource allocation.” Several “public health” palliative care demonstration projects are now in operation—​in Africa and India as well as in the wealthier nations. The other approach seeks recognition for palliative care within human rights legislation (Harding 2006)—​for example, by incorporating access to palliative medication into a resolution of the United Nations Office of the High Commissioner for Human Rights (2004). Item 7c calls on states “to promote effective access to such preventive, curative or palliative pharmaceutical products or medical technologies.” These two approaches are clearly related and suggest considerable potential for development. But they emerge at a time when major questions are being asked about the value of high-​level legislation about human rights, which are still easily flaunted by many governments and which appear to do little to support the efforts of activists on the ground (Hopgood 2013). Much more needs to be done, not only to demonstrate the role and efficacy of palliative care in relieving physical and psychological distress but also to promote community cohesion and personal resilience. A focus on “health-​promoting palliative care” and the creation of “compassionate communities” has also emerged, drawing on the discourses of the new public health (Kellehear 1999). This is leading to significant take-​up and engagement by civil society organizations and policy activists; so far the level of interest is encouraging, but the impacts and outcomes remain unclear.

The Contested Space of Hospice and Palliative Care Specialty recognition constituted a major milestone in the development of hospice and palliative care and continues to be taken up in several countries, although progress has been rather slow. Specialization signaled a wider shift away from terminal and hospice

Hospice and Palliative Care    421 care toward the broader concept of palliative care. Many claim that palliative care specialization, integration into the mainstream health system, and the development of an evidence-​based model of practice and organization are crucial to long-​term viability. Others mourn the loss of early ideals associated with hospice and regret an emphasis on physical symptoms at the expense of psychosocial and spiritual concerns. In short, there have been claims that medicalization and reutilization (James and Field 1992) are at work or even that the putative “holism” of palliative care philosophy masks a new, more subtle form of surveillance of the dying and bereaved in modern society (Clark and Seymour 1999). Interestingly, some of these concerns have come from within medicine and the new specialty itself. As early as 1992, concerns were raised about the danger of palliative care physicians becoming “symptomatologists,” losing sight of the key question that the author saw as “symptom control—​for what?” Likewise, questions were raised about whether palliative medicine is really specialist territory and not more properly the domain of the generalist. Why had there been so little discussion on why specialization in palliative medicine came about, whether it is the most appropriate way to address acknowledged deficiencies in care, and whether it can be sustained in the long term? Examining the factors that have contributed to the evolution of palliative medicine as a specialty was: “its future is in doubt” (Fordham, Dowrick, and May 1998). Definitional problems raise further concerns. A 2008 review by Pastrana and colleagues focused on definitions of the term “palliative medicine” and “palliative care” in two languages found a total of 37 English and 26 German versions, confirming “lack of a consistent meaning” (Pastrana et al 2008) associated with the field. At the same time, there are surprisingly few useful definitions of “end-​of-​life care,” and a 2012 volume offering an international public health perspective on the subject offers no clarification, often using “end-​of-​life care” interchangeably with “palliative care” (Cohen and Deliens 2012). “Assisted dying” and “euthanasia” are also terms that have proved notoriously difficult to define and are subject to multiple interpretations (Materstvedt et al. 2003). The issue here, however, is not so much the question of definition as how these various positions relate to one another and contribute to the “deathways” of any given culture or jurisdiction. Interestingly, in the United States, the full title of the specialty is “hospice and palliative medicine” and the major related professional body is the American Academy of Hospice and Palliative Medicine. This provides a broad church for physicians of differing clinical orientations and roles to join together professionally, but it may suggest greater consensus than actually exists between two fields that are increasingly more developed. There are indications that some end-​of-​life care proposals and interventions are meeting with negative reactions from public and professional groups. This can apply to palliative care programs (sometimes seen as a form of health-​care rationing or as “back-​ door” euthanasia) as well as to more overt proposals for the legalization of assisted dying (e.g., resisted by disability and religious groups). The provision of appropriate end-​of-​life care interventions at the population level is therefore challenged not only by available resources and expertise but also by conflicting claims, value positions, and viewpoints. Is this an expected and perhaps necessary juncture in the evolution of a fledgling field?

422   David Clark Or does it mask some deeper contradictions that will inhibit the development of appropriate care for those facing serious and life-​shortening illness? There is a need for a new approach to research and scholarship on these issues. Despite the complexity and importance of this terrain, recent social and cultural research on end-​of-​life issues has been either too remote to be relevant to policy and practice or too embedded in narrow areas of inquiry to stimulate innovation. Meanwhile, clinical research can appear bogged down in a triumph of methodological preoccupations over substantive findings (the ongoing vogue for systematic reviews is a good example). The field requires more interdisciplinarity, an opening up to new theoretical perspectives and methodologies, and better engagement with the global challenges of end-​of-​life care in the 21st century. This approach forces us to go beyond end-​of-​life interventions that involve the delivery of clinical care. We need to develop a fuller understanding of a broad range of legal, policy, philanthropic, and development-​based interventions for end-​of-​ life provision. Many examples of such are now available to us in the global context—​but we lack a taxonomy and an analysis of their goals, achievements, and impacts. It is just this complex array of actions that currently comprise the overall response to hospice, palliative care, and end-​of-​life needs at a time of huge global challenge and complexity. The care of the dying is finally moving to the center of global thinking and discourse. A paradigm shift in our approach is badly needed if we are to produce future interventions that are robust, acceptable, and sustainable.

References Center to Advance Palliative Care. 2011. 2011 Public Opinion Research on Palliative Care. New  York:  Author. http://​www.capc.org/​tools-​for-​palliative-​care-​programs/​marketing/​ public-​opinion-​research/​2011-​public-​opinion-​research-​on-​palliative-​care.pdf; accessed December 1, 2013. Chen, C. H., S. T. Tang, and C. H. Chen. 2012. Meta-​analysis of cultural differences in Western and Asian patient-​perceived barriers to managing cancer pain. Palliative Med 26(3):206–​221. Cherny, N. 2012. Controversies in oncologist-​patient communication: A nuanced approach to autonomy, culture and paternalism. Oncology 26(1):37–​41. Clark, D. 2000. Palliative care history: A ritual process. Eur J Palliative Care 7(2):50–​55. Clark, D., and J. Seymour. 1999. Reflections on palliative care: Sociological and policy perspectives. Buckingham: Open University Press. Cohen, J., and L. Deliens, eds. 2012. A public health perspective on end of life care. Oxford: Oxford University Press. Corr, C. A., C. M. Nabe, and D. M. Corr. 2009. Death and dying, life and living, 6th ed. Belmont: Wadsorth. Crawley, L. P. A. Marshall, B. Lo, and B. A. Koenig. 2002. Strategies for culturally effective end-​ of-​life care. Annals of Internal Medicine 136(9):673–​679. Douglas, C. 1992. “For all the saints.” British Medical Journal 304(6826):479. Doyle, D. 2005. Palliative medicine: The first 18 years of a new sub-​specialty of general medicine. The Journal of the Royal College of Physicians of Edinburgh 35:199–​205.

Hospice and Palliative Care    423 Economist Intelligence Unit. 2010. The quality of death:  Ranking end-​of-​life care across the world. New  York:  Author. http://​www.eiu.com/​site_​info.asp?info_​name=qualityofdeath_​ lienfoundation&page=noads; accessed December 1, 2013. Fordham, S., C. Dowrick, and C. May. 1998. Palliative medicine: Is it really specialist territory? Journal of Royal Society of Medicine 91:568-​572. General Assembly of the United Nations. 2011. High-​level meeting on non-​communicable diseases. New  York:  Author. http://​www.un.org/​en/​ga/​president/​65/​issues/​ncdiseases.shtml, accessed December 1, 2013. Gomes, B., J. Cohen, L. Deliens, and I. Higginson. 2011. International trends in circumstances of death and dying amongst older people. In Living with ageing and dying: Palliative and end of life care for older people, edited by M. Gott and C. Ingleton. Oxford: Oxford University Press. Graham, F., and D. Clark, D. 2008. Barriers to cancer pain relief: An international perspective on drug availability and service delivery. In Cancer pain, 2d ed., edited by N. Sykes, M. Bennett, and C. S. Yuan. London: Hodder Arnold, 81–​92. Graham, F., S. Kumar, and D. Clark. 2010. Barriers to the delivery of palliative care. In Oxford textbook of palliative medicine, 4th ed., edited by G. Hanks et al. Oxford: Oxford University Press, 125–​134. Gunaratnum, Y. 2007. Intercultural palliative care:  Do we need cultural competence? International Journal of Palliative Nursing 13(10):470–​477. Harding R. 2006. Palliative care: A basic human right. id21 Insights Health 7(3):1–​2. Higginson, I. 2011. The global palliative care lottery. Can we overcome it? Journal of Palliative Medicine 14(4):384–​385. Hopgood, S. 2013. The endtimes of human rights. New York: Cornell University Press. Humphreys, C. 2001. “Waiting for the last summons”: The establishment of the first hospices in England 1878–​1914. Mortality 6(2):146–​166. James, N., and D. Field. 1992. The routinization of hospice: Charisma and bureaucratisation, Social Science and Medicine 34(12):1363–​1375. Kellehear, A. 1999. Health promoting palliative care. Oxford: Oxford University Press. Kellehear, A. 2005. Compassionate cities: Public health and end of life care. London: Routledge. Lynch, T., S. Connor, and D. Clark. 2013. Mapping levels of palliative care development:  A  global update. Journal of Pain and Symptom Management 45(6):1094–​1106. doi: 10.1016/​j.jpainsymman.2012.05.011 Materstvedt, L. J., D. Clark, J. Ellershaw R. Førde, A.-​M. Boeck Gravgaard, H. C. Müller-​Busch, J. Porta i Sales, and C.-​H. Rapin. 2003. Euthanasia and physician-​assisted suicide: A view from an EAPC Ethics Task Force. Palliative Medicine 17:97–​101. McLean, S. 2007. Assisted dying: Reflections on the need for law reform. Abingdon: Routledge-​ Cavendish. Mount, B. 1997. The Royal Victoria Hospital Palliative Care Service:  A  Canadian experience. In Hospice care on the international scene, edited by C. Saunders and R. Kastenbaum. New York: Springer. Munk, W. 1887. Euthanasia: or, medical treatment in aid of an easy death. London: Longmans, Green and Co. National Institutes of Health. 2004. National Institutes of Health State-​ of-​ the-​ Science Conference: Statement on improving end-​of-​life care. December 6–​8. Pastrana, T., et al. 2008. A matter of definition—​Key elements identified in a discourse analysis of definitions of palliative care. Palliative Medicine 22:222–​232.

424   David Clark Patrizi, P., E. Thompson, and A. Spector. 2011. Improving care at the end of life: How the Robert Wood Johnson Foundation and its grantees built the field. Princeton, NJ:  Robert Wood Johnson Foundation. Rosenberg, C. E. 1987. The care of strangers:  The rise of America’s hospital system. New York: Basic Books. Saunders, C. 1962. “And from sudden death … ” Nursing Times, August 17, 1045–​1046. Sepúlveda C, A. Marlin, T. Yoshida, and A. Ullrich. 2002. Palliative care: The World Health Organization’s global perspective. Journal of Pain and Symptom Management 24(2):91–​96. Sleeman, K. 2013. The Liverpool care pathway:  A  cautionary tale. British Medical Journal 347:f4779. Stjernsward, J., and D. Clark. 2003. Palliative medicine—​A global perspective. In Oxford textbook of palliative medicine, 3d ed., edited by D. Doyle, G. W. C. Hanks, N. Cherny, and K. C. Calman. Oxford: Oxford University Press, 1199–​1224. United Nations Department of Economic and Social Affairs/​Population Division. 2009. World population ageing 2009. Report No. ESA/​P/​WP/​212. New York: Author. United Nations Office of the High Commissioner for Human Rights. Commission on Human Rights Resolution 2004/​26:  New  York:  Author. http://​www.ohchr.org/​english/​, accessed December 1, 2013. Wood, W. R., and J. B. Williamson. 2003. Historical changes in the meaning of death in the western tradition. In Handbook of death and dying, Vol. 1, edited by C. D. Bryant Thousand Oaks, CA: SAGE. Worcester, A. 1935. The care of the aged, the dying and the dead. Springfield, IL:  Charles C Thomas. World Health Organization (WHO). 1990. Cancer pain relief and palliative care. WHO Technical Report Series 804. Geneva: Author. World Health Organization (WHO) Regional Office for Europe. 2012. Health 20-​20: A European policy framework supporting action across government and society for health and wellbeing. Copenhagen: Author. World Medical Association. 2011. WMA declaration on end-​of-​life medical care. New York: Author. http://​www.wma.net/​en/​30publications/​10policies/​e18/​index.html?World%20Medical%20 Association%20declaration%20on%20end%20of%20life%20medical%20care; accessed December 1, 2013. Wright, M., J. Wood, T. Lynch, and D. Clark. 2008. Mapping levels of palliative care development: A global view. The Journal of Pain and Symptom Management 35(5):469–​485.

Chapter 26

P ot ential Peri l s to t h e Prom ise of Spe c ia lt y Palliativ e  C a re Robert M. Arnold

Reading the recent Institute of Medicine report “Dying in America” was, for me, a disheartening experience: most people who are dying in the United States today are still suffering in almost the same ways as did people who died twenty years ago (Connors et al. 1995; Institute of Medicine 2014). I was disheartened but not surprised because I have come to believe that there are several reasons why, despite the remarkable growth of specialty palliative care (SPC), we are failing to improve the care of most people living with and dying from serious illness. As one who has devoted his career to helping build the field of SPC, seeing the lack of progress, trying to understand what happened, and realizing that SPC may be part of the problem is painful. But, for the sake of people still dying alone, in pain and suffering, I believe I need to share my viewpoint. Twenty years ago, the largest study of end-​of-​life care, SUPPORT, was published. The study randomized more than 4,000 patients with a projected mortality of 50% at six months at five U.S. hospitals. The randomized controlled trial compared usual care to an intervention consisting of prognostic information, information about patient and family care preferences, and access to a nurse interventionist to facilitate communication. The study found that these interventions made no difference in patients’ pain, early documentation of do not resuscitate (DNR) orders, days in the intensive care unit (ICU) pre-​death, or hospital resource use (Connors et al. 1995). More disturbingly, in more than 1,000 peer-​reviewed citations, SUPPORT documented that pain and other symptoms were seriously undertreated, that conversations about goals of care were uncommon, and that caregivers suffered both psychologically and financially from their loved one’s illness (Covinsky 2010). The negative results of SUPPORT galvanized efforts to improve end-​of-​life care. The Robert Wood Johnson Foundation, the American Medical Association, and George Soros’s Open Society Foundation all developed ambitious programs designed to improve the care of dying patients (Covinsky 2010).

426   Robert M. Arnold Against this backdrop, the growth of SPC within the United States over the past twenty years is amazing. The American Board of Medical Specialties recognized a new medical specialty, palliative care. New fellowship programs have been starting at major academic programs across the country; similarly, nursing schools have been starting specialized programs in palliative nursing, and the clinical need for SPC has grown exponentially (Wikipedia n.d.) The Institute of Medicine has convened two reports focusing on death and dying that called for increased palliative care, and the American Society of Clinical Oncology has issued a provisional clinical opinion recommending that SPC be offered along with concurrent oncological treatment for patients with metastatic cancer (Field 1997; Foley and Gelband 2001; Smith et al. 2011). In the past few years, a number of popular books on death and dying have been released and made it to the New York Times best-​seller list, and a palliative care physician received a MacArthur “genius” award for her work publicizing and promoting SPC in hospitals across the country (Gawande 2014; MacArthur Foundation 2008). And yet, for most seriously ill patients, little has changed. The experience of the average seriously ill or dying patient seems to be about the same as it was twenty years ago (Singer et al. 2015; Teno et al. 2013). Some have made the argument that the solution is to further increase the number of SPC programs. I will argue that this is a mistake. In fact, I believe the rapid growth of SPC merely illustrates the limitations of American medicine in the early twenty-​first century. In particular, the growth of SPC may distract attention from solutions that are more likely to improve care for most seriously ill patients. My argument requires distinguishing between generalist and specialty palliative care. Generalist palliative care (GPC) is the care of seriously ill patients provided by any physician (Quill and Abernethy 2013); that is, the care that each medical specialty (oncology, cardiology, geriatrics, nephrology, etc.) should provide to seriously ill patients. Good GPC would mean that a clinician can treat the common medical and psychological symptoms that seriously ill patients have, discuss advance directives and goals of care, inquire about spiritual and existential issues, and appropriately refer patients to hospice. Finally, a generalist should know when to refer a patient to a specialist in palliative care when the patient is not improving or when the issues are too complex for the generalist to handle. SPC is provided by clinicians who have had extra training in palliative care, spend a significant part of their job providing palliative care consultations, and serve as an extra level of support for other clinicians. The literature on palliative care typically confuses GPC and SPC. Sometimes, when authors call for more palliative care, they are urging more SPC consultations. For example, this is the focus of ASCO’s provisional clinical opinion: Based on strong evidence from a phase III RCT, patients with metastatic non–​small-​ cell lung cancer should be offered concurrent palliative care and standard oncologic care at initial diagnosis. While a survival benefit from early involvement of palliative care has not yet been demonstrated in other oncology settings, substantial evidence demonstrates that palliative care—​when combined with standard cancer care or as the main focus of care—​leads to better patient and caregiver outcomes. (ASCO Institute for Quality 2012).

Specialty Palliative Care   427 Other times, when people are calling for more palliative care what they mean is more attention to a set of concerns that are particularly important to patients who have serious life-​limiting illnesses. For example, when Brennan calls for palliative care as a human right (Brennan 2007), he is arguing for more attention to pain control and symptom management and more attention to communication about goals of care and advance directives rather than more SPC consultations. This causes confusion because one is never sure whether one is talking about the specialty or the general domains of care that are important for the care of all seriously ill patients. This ambiguity causes confusion as to whether the solution is more specialty consultations or better training of all health care providers. I am going to argue that, at this time in history, what is needed is more GPC. In fact, the rapid growth of SPC can have negative consequences for GPC. And if we are going to do a better job caring for seriously ill patients, we have to be alerted to how SPC can (1) allow the health care system to continue to deny death and (2) “ghetto-​ize” and de-​skill GPC.

The Problem: SPC Grows While Dying Stays the Same Morrison et al. do a nice job reviewing the history of hospice, the start of palliative care, and how it has grown over the past twenty-​five years. The development of SPC consultative services has been justified by a growing literature showing that these consultative services improve clinical care, enhance patient and family satisfaction, and lower health care costs. One study, published in the New England Journal of Medicine, even showed that palliative care/​oncologist co-​management of patients with metastatic lung cancer improved survival (Temel et  al. 2010). Similar articles showing improved care when patients were seen by SPC were published in Lancet, the Journal of the American Medical Association, and the Journal of Clinical Oncology (Zimmerman et al. 2014). Based on this literature, the American Society of Clinical Oncology recently called for better integration of SPC into the care of cancer patients (Bakitas et al. 2015). Similar proposals have been suggested for patients in the ICU and for patients with heart failure and renal failure. Another stream of data shows that SPC can be associated with lower hospital costs. A series of articles show that SPC decreased costs both for patients who die in the hospital and those who survive. Studies in ambulatory and home-​based settings also suggest that SPC decreases health care utilization (Brumley, Enguidanos, and Cherin 2003; Klinger et al. 2013; Penrod et al. 2006; Zimmerman et al. 2014). These data have facilitated the growth of SPC. The Veterans Administration now requires all hospitals to have a palliative care consultative service. US News and World Report includes the presence of SPC as one of its quality criteria, and having an SPC is required if one is going to be a Comprehensive NCI Cancer Center. In 2000, only 24.5% of hospitals with more than fifty beds had palliative care consult services. In 2013,

428   Robert M. Arnold 72.3% had consultative services (CAPC 2015). Despite the increase in SPC, the public health data suggest that dying in America has not changed much. The second Institute of Medicine report, written sixteen years after the first one, found more SPC programs but little improvement in patient care overall, including among patients who are not receiving SPC consultations (Field 1997; Institute of Medicine 2014). Empirical data suggest that patients in the last years of life continue to have similar amounts of untreated physical and psychological symptoms (Singer et al. 2015; Temel et al. 2010). Although the number of people who die in the hospital has decreased slightly, the number of people admitted to the ICU and the number of transitions (transfers from one location to another) patients experience in the last six months of life is unchanged (Gozalo et al. 2011; Temel et al. 2010). Conversations about patients’ goals of care are still infrequent and occur only in the last week of life. A study looking at how house staff talked about code status in 2014 found little difference from how they had the conversations in 1991 (Teno et al. 2013). Although hospice use is up, the median length of stay in 2013 was only 18.5 days (it was 47 in 1995) (Field 1997; National Hospice and Palliative Care Organization 2014). Could it be that there is still not enough SPC? Maybe a greater increase in SPC is the answer. After all, if the data suggest that SPC improves care, maybe the problem is that there are just not enough of these specialists. The problem is that there will never be enough SPC consultants. Although there is no good estimate of the number of patients with serious illness and palliative care needs, some have said that it is 15–​25% of all hospitalized patients. There are about 1,658,370 new cancer cases expected to be diagnosed in the United States in 2015 (Cancer Facts and Figures 2015); a total of 95,889 patients on hemodialysis are aged 65–​74 years and 88,149 are older than 75 years (U.S. Renal Data System 2014). Even a mature SPC service will only see 15–​25% of the patients who die in the hospital. There are currently about 5,000 board-​certified palliative care physicians, and, of these, half work full-​time providing SPC. Roughly 200 new specialists complete their fellowships each year. Workforce studies estimate that the shortage of palliative care physicians is more than 10,000. A recent study found that 59% of palliative care clinicians reported emotional exhaustion, a finding that correlates with higher rates of clinicians leaving the field. And these estimates are focused on hospital-​based consultative services and do not take into account the growing need for SPC in the ambulatory, home, or long-​term settings. (There are fewer than five nurse practitioner [NP] or physician assistant [PA] palliative care–​specific programs in the country, and they are not likely to help significantly with the shortage; see Chiarella and Duffield 2007; Lupu 2010). Even if there were enough SPC programs, having SPC co-​manage all patients with a serious illness seems like a bad idea. It would mean that patients with a serious illness would have a primary care doctor, a palliative care clinician, and whatever subspecialists who were involved in their care. This is likely to lead to coordination problems, with each doctor ordering tests, changing medications, and giving slightly different explanations to the patient. Data already suggest that too many doctors are seeing patients and that their care is uncoordinated. Co-​management would mean that patients would have higher co-​ pays and spend more of their time seeing doctors (rather than living their lives).

Specialty Palliative Care   429 Given this reality, will most GPC be delivered by non-​SPC clinicians, if at all? My worry is that the promotion of SPC decreases the quality and quantity of GPC delivered.

The Cloud Surrounding the Silver Lining: The Marketing of Palliative Care and the Exclusion of Death Understanding that, for the most part, neither clinicians nor patients nor families are interested in “a better dying” as long as life can be prolonged, the Center for Palliative Care recommended that SPC de-​emphasize its care of dying patients. (The ability of SPC to prolong the lives of patients with advanced cancer, based on the results of a single-​ institution study, has been widely promulgated. The Center to Advance Palliative Care (CAPC) and the American Cancer Society talk euphemistically about serious illness and do not mention death when defining palliative care. Rather than talking about the limits of medicine and/​or trying to change the culture of medicine, palliative care should focus on more palatable topics such as its ability to promote quality of care for seriously ill patients (Arnold 2002). For some, this marketing is not going far enough. Bruera has published a series of studies showing that oncologists would be more likely to consult SPC if they were called “supportive care” than “palliative care” (Fadul et al. 2009). They feel that palliative care is a barrier to referral because it scares patients and decreases hope. The argument for de-​emphasizing death in the marketing of SPC is of consequence. More SPC is good for patients because it results in better care at lower costs. To get other clinicians to consult SPC, however, one should not talk about death and dying. Talking about death as the purview of SPC will lead other clinicians to say that it is “not time for a SPC consult.” Instead, the need for consultation has to be framed in terms of physical symptoms or family support. There are problems with this. First, from an empirical point of view, it is unclear the degree to which SPC’s marketing is successful. From the patient point of view, their willingness to see palliative care depends on whether the other doctors recommend it, not how it is marketed. And it is unclear if the death-​avoidance marketing fools doctors. Regardless of what you call SPC, the reality is that clinicians care for seriously ill patients who are likely to die. And one of the core roles of SPC is to talk to patients about their goals should they become sicker and to negotiate the transition to hospice. There is no reason to think that, whatever we call these clinicians (palliative care, supportive care, or something else), over time, they will be viewed as the clinicians who deal with death and dying. The attempt to find a name for what SPC does that is not tainted with death and dying seems like the search for the holy grail—​time-​consuming and unlikely to succeed. The reason is that SPC does take care of patients who are likely to die.

430   Robert M. Arnold Second, I worry that this fear of talking about death and dying will leave both the public and health care providers more, rather than less, scared of patients who are dying. If even the specialists who care for these patients cannot talk about it, imagine how scary it must be. Typically, things that cannot be spoken about in a society are taboo. They are so disruptive that just talking about them would be violating social norms. I cannot imagine that this is the view of death and dying that SPC providers want to promote among their colleagues. Might this increase the fear of dying among non-​SPC clinicians and lead them to be even less willing to talk to patients about these issues? This is an area that should be studied. Finally, I wonder what effect this marketing has on the clinicians who practice SPC. On one hand, the reality is that the majority of patients who SPC practitioners see are going to die in the next year or two. SPC clinicians know this and, in fact, often choose to do this work because they believe in the importance of caring for these patients. On the other hand, they cannot acknowledge it or talk about it, at least when they are talking to other clinicians. And they are being asked to change their identities to a euphemism like “supportive care.” At my institution, the SPC physicians were clearly upset about being asked to change their name to “supportive care.” They thought that it was disrespectful of their training and their identity. They wondered why they, among all the doctors in the hospital, could not talk about or be proud of what they do. They were angry that they could not express pride about their role and believed that these proposed changes made them the “pariahs” of medicine. A recent study found that 62% of palliative care clinicians met criteria for burnout (Kamal et al. 2014). High scores were associated with younger physicians and those who work in isolation. One possible reason for these high scores is that the clinicians are working in an environment in which their work is devalued. SPC clinicians cannot talk to other clinicians about what their work actually is. In addition, other clinicians want SPC providers to “hide” their role. Living this double identity may make it even more difficult to cope with the work of caring for the dying.

The Deskilling of Generalists and the “Ghettoization” of Palliative Care Regardless of how it is marketed, SPC consultants will never be able to see the majority of seriously ill and dying patients. The only plausible solution if one wants to improve the care of seriously ill patients in America is to improve the care provided by the generalists and other specialists. And so, a critical question asks how SPC has affected the quality and amount of GPC. I am worried that the growth of SPC worsened generalists’ skill levels rather than improved them. First, the need to develop a specialty implies that there is a level of knowledge and skills involved in palliative care that a generalist does not have. Second,

Specialty Palliative Care   431 the marketing of SPC also implies that it requires a higher level of skills. Finally, the studies of SPC show that patients do better when SPC is involved. Thus, if a doctor wants his or her patient to get the best care, he or she will consult SPC rather than trying to provide the care personally. Compounding this issue is the fact that too many doctors do not believe that palliative care is central to being a doctor. This encourages “turfing” of the problem to the specialist. When I was a resident, this is what happened when a patient with serious psychiatric illness or gynecological problems was admitted to the hospital. Rather than pay any attention or try to understand the patient’s issues and decide what to do, we just called the specialists or subspecialists who were responsible for that body part or problem—​in this case the psychiatrist or gynecologist. We would not do a psychiatric history or gynecologic exam for four reasons. First, we felt inadequate dealing with the problems both because we did not see them that frequently and because we felt we did not have sufficient training. Second, we did not really believe it was our job. “Women’s problems” were not central to our identity as internists. Third, we did not like doing this work because it was difficult. Talking to psychiatrically ill patients was time-​consuming and often raised emotional difficulties that we did not like thinking about—​addiction, abuse, and poverty. Fourth, we rationalized that the specialty services could do it faster and better. The effect of these referrals is that we did never developed our skills in performing pelvic exams or taking psychiatric histories. We finished our training and were no better at doing them than when we were medical students (to be honest, we were worse because, over the three years of training, our skills decayed). And then, we were out in practice where it is hard to get our patients seen because of the shortage of, for example, good psychiatric care. At this point, we had three options. First, bemoaning the lack of psychiatrists, one could refer the patient and hope he got seen by a psychiatrist at some point. Second, we could just avoid dealing with psychiatric care and hope that the problems would go away. Finally, we could muddle along, wishing we had received better training when we were residents. The sad state of generalists’ ability to practice or provide good psychiatric care is well documented (Stein 2004; Szabo 2013). I see the same thing happening in palliative care. Rather than have the house staff try to talk to the family or handle bad news, they just consult SPC. I often am asked in to talk to the family before the primary team has even met with the family because “that is your job.” When asked what the patient had said about his or her disease getting worse, I am told, “Oh, I was hoping you would talk to them and negotiate what comes next.” The primary service is often too busy to come to a family meeting and so tells us about the medical problem and sends the SPC consultant to do the work. Patients with palliative care needs become ghettoized. There are a variety of reasons this might be happening. The kindest explanation is that primary service providers realize that SPC providers are better at doing it than they are, and they want the patient to get the best possible care. Second, it may be that residents do not get enough training in GPC and avoid it because they feel inadequate. (If this is the reason, then it may be that SPC will lead to improved generalists’ ability to do palliative

432   Robert M. Arnold care because there has been an explosion of training both in medical school and residency about palliative care; see Bickel-​Swenson 2007; Mullan, Weissman, Ambuel, and Von Gunten 2004). Third, residents may not like having emotionally difficult conversations or writing prescriptions for opioids, so it is easier for them if they just give the tasks to someone else. Fourth, and this is unique to palliative care, the economics of SPC has led it to actively promote its services. If I want to grow a consult service, I need to market SPC and encourage the demand for SPC to fund my doctors’ time. And so SPC has actively promoted its access and encourages doctors to consult these specialists. Finally, like my view of psychiatry, most doctors do not believe it is their job. They still see palliative care as dealing with dying patients, and, as doctors and nurses, they want to be about “making the patient better.” So when there is “nothing more to do,” they refer the problem to somebody else. The problem is that there are not enough SPC providers to see all of the patients. So when the medical student leaves residency without the skills to do good GPC, he or she is forced, as I was with psychiatry, to muddle along. There are some data to support, if not the deskilling, then the fact that the growth of palliative care has not improved the skill sets of residents. In 1995, James Tulsky conducted a study at the University of California, San Francisco (UCSF) where he audiotaped doctors, largely house staff, talking to patients about code status decisions. Sixteen years later, Wendy Anderson conducted the same study at the same institution. Dr. Tulsky found that the conversations were not about goals, they did not provide information about risks and benefits, and they were largely devoid of any attention to the patient’s emotional state (Tulsky, Chesney, and Lo 1995). Sixteen years later, Wendy Anderson found the same thing. This is despite the fact that UCSF developed a nationally recognized palliative care program that has been responsible for training clinicians across the western United States about how to develop SPC programs (Anderson, Chase, Pantilat, Tulsky, and Auerbach 2011). It is not a good idea for doctors to try to refer all palliative care issues to SPC consultants, particularly in academic settings. SPC providers need to view their job as broader than just taking care of individual patients and to think about the pluses and minuses of marketing its services to learners. As educators, it may be that they need to not respond to every consult and instead to remind the consulting team that, although they will “teach,” they will not take over patient care. This model, however, will require some changes in medicine more generally. First, it would require paying SPC for teaching rather than having a salary based on how many patients the service sees. Second, it would mean changing the way that SPC services are funded. Currently, subspecialists are evaluated and paid by the number of consultations they do. A new model would encourage SPC providers to help the generalist do a better job of providing care rather than doing it themselves. In this model, SPC providers would be integrated into services as teachers on services that see patients with a lot of palliative care needs. For example, one can imagine an SPC clinician rounding daily in an ICU, helping the primary team provide the care the patients’ need (Mosenthal et al. 2012). An SPC consultant would only be called if the intensive care team could not meet a patient’s needs.

Specialty Palliative Care   433

A Population Health View of GPC The problem of ghettoization and deskilling is indicative of a larger issue—​that the attention and research on palliative care has focused on the specialty service rather than the care of all patients with a serious illness. As palliative care developed its own specialty, the research literature has focused largely on SPC’s impact. The number of journals focusing on SPC has increased, and articles have focused on how hospital, ambulatory, and home care SPC have improved patients’ and their caregivers’ lives. Even in the lay media outlets, the conversation has focused on SPC. In the past six months, there have been articles focusing SPC programs in the New York Times, Wall Street Journal, the Pittsburgh Post-​Gazette, and Harvard magazine. These articles focus on the importance of SPC and what SPC can do to improve the quality of life for patients and their families. Websites such as www.getpalliativecare.com urge patients and their families to ask for SPC. The CAPC has raised millions of dollars from philanthropists to promote SPC. The Center worked with national organizations such as the Joint Commission and the National Quality Forum to promote standards for SPC. Their mission was to increase access to SPC, and they have succeeded beyond the field’s wildest dreams. It should be noted that, recently, the CAPC has started to devote attention to promoting GPC. The American Academy of Hospice and Palliative Medicine has doubled its membership as it has promoted the need for SPC and the training of its practitioners. There is nothing wrong with the growth of SPC per se. However, from a public health point of view, it may not be the right place to devote all of our resources. That is true for two reasons: first, the overselling of SPC sets the field up to fail to reach its stated goal—​ to improve the lives (and deaths) of as many seriously ill people as possible. As noted earlier, there will never be enough SPC doctors to see all of the patients who have GPC needs. Encouraging patients and health care providers to continue to use SPC will just frustrate SPC providers as SPC becomes overwhelmed and pushes back on consultations that are within the domain of generalists. Second, from a public heath point of view, we should not be concerned with who cares for patients as long as their needs are met. For example, think about patients with coronary artery disease. Our society cares about whether these patients get aspirin and statins appropriately and about patients’ mortality and morbidity after a myocardial infarction. They have less interest in whether they were seen by a cardiologist. (In fact, if referrals to a cardiologist are too high, this is a problem from a public health point of view because it indicates a lack of generalist education and inefficiency). Similarly, from a public health point of view, the question is whether the care of seriously ill and dying patients is getting better. The data quoted earlier suggest that the answer to whether the care of seriously ill and dying patients is getting better is “probably not.” Thus, rather than celebrating SPC, we should be questioning whether it is the best model to improve the care of most patients.

434   Robert M. Arnold It may be that the development of SPC is an intermediate stage to improving the care of all patients. Spending money on GPC without having a cadre of leaders in palliative care might be too diffuse a solution to have any impact. On the other hand, it may be that developing SPC is not a strong enough intervention to change the culture of medicine and raise the level of care for all patients with a serious illness. To determine which of these possibilities is accurate, we need to spend more research dollars focusing on how all seriously ill and dying patients are doing, not just those seen by SPC providers. Our focus should move from SPC to GPC and its impact on all patients. As much as we hold generalists responsible for the care for diabetic patients or patients with hypertension, we should develop standards to ensure that the majority of patients who have palliative care needs have those needs met by their non-​SPC clinicians providing GPC. Only when those needs can’t be met should they be referred to SPC. This should not be that hard to do. Research has begun to define a group of patients for whom palliative care is at the forefront of their care needs. For example, one way to define the population of patients who need GPC is to ask the question, “Would you be surprised if the patient is admitted to the intensive care unit or dies?” Studies of various seriously ill patient populations show that when clinicians answer “no” to this question, patients have a significantly higher rate of morbidity and mortality (Moss et al. 2010). Alternatively, other studies have used other criteria to define patients who are at high risk of hospitalization or death (Weissman and Meier 2011). Having defined the population of patients in need of GPC, funding should be used to focus on whether interventions can improve the care of entire patient populations. This would force the small cadre of SPC providers to be as concerned with population health as they are with how they deal with individual patients. Rather than promoting SPC, the focus of our efforts should be to improve the care of all seriously ill or dying patients.

Critique and Conclusion Proponents of SPC may read this chapter and think that I am blaming SPC for not solving the problems of modern American medicine (or society). As a friend put it, “It is not fair to blame a flashlight for the darkness.” And I do understand that sentiment. The problem is that SPC leaders should care less about their own small flashlight and more about the all-​encompassing darkness. Society has spent a good deal of money promoting SPC, building its infrastructure, developing SPC fellowships, and more. I wonder if this is the best way to light up a room. It is an open question as to whether the use of fancy LED flashlights is a better way to light up the room than lighting multiple candles. This is particularly true if the building of flashlights has diminished the general population’s interest or skill in lighting candles. Finally, many will complain that I  have posited a false choice between promoting SPC and helping all doctors, regardless of specialty, provide better palliative care (GPC). Although this is true in theory, I think in an era of limited resources and energy,

Specialty Palliative Care   435 society needs to be intentional about where to focus its efforts. The problem is misplaced priorities. The literature, for the most part, has focused only on how SPC has improved the care of patients who received an SPC consultation rather than testing interventions that promote GPC and looking at their effect on all patients. If SPC is only going to touch a minority of seriously ill patients, then researchers should be spending more energy focusing on the larger group of seriously ill patients who do not receive SPC. When the negative results of SUPPORT were published, the investigators believed that more forceful interventions were required to improve the care of seriously ill patients. One investigator reflected that, “Several of the SUPPORT leaders noted that the efforts to improve end-​of-​life care were inhibited by a medical structure that made it very difficult to provide high quality end-​of-​life care and argued that improvements were unlikely without a major societal commitment to improve end-​of-​life care” (GeriPal 2010). The question is whether, over the past twenty years, this statement is still true. Has the growth of SPC changed the underlying medical structure that will lead to improved care of all patients or has it merely served as a practice, which, while improving the care of some patients, does not lead to larger changes in the culture of medicine? There is a way that SPC can lead to better GPC. SPC will change from a consultation service that cares for one patient at a time to a specialty that focuses on culture change, quality improvement, and population health. First, this will require a different kind of training than most clinicians receive. Rather than making quality improvement a small part of fellows’ training, it will require fellows to receive as much training in health care administration, changing clinicians’ behaviors, information technology, and measuring outcomes as they currently get in pain management or communication skills. Second, the incentives that support SPC need to change. Rather than fiscally rewarding clinicians based on how many patients they see, which promotes encouraging consultation, incentives need to reward clinicians for helping other health care providers meet population-​wide metrics. Third, SPC needs to be embedded into the health care organizational structure so that SPC clinicians can work to improve the routines that drive clinical care. For SPC to change clinical care, the metrics they care about need to be the ones that health care institutions care the most about. Only then will SPC be able to ensure that GPC is sufficiently robust to revolutionize medical care and ensure that all seriously ill patients get appropriate care.

Acknowledgments I am thankful to Debbie Seltzer for making this chapter readable and to Yael Schenker and Chris Feudtner for helping me clarify the argument and present it in a less apologetic way. The title was Chris’ suggestion.

436   Robert M. Arnold

References Anderson, W. G., R. Chase, S. Z. Pantilat, J. A. Tulsky, and A. D. Auerbach. 2011. “Code status discussions between attending hospitalist physicians and medical patients at hospital admission.” Journal of General Internal Medicine 26(4): 359–​366. PMID: 21104036 Arnold, R. M. 2002. “Rose by any other name.” Journal of Palliative Medicine 5(6): 807–​811. http://​online.liebertpub.com/​toc/​jpm/​5/​6. ASCO Institute for Quality. 2012. ASCO provisional clinical opinion: The integration of palliative care into standard oncology care. http://​www.instituteforquality.org/​asco-​provisional-​ clinical-​opinion-​integration-​palliative-​care-​standard-​oncology-​care. Accessed May 5, 2015. Bakitas, M. A., T. D. Tosteson, Z. Li, et al. May 1, 2015. “Early versus delayed initiation of concurrent palliative oncology care: Patient outcomes in the ENABLE III Randomized Controlled Trial.” Journal of Clinical Oncology 33(13): 1438–​1445. http://​jco.ascopubs.org/​content/​33/​13/​ 1438.abstract. Bickel-​Swenson, D. 2007. “End-​of-​life training in U.S. medical schools: A systematic literature review.” Journal of Palliative Medicine 10(1):  229–​235. http://​www.ncbi.nlm.nih.gov/​ pubmed/​17298271. Brennan, F. 2007. “Palliative care as an international human right.” Journal of Pain and Symptom Management 33(5): 494–​499. doi:10.1016/​j.jpainsymman.2007.02.022. http://​www. jpsmjournal.com/​article/​S0885-​3924(07)00155-​8/​abstract. Accessed May 5, 2015. Brumley, R. D., S. Enguidanos, and D. A. Cherin. 2003. “Effectiveness of a home-​based palliative care program for end-​of-​life.” Journal of Palliative Medicine 6(5): 715–​724. http://​www. ncbi.nlm.nih.gov/​pubmed/​14622451. Cancer Facts and Figures–​2015. American Cancer Society. 2015. http://​www.cancer.org/​acs/​ groups/​content/​@editorial/​documents/​document/​acspc-​044552.pdf. Center to Advance Palliative Care Metrics and Measurement in Palliative Care (CAPC). 2015. Metrics and measurement in palliative care. https://​www.capc.org/​topics/​metrics-​and-​ measurement-​palliative-​care/​. Accessed April 30, 2015. Chiarella, M., and C. Duffield. 2007. “Workforce issues in palliative and end-​of-​life care.” Journal of Hospice & Palliative Nursing 9(6):  334–​341. doi:10.1097/​01.NJH.0000299315.07139.72. http://​journals.lww.com/​jhpn/​Abstract/​2007/​11000/​Workforce_​Issues_​in_​Palliative_​and_​ End_​of_​Life.17.aspx. Connors, A. F., Jr., N. V. Dawson, N. A. Desbiens, et al. 1995. “A controlled trial to improve care for seriously iii hospitalized patients—​The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT).” Journal of the American Medical Association 274(20): 1591–​1598. doi:10.1001. http://​www.ncbi.nlm.nih.gov/​pubmed/​7474243. Accessed May 5, 2015. Covinsky, K. November 29, 2010. The Legacy of SUPPORT: 15 years later GeriPal—​A geriatrics and palliative care blog. http://​www.geripal.org/​2010/​11/​legacy-​of-​support-​15-​years-​later. html. Fadul, N., A. Elsayem, J. L. Palmer, et al. 2009. “Supportive versus palliative care: What’s in a name?: A survey of medical oncologists and midlevel providers at a comprehensive cancer center.” Cancer 115(9): 2013–​2021. doi:10.1002/​cncr.24206. http://​www.ncbi.nlm.nih.gov/​ pubmed/​19235253. Field, M. J. 1997. Approaching death: Improving care at the end of life. Committee on Care at the End of Life, Institute of Medicine. Washington, DC: National Academies Press, p. 169. http://​ www.nap.edu/​catalog/​5801/​approaching-​death-​improving-​care-​at-​the-​end-​of-​life.

Specialty Palliative Care   437 Foley, K. M., and H. Gelband. 2001. Improving palliative care for cancer. Washington, DC: National Academies Press. http://​www.nap.edu/​catalog/​10149/​improving-​palliative-​care​for-​cancer. Gawande, A. 2014. Being mortal: Medicine and what matters in the end. Deckle Edge. http://​ www.amazon.com/​dp/​0805095152/​?tag=mh0b-​20&hvadid=7000020933&hvqmt=e&hvb mt=be&hvdev=c&ref=pd_​sl_​6ywkunzsyo_​e. GeriPal. November 29, 2010. The Legacy of SUPPORT. 15 years later. http://​www.geripal.org/​ 2010/​11/​legacy-​of-​support-​15-​years-​later.html. Accessed May 4, 2015. Gozalo, P., J. M. Teno, S. L. Mitchell, J. Skinner, J. Bynum, D. Tyler, and V. Mor. 2011. “End-​of-​ life transitions among nursing home residents with cognitive issues.” New England Journal of Medicine 365:  1212–​1221. doi:10.1056/​NEJMsa1100347. http://​www.nejm.org/​doi/​full/​ 10.1056/​NEJMsa1100347. Institute of Medicine. 2014. Dying in America: Improving quality and honoring individual preferences near the end of life. Washington, DC: National Academies Press. http://​www.iom. edu/​Reports/​2014/​D ying-​In-​America-​Improving-​Quality-​and-​Honoring-​Individual-​ Preferences-​Near-​the-​End-​of-​Life.aspx. Kamal, A., J. Bull, S. Wolf, et al. 2014. “Burnout among palliative care clinicians in the United States:  Results of a national survey.” Journal of Clinical Oncology. Accessed May 4, 2015. http://​meetinglibrary.asco.org/​content/​132497-​144. Klinger, C. A., D. Howell, D. Marshall, D. Zakus, K. Brazil, and R. B. Deber. 2013. “Resource utilization and cost analyses of home-​based palliative care service provision: The Niagara West end-​of-​life shared-​care project.” Palliative Medicine 27(2):  115–​ 122. doi:  10.1177/​ 0269216311433475. http://​www.ncbi.nlm.nih.gov/​pubmed/​22249926. Lupu, D. 2010. “Estimate of current hospice and palliative medicine physician workforce shortage.” Journal of Pain and Symptom Management 40(6):  899–​911. doi:10.1016/​ j.jpainsymman.2010.07.004. http://​www.ncbi.nlm.nih.gov/​pubmed/​21145468. MacArthur Foundation. 2008. MacArthur Fellows Program/​Meet the class of 2008. http://​ www.macfound.org/​fellows/​805/​. Accessed May 5, 2015. Mosenthal, A. C., D. E. Weissman. J. R. Curtis, et al. 2012. “Integrating palliative care in the surgical and trauma intensive care unit:  A  report from the Improving Palliative Care In The Intensive Care Unit (IPAL-​ICU) project advisory board and the Center to Advance Palliative Care.” Critical Care Medicine 40: 1199–​1206. doi: 10.1097/​CCM.0b013e31823bc8e7. http://​www.ncbi.nlm.nih.gov/​pubmed/​22080644. Moss, A. H., J. R. Lunney, S. Culp, et al. 2010. “Prognostic significance of the surprise question in cancer patients.” Journal of Palliative Medicine 13(7): 837–​840. Mullan, P. B., Weissman, D. E., Ambuel, B., and Von Gunten, C. 2004. “End-​of-​life care education in internal medicine residency programs:  An interinstitutional study.” Journal of Palliative Medicine 5(4):  487–​496. doi:10.1089/​109662102760269724. http://​online.liebertpub.com/​doi/​abs/​10.1089/​109662102760269724. National Hospice and Palliative Care Organization. October 1, 2014. NHPCO’s facts and figures:  Hospice care in America. http://​www.nhpco.org/​sites/​default/​files/​public/​Statistics_​ Research/​2014_​Facts_​Figures.pdf. Accessed May 6, 2015. Penrod, J. D., P. Deb, C. Luhrs, et al. 2006. “Cost and utilization outcomes of patients receiving hospital-​based palliative care consultation.” Journal of Palliative Medicine 9(4): 855–​860. http://​www.ncbi.nlm.nih.gov/​pubmed/​16910799. Quill, T. E., and A. P. Abernethy. 2013. “Generalist plus specialist palliative care—​Creating a more sustainable model.” New England Journal of Medicine 368:  1173–​1175. doi:  10.1056/​ NEJMp1215620. http://​www.nejm.org/​doi/​full/​10.1056/​NEJMp1215620. Accessed May 5, 2015.

438   Robert M. Arnold Singer, A. E., D. Meeker, J. M. Teno, J. Lynn, J. R. Lunney, and K. A. Lorenz. 2015. “Symptom trends in the last year of life from 1998 to 2010: A cohort study.” Annals of Internal Medicine 162(3): 175–​183. doi:0.7326/​M13-​1609. http://​annals.org/​article.aspx?articleid=2107746. Smith, T. J., S. Temin, E. R. Alesi, et al. 2011. “American Society of Clinical Oncology provisional clinical opinion:  The integration of palliative care into standard oncology care.” Journal of Clinical Oncology 30(8): 880–​887. http://​jco.ascopubs.org/​content/​early/​2012/​02/​ 06/​JCO.2011.38.5161.abstract. Stein, M. B. 2004. “Quality of care for primary care patients with anxiety disorders.” American Journal of Psychiatry 161(12): 2230–​2237. doi: http://​dx.doi.org/​10.1176/​appi.ajp.161.12.2230. http://​ajp.psychiatryonline.org/​doi/​abs/​10.1176/​appi.ajp.161.12.2230. Szabo, L. January 8, 2013. “State report card on mental health care.” USA Today. http://​www. usatoday.com/​story/​news/​nation/​2013/​01/​07/​states-​mental-​health/​1805023/​. Accessed May 4, 2015. Temel, J. S., J. A. Greer, A. Muzikansky, et  al. 2010. “Early palliative care for patients with metastatic non-​small-​cell lung cancer.” New England Journal of Medicine 363:  733–​742. doi: 10.1056/​NEJMoa1000678. http://​www.nejm.org/​doi/​full/​10.1056/​nejmoa1000678. Teno, J. M., P. L. Gozalo, J. P. W. Bynum, et al. 2013. “Change in end-​of-​life care for Medicare beneficiaries site of death, place of care, and health care transitions in 2000, 2005, and 2009.” Journal of the American Medical Association 309(5): 470–​477. doi:10.1001/​jama.2012.207624. http://​jama.jamanetwork.com/​article.aspx?articleid=1568250. Tulsky, J. A., M. A. Chesney, and B. Lo. 1995. “How do medical residents discuss resuscitation with patients?” Journal of General Internal Medicine 10(8): 436–​442. PMID: 7472700. U.S. Renal Data System. 2014. Chapter 1: Incidence, prevalence, patient characteristics, and treatment modalities (Table  1.5). http://​www.usrds.org/​2014/​view/​v2_​01.aspx. Accessed April 30, 2015. Weissman, D. E., and D. E. Meier. 2011. “Identifying patients in need of a palliative care assessment in the hospital setting: A consensus report from the Center to Advance Palliative Care.” Journal of Palliative Medicine 14(1): 17–​23. doi:10.1089/​jpm.2010.0347. http://​www.ncbi.nlm. nih.gov/​pubmed/​21133809. Wikipedia. n.d. Palliative care. http://​en.wikipedia.org/​w/​index.php?title=Palliative_care&oldid=65702 6207 . Zimmerman, C., N. Swami, M. Krzyzanowska, et al. 2014. “Early palliative care for patients with advanced cancer: A cluster-​randomized controlled trial.” Lancet 383(9930): 1721–​1730. doi:  http://​dx.doi.org/​10.1016/​S0140-​6736(13)62416-​2. http://​www.thelancet.com/​journals/​ lancet/​article/​PIIS0140-​6736%2813%2962416-​2/​abstract.

Chapter 27

Marketi ng Palliativ e  C a re Bridget Tracy and Rolfe Sean Morrison

Introduction Palliative care has come a long way in its short existence. Thanks to a growing body of research, philanthropic support of major foundations driving change in the US health-​care system, and new policies reflecting governmental support of palliative medicine, an increasing number of seriously ill persons and their families have access to high-​quality palliative care. Nevertheless, palliative care is still accessed by a minority of persons with serious illness who could benefit from this care. Whereas a number of factors contribute to this issue, including a lack of available programs, providers, and reimbursement, even well-​established, adequately staffed, and fiscally strong programs struggle to reach all persons in need. Why is this? In this article, we argue that a major limitation to the growth of palliative care has been the lack of a successful social marketing platform directed to consumers. Furthermore, we argue that the early language and messaging developed by palliative care professionals failed to consider basic principles of social marketing and, as such, hindered the growth of the field. We explore how palliative care, which developed as a movement by health-​care professionals to address well-​documented and pervasive poor-​quality care for dying patients, failed to capture widespread public support and engagement; describe how early messaging by palliative care professionals hindered referrals to palliative care by other health-​care professionals; and conclude by describing how recent attempts to use principles of audience research and targeted social marketing have led to rapid spread and uptake of palliative care services in the United States.

440    Bridget Tracy and Rolfe Sean Morrison

The Modern Hospice Movement The discovery of penicillin and integration of antibiotics into routine medical practice in the 1950s and the subsequent rapid advancements in medical treatments over the succeeding decades changed how medicine defined death and dying. Care of patients with serious illness shifted from the home and into the hospital, and the dying process became increasingly medicalized. Concerns about suffering at the end of life and abandonment of patients when medical treatments were no longer effective led to the development of the modern hospice. The term “hospice” stems from the Latin hospitum, meaning a place to host, receive, and entertain guests. The original hospices, in medieval times, were houses of rest or shelter for pilgrims to the Holy Land. The modern hospice (places to care for the dying) were established by religious orders in Europe (Meier, Isaacs, and Hughes, 2010). It was in one of these hospices that Cicely Saunders, an English nurse, social worker, and physician, was first exposed to end-​of-​life care. She, along with Florence Wald in New Haven and Elisabeth Kübler-​Ross in Chicago, launched the modern hospice movement. Saunders founded St. Christopher’s Hospice in London that not only cared for the dying but mandated education and research at the core of its mission. Wald, having studied the work of Saunders, brought the concept of hospice to the United States, establishing the Connecticut Hospice in 1974. During the 1970s and early 1980s, hospice evolved as an alternative to mainstream medical care. Focused almost exclusively on patients who elected to forego disease-​ directive treatments, or those for whom treatments were no longer effective, hospice evolved in the minds of both medical professionals and the public as an alternate care system for the dying. As the hospice movement grew, the government took notice. In 1979 a demonstration program examined twenty-​six hospice sites and fourteen usual care sites across the country and found that hospice care did not deprive patients of care, often lowered costs, and allowed patients to spend more time at home and less in hospitals or other acute care settings (Connor, 2007). The results of the National Hospice Study led to the enactment of the Medicare Hospice Benefit in 1982. Subsequent concerns about expenditures under the new hospice benefit led to the establishment of strict eligibility criteria in 1986 that limited access to hospice to persons with a prognosis of six months or less if the disease followed its usual course, those willing to forego curative treatments, and those willing to forego Medicare Part A. Furthermore, although patients under hospice are allowed to keep their primary physician, in practice misunderstandings and difficulties in billing for services (primary physicians are required to bill hospice rather than Medicare for their services) leads most patients to be cared for in a separate care system—​isolated from their long-​term providers. In summary, the establishment of the hospice benefit—​designed to both provide comfort and limit costs in the last six months of life—​cemented in language and in the public and professional’s mindset the false perception that medicine had two mutually exclusive goals—​life prolongation and cure or comfort.

Marketing Palliative Care   441

Development of Palliative Care A decade after enactment of the Medicare Hospice benefit, concerns about changes in structures and delivery of health care and the recognition that the majority of Americans were still dying without accessing their hospice benefit led a group of investigators and the Robert Wood Johnson Foundation to initiate “The Study to Understand the Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT; SUPPORT Principal Investigators 1995). SUPPORT was designed to describe and then subsequently improve care for persons with serious illness in American hospitals. Whereas the SUPPORT intervention (communication of patient preferences and prognosis to treating physicians) failed to achieve demonstrable changes in practice patterns, its observational data provided a wealth of information regarding the symptom burden, quality of care, and family experience of persons cared for in five major US teaching hospitals. Specifically, results from SUPPORT demonstrated that high levels of pain were common across all diagnostic categories, patients consistently received treatments in conflict with prior stated goals, hospital deaths were common despite wishes to die at home, and families were both impoverished and extremely burdened despite being appropriately covered by medical insurance. SUPPORT’s publication and subsequent media coverage in 1995, coupled with publicity surrounding Jack Kevorkian and the movement to legalize physician-​aid-​in dying, led to a series of initiatives focused on enhancing care for the majority of seriously ill persons being cared for outside of hospice. Philanthropic investment of almost $300 million led George Soros’ Open Society Institute’s Project and the Robert Wood Johnson Foundation, focused on educational, clinical, and research initiatives, took the principles of hospice care (interdisciplinary teams providing pain and symptom management; emotional, psychological, and spiritual support; and practical support for families and other loved ones from diagnosis through bereavement) and integrated them into mainstream medicine. The new US specialty of palliative care thus evolved to provide care for persons with serious illness regardless of prognosis and in conjunction with all other appropriate medical treatments, including those to prolong life.

Enhancing Access to Palliative Care Despite a solid body of evidence of evidence that palliative care improves clinical quality, enhances patient and family satisfaction, lowers health-​care costs, and enhances survival, palliative care is considerably underutilized. More than one-​third of US hospitals do not have palliative care teams, and even the hospitals that do see only a fraction of those in need (“Growth of Palliative Care in U.S. Hospitals 2012 Snapshot,” 2012; Voelker, 2011). Palliative medicine consultation, a frequent pathway for referral to hospice care, occurs late in hospitalization; 30 percent of hospice patients are not referred

442    Bridget Tracy and Rolfe Sean Morrison until the last week of life and 10 percent are referred in the last twenty-​four hours (Teno et al., 2007). African Americans, Asian Americans, and Latinos are less likely to receive hospice care. There is wide geographic variation in access to palliative care, as well as the quality of care offered. For all of the benefits palliative care delivers to hospitals, clinicians, and patients and their families, it is remarkable that the specialty has failed to capture the endorsement of patients, families, clinicians, and policymakers.

Marketing a New Specialty What are the barriers to access to palliative care? One answer to this question lies in palliative care’s origins within the hospice movement. Palliative care’s model of comprehensive, interdisciplinary support for patients and families is adapted from the hospice model. In fact, many clinicians and researchers came to the field of palliative care out of an interest in helping Americans at the end of life to have a good death. In communicating with patients, families, clinicians, and policymakers, the language that has been often used has been adopted from hospice. Specifically, it has included such phrases as “improving care at the end of life,” achieving a peaceful death,” “focusing on comfort,” and “dying with dignity,” for example. The problem is that many of the patients who would benefit from palliative care are not dying, and even among the subset who are, no one wants to die, and very few, including clinicians, are able to accept that they are dying until death is imminent. If palliative care is going to meet the needs of these increasing numbers of persons living with serious illness, it needs to be decoupled from end-​of-​life care, terminal care, and hospice

Principles of Social Marketing Understanding the values, fears, and desires of the patient and the family and matching treatments to those values and goals is at the heart of what palliative care offers. Similarly, the science of social marketing focuses on influencing behavior change to benefit individuals or communities through an in-​depth understanding of their needs. Successful social marketing involves (1)  problem definition or identification of the desired behavior change; (2) market research to understand the needs, values, and opinions of the target audience; (3) creation of a market strategy that aligns with the target audience; (4) execution of that strategy; and (5) evaluation of the strategy and adjustment as needed (Morrison and Merkin, 2011). With the principles of social marketing in mind, it is clear that early messages about palliative care were not matched to the needs of its target population. Our discomfort with death and our reluctance to discuss the dying process are evident throughout human history and likely have deep biological and evolutionary roots. Indeed, humankind’s fear of death and dying can be traced from Homer (“Better to flee from death than feel its grip”) to Shakespeare (“The weariest and most loathed worldly life that age, ache, penury, and imprisonment can lay on nature is a paradise to what we fear of death”) to

Marketing Palliative Care   443 Woody Allen (“I don’t want to achieve immortality through my work. I’d rather achieve it by not dying”). Not surprisingly, attempts to engage the public in services aimed at helping them die a better death—​a highly unwanted outcome—​have been ineffectual. What then is required to better engage the public, clinicians, and policymakers to enhance care of the seriously ill and those nearing or at the end of life?

Steps Forward Successful marketing strategies offer the audience a benefit that they perceive as valuable. The value of a benefit is subjective, as different audiences will perceive different benefits from each. In 2011 the Center to Advance Palliative Care and the American Cancer Society conducted the first social marketing research study to better grasp the public’s knowledge of palliative care and to develop an audience-​specific and patient-​centered definition for the specialty. The sample included 800 adults over the age of twenty-​five, oversampling seniors aged sixty-​five and over. Public support was greatest for the following definition: “Palliative care is specialized medical care for people with serious illness. This type of care is focused on providing patients with relief from the symptoms, pain, and stress of a serious illness—​whatever the diagnosis. The goal is to improve quality of life for both the patient and the family. Palliative care is provided by a team of doctors, nurses, and other specialists who work with a patient’s other doctors to provide and extra layer of support. Palliative care is appropriate at any age and at any stage in a serious illness, and can be provided together with curative treatment.” Of note, once this definition was developed and presented to consumers, 95 percent agreed that it was important for patients with serious illness and their families to be educated about palliative care, 92 percent said they would likely consider palliative care for a loved one if he or she had a serious illness, and 92 percent said it was important that palliative care services be made available at all hospitals (Center to Advance Palliative Care, 2011). Messaging to clinicians, health-​care institution administrators, and payers and policymakers builds on the key messages outlined here to patients but needs to take into account the additional needs of these audiences. For example, clinicians are under increasing pressure to see more patients in less time while at the same time meeting an increasing number of mandated quality metrics. Thus marketing palliative care to physicians should stress the added level of support that palliative care teams provide. Conversely, marketing to physicians should not stress elements they are not providing (e.g., pain and symptom management, enhanced communication) but rather should focus on the assistance that palliative care teams provide with complex care coordination, accessing and managing resources for outpatient support, and collaboration with multiple disciplines to provide patients with superior discharge plans that keep them out of hospitals and doctors’ offices (Enguidanos, Vesper, and Lorenz, 2012). Messaging should also stress the fact that patients who receive palliative care as part of their medical treatment report higher levels of satisfaction with their physicians, health-​care team, and hospitals (Ringdal, Jordhoy, and Kaasa, 2002; Temel et al., 2010).

444    Bridget Tracy and Rolfe Sean Morrison Hospital administrators and policymakers are concerned with meeting quality standards, containing costs, and maintaining the satisfaction of those they serve, be they patients, staff, or constituents. Thus the appropriate communication to these stakeholders is that palliative care teams partner with the services in the system to enhance quality of care at no increase in cost. Research from hospitals with palliative care programs demonstrate that costs are lower for this patient population, patients are transitioned to more appropriate levels of care earlier in the course of their hospital stay, readmissions are reduced, and hospice referrals are increased.

Marketing Approaches Finally, translating these findings into successful campaigns and publications for social marketing requires an awareness and selection of the right outlets. Over the past few years, more targeted campaigns with messages touting palliative care’s ability to treat the whole person, improve the quality of life for patients, and increase quality for the greater system have garnered success. These campaigns focus on positive images and experiences. They use the phrasing tested in polling paired with positive images to sell the known benefits of palliative care and avoid negative images mentioning death and dying. An example from the American Cancer Society is shown in Figure 27.1.

Figure 27.1  The American Cancer Society’s “Dancer” advertisement. An example of a successful social marketing campaign.

Marketing Palliative Care   445

Conclusion Targeted social marketing has facilitated the rapid spread and uptake of palliative care services in the United States, improving the care of people living with serious illness and taking the field beyond its roots in hospice care. Through continued strategic planning, research, and vision, we hope that all patients and families will know to request palliative care in the setting of serious illness, that all health-​care professionals will have the skills and knowledge to provide palliative care, and that all health-​care institutions will be able to support and deliver high-​quality palliative care.

References Center to Advance Palliative Care. 2011. 2011 public opinion research on palliative care: A report based on research by public opinion strategies. Accessed January 30, 2014, https://​www.capc. org/​media/​filer_​public/​18/​ab/​18ab708c-​f835-​4380-​921d-​fbf729702e36/​2011-​public-​opinion-​ research-​on-​palliative-​care.pdf Connor, S. R. 2007. Development of hospice and palliative care in the United States. Omega (Westport) 56(1): 89–​99. Enguidanos, S., E. Vesper, and K. Lorenz. 2012. 30-​day readmissions among seriously ill older adults. Journal of Palliative Medicine 15(12): 1356–​1361. Growth of palliative care in U.S. hospitals 2012 Snapshot. 2012. Accessed July 15, 2014, http://​ reportcard.capc.org/​pdf/​capc-​growth-​analysis-​snapshot-​2011.pdf Meier, D. E, S. L. Isaacs, and R. G. Hughes. 2010. Palliative care: Transforming the care of serious illness. 1st ed. San Francisco, CA: Jossey-​Bass. Morrison, R. S, and H. Merkin. 2011. Marketing our message. Journal of Palliative Care 27(1): 5. Ringdal, G. I., M. S. Jordhoy, and S. Kaasa. 2002. Family satisfaction with end-​of-​life care for cancer patients in a cluster randomized trial. Journal of Pain and Symptom Management 24(1): 53–​63. SUPPORT Principal Investigators. 1995. A controlled trial to improve care for seriously ill hospitalized patients: The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). JAMA 274(20): 1591–​1598. Temel, J. S., J. A. Greer, A. Muzikansky, et al. 2010. Early palliative care for patients with metastatic non-​small-​cell lung cancer. New England Journal of Medicine 363(8): 733–​742. Teno, J. M., Shu, J. E., D. Casarett, C. Spence, R. Rhodes, and S. Connor. 2007. Timing of referral to hospice and quality of care: Length of stay and bereaved family members’ perceptions of the timing of hospice referral. Journal of Pain and Symptom Management 34(2): 120–​125. Voelker, R. 2011. Hospital palliative care programs raise grade to B in new report card on access. JAMA 306(21): 2313–​2314.

Index

AACN. See American Association of Critical Care Nurses abortion conscience-​based refusals for,  103 conscience clause legislation for,  103 federal funding and,  87 physician’s moral and religious objection to, 87 prenatal fetal screening influencing,  43 reasoning for,  43 training for,  102 abuse child, 46 PAS potential for,  384, 386 of physician aid-​in dying,  22 acceptable euthanasia,  358 ACCP. See American College of Chest Physicians ACLU. See American Civil Liberties Union Act 69, physician protection by,  105 active euthanasia (lethal injection),  3, 24 actively hastening death,  18–​19 advertising blitz against legalization of,  22–​23 consent for,  21 as criminal homicide,  21 as end-​of-​life controversy,  20–​21 increased awareness regarding,  23 as legally unacceptable,  21–​22 moral culpability and,  23–​24 patient suffering and,  21–​22, 378–​379 public support for,  22 active versus passive end-​of-​life decision,  13 acute infant suffering,  45, 47 acute neonatal illness, counseling and,  44–​45 Adams, John Bodkin,  115 adjustment disorder with depressed mood, 395

advance care planning by African Americans and Hispanics,  312 health care professional discussion on, 320 regarding nutrition and hydration,  143 advance directives for continuous sedation,  116 court system on,  16 decision-​making capacity and,  15 existence of,  286 legislation passed supporting,  17 living will as,  15–​16 movement for,  31, 32 national database for,  280 social media linked with,  280, 286–​87, 288n19 uncertainty surrounding,  16 advance notification, for conscience-​based objections,  95–​96 Aesculapian power,  145 African Americans advance care planning by,  312 aggressive treatment for,  247–​48, 316 cultural beliefs, values and preferences of,  309, 314–​15 culture of,  309 end-​of-​life care and,  312, 315–​16 family importance of,  312 health care system mistrust and,  317 hospice and palliative care services use,  313–​14 pain management of,  311 spiritual beliefs of,  316 aggressive treatment for African Americans,  247–​48, 316 in end-​of-​life care,  230 financial cost of,  242, 245 for Hispanics,  247–​48, 316

448   Index aggressive treatment (Cont.) for newborns with life-​threatening illness, 45 poor prognosis, limited life expectancy and,  321–​22 request for,  321–​22 Aikenhead, Mary,  412 All About My Mother (Almodovar),  55–​57 Almodovar, Pedro,  55–​57 Alzheimer’s disease,  109 end stage of,  135–​36 ERC and,  361 euthanasia and,  363 KNMG and,  361 nutrition and hydration during end stage of,  134–​35 PAD and,  352 AMA. See American Medical Association American Academy of Hospice and Palliative Medicine,  280, 373, 421 American Academy of Pediatrics,  46, 49, 50 American Association of Critical Care Nurses (AACN), 79 American Cancer Society,  429, 443–​44, 444f American Civil Liberties Union (ACLU),  269 American College of Chest Physicians (ACCP), 79 American College of Physicians (2012),  78 American hospitals, dying in,  294–​96 American Medical Association (AMA),  30, 77, 89, 96, 111, 425 on continuous sedation,  112 House of Delegates Health Policy statement, 97 American Nurses Association,  96 American Society of Clinical Oncology, 249, 426 American Thoracic Society (1991) (ATS),  77, 79, 106n8 Ancient Greece, futility origins of,  66 And a Time to Die: How American Hospitals Shape the End of Life (Kaufman), 294, 296 Anderson, Tom,  284 anencephaly,  46–​47 Angell, Marcia,  374 ANH. See artificial nutrition and hydration

Annals of Internal Medicine, 236 “The Aquarium” (Hemon),  61 Aquinas, Thomas,  387–​88, 388n2 artificial nutrition and hydration (ANH),  114 of Bouvia,  269–​70 continuous sedation and,  119–​20 Aspinal, F.,  249 ATS. See American Thoracic Society autonomy, 296. See also patient autonomy bioethics revolution and,  250 of death,  390 disability and,  260 exercise of,  268 justice theories conflicting with,  251–​52 in medical decision making,  8 PASE and,  371 precedent, 261 PVS and,  262 reorienting,  331–​32 soft paternalism and,  250–​54 Awareness of Dying (Glaser and Strauss),  294, 296, 301, 305 Baby Doe Amendment,  46 Baby Doe case legal culpability from,  46 quality of life influenced by,  45–​46 trisomy21 and, 45–​46 Baby K case, influencing personhood and futility,  46–​47 baggage cultural and professional,  194 prior loss experiences causing,  204–​6 surrounding death,  194, 197–​98 Balz, J. P.,  253 Bardin, J. C.,  162 Barnato, A. E.,  244, 247 Barnes, Julian,  228 Battin, Margaret,  117 Belmont Report, 331 “Beloved Brother” case, regarding surrogate decision making,  32–​33 bereavement, embarrassment of,  228 best interests,  281 Billings, J. A.,  118 bioethics, 67, 250 BIS monitor, for consciousness,  118

Index   449 Block, S. D.,  118 BlogPulse, 279 Blumenthal, Jeremy,  268 body civilizing process of,  235 embarrassment for functions of,  234–​35 shame of,  234 social and cultural norms surrounding,  235–​36 treatment refusal and integrity of,  20 Born Alive Infant Protection Act of 2002,  47 Bosscher, Hilly,  375 Bouvia, Elizabeth,  275, 278n38 ANH of,  269–​70 critics on,  271 disability activists on,  270 on her disability,  270 self-​loathing of,  272 standard mental competence of,  269 bowel obstruction,  138 Bowers, Tim,  273 brain death criterion,  100, 292 brain injury families response to,  155 patients’ displacement with,  155 politicization of,  155 of Quinlan, K.,  155–​56 right to die and,  155–​56 traumatic, 159 Brauer, Karen,  93 Brock, Dan,  93 Brongersma court case on euthanasia,  347–​48 Buckman, Robert,  229 Bush, George W.,  102 Buskes, J. J.,  343 Bynum, J. P. W.,  229 Callahan, Daniel,  154 Campbell, Amy,  268 Canadian Society for Palliative Care Physicians, 111 cancer, nutrition and hydration during,  136 Cancer Care Outcomes Research and Surveillance Consortium,  184 capacity to appreciate,  399 assessment of,  396–​97 demonstration threshold for,  397

depression and,  398–​99 elements of,  397 CAPC. See Center to Advance Palliative Care capital punishment,  368 cardiopulmonary resuscitation (CPR),  15, 66 caregiver for childhood illness,  60 moral distress among,  62–​63 Carver, Raymond,  63 CASBI. See Consortium for the Advanced Study of Brain Injury Cassell, Eric,  2, 225 Catholic-​Christian perspective on communal responsibility,  336–​37 of compassionate care,  335 dualism and,  334 of end-​of-​life care,  330–​31, 334 Catholic Health Care Services, ethical and religious directives for,  140–​41 causation, forgoing life-​sustaining treatment and,  19, 109, 410 Cellarius, V.,  119 Centers for Medicare and Medicaid Services,  243 Center to Advance Palliative Care (CAPC), 429, 443 Central Committee on Late Terminations of Pregnancy and Life-​Ending of Neonates, 355 Chabot, B.,  357 Chabot case on euthanasia,  346 checks (SCEN), for Netherland PAD quality control,  352–​54 child abuse,  46 childbirth, 43 childhood death,  2 familial isolation and,  55 silver lining search and,  63 small good things needed after,  62–​64 childhood illness care for,  62–​63 caregiver’s experience of,  60 clinical research for,  57 familial isolation and,  55 hope needed during,  57 incurability of, 57 isolation from,  61 moral distress accompanying,  60, 62

450   Index children care for,  62–​63 death and dying in,  2 despair, hope for dying,  56–​57 life-​sustaining treatment withheld for disabled, 46 suffering of, 61 choice architecture,  252 Christakis, Nicholas,  170 Christians embodiment notion of,  333 nutrition and hydration beliefs of,  140–​41, 333 Church Amendment,  87, 102 City of Cleburne, Texas, et al. v. Cleburne Living Center et al., 271 Civil Rights Act (1964), Title VII,  105 Clark, David,  199–​200 clinical practice, court system influencing, 8, 25 clinician burnout of,  430 conflict resolution by,  79–​82 conscience-​based objections by,  91 conventional compromise by,  93 family support by,  50–​51 good stewardship promotion by,  71 interdisciplinary hospital committee review provided by, 81 medical professional integrity protected by, 70 moral distress avoidance by,  71 negotiated agreement attempt by,  80 no-​exemption approach for,  91 palliative care marketing to,  443–​44 patient autonomy respect by,  70 patient suffering avoidance by,  70 physiological futile, proscribed, and discretionary treatment refusal by,  79–​80 process notice given to surrogates by,  80 professional obligations of,  97, 99 second medical opinion requested by,  80 skepticism of,  100–​101 stereotyping by,  309 surrogate transfer offered by,  81 treatment request resistance by,  70–​7 1

Coats Amendment,  102 Code of Ethics for Nurses (American Nurses Association), 96, 98 Code of Medical Ethics (AMA),  96–​97, 111 cognitive disabilities,  275n1 concept of good and,  265–​66 decision-​making capacity and,  261–​68 euthanasia and,  355–​56 moral standing of people with,  266, 276n6 self-​governance and,  261 trisomy 21 and,  267–​68 cognitive issues, as futility dispute cause,  69 coma. See permanent vegetative state; unresponsive wakefulness syndrome comfort feeding only,  136 Committee on the Acceptability of Life-​ Ending Treatment,  348–​49 communal altruism,  56 communication as cause of futility disputes,  75 cultural norms influencing,  209–​10 deficiencies in,  229 denial influencing,  210–​11 on dying,  229–​30, 236, 199–​200 with dying patients,  202–​15 emotional tone and,  282 ethnicity influencing,  310–​11 interpreters used for,  311 LEP and,  311, 321 phrases used in,  212 race influencing,  310–​11 readiness for,  210–​11 on social media,  282–​83 toolbox for,  211–​12 communication, of prognosis cultural differences in,  180 ethical issues and,  178–​79 forcing of,  180–​81 hope relayed during,  179 individualization of,  179–​80 informed consent and,  179 median estimation used for,  182 self-​fulfilling prophecy and,  185 sensitivity of,  179 truth needed for,  182–​84 uncertainty disclosure during,  182, 183t understanding needed during,  181–​82

Index   451 communication style audience needs identified for,  209–​10 authentic,  208–​9 identification of,  208–​9 information receiving and,  210 phrases and,  212 training on,  207–​8 trial and error and,  211–​12 Compassion and Choices,  373, 376 compassionate cannibalism,  292 compassionate care call to,  335–​36 Catholic-​Christian perspective of,  335 embodiment and,  333–​34 professional-​patient relationship in,  336 suffering and,  335, 338–​39 competent patient, forgoing life-​sustaining treatment and,  30 Comprehensive NCI Cancer Center,  427 conflict management models, no-​exemption approach similar to,  92 conflict resolution extramural appeal opportunity for,  81 fair process for,  82–​83 independent review role in,  82 interdisciplinary hospital committee review for, 81 negotiated agreement attempt for,  80 prevention strategy implementation for,  79 process decision implementation for,  81–​82 request refusal and,  79–​80 second medical opinion and,  80 surrogates process notice for,  80 Conroy case,  11–​12 conscience absolutism,  91 conscience-​based objection and,  92 moral integrity and,  93 conscience-​based objection advance notification for,  95–​96 burdens from,  96 Code of Ethics for Nurses on, 98 conscience absolution and,  92 to DCDD,  96, 101 disclosure of, 95 exemptions for,  91–​102 futility approach to,  76

middle ground approach to,  93–​94 no-​exemption approach to,  91–​92 participation and,  92, 106n4 reasonable accommodation approach to,  94–​99 referrals from,  94 conscience-​based referral,  106n8 conscience clauses and,  104 patient autonomy impacted by,  97–​98 timing of,  94–​95 vulnerability and burdens from,  95 conscience-​based refusals for abortion,  103 clinician’s skepticism and,  100–​101 constraints on,  99–​101 false clinical beliefs influencing,  100 due to invidious discrimination,  99, 106n9, 106n10 moral weight of,  100 due to unethical treatment values,  99 conscience clauses for abortion,  102 conscience-​based referral,  104 end-​of-​life care and,  104–​5 explicit exemptions in,  104 law and public policy and,  102–​5 for treatment and services,  103 conscientious objection court decisions on,  87 end-​of-​life care and,  88 examples of,  88–​90 by health care professionals,  1 health care scope of,  87–​88 life-​sustaining treatment and,  87–​88 management of,  105–​6 for PAS,  89, 348 professional organizations scope of,  98 reproductive health and,  88 consciousness degree of,  118 ethical obligation to identify,  162 families recognizing evidence of,  164 pain perceived during,  155 Consortium for the Advanced Study of Brain Injury (CASBI),  155 Constitutional rights, physician aid-​in dying and, 23

452   Index continuous sedation,  2, 109 as acceptable,  111 advance directive for,  116 AMA on,  112 ANH and,  119–​20 assessment for,  118 consensus and guidelines for,  110–​11 criticism on,  117 death under,  113, 119 debates on,  118–​21 double effect doctrine and,  114–​15 in foreign countries,  113 indications for,  120 justifications commonly invoked for,  114–​18 Netherland guidelines for,  110–​11, 120 patient autonomy and,  116–​17 patient consent for,  114, 116–​17 prevalence and incidence of,  113 proportionality of,  117–​18 psychological suffering as indication for,  120–​21 research on,  112–​13 restriction of,  119 as slow euthanasia,  118 as state of art,  112–​14 for suffering,  110 terminology and definition of,  111–​12 conventional compromise,  93–​94 Convention on the Rights of Persons with Disabilities (CRPD) (United Nations), 266 The Conversation Project,  237, 238 Council on Ethical and Judicial Affairs (AMA), 89 counseling, acute neonatal illness and,  44–​45 court system on actively hastening death,  21 on advance directives,  16 clinical practice influenced by,  8, 25 decision-​making capacity determined by, 11 on DNR orders,  15 end-​of-​life controversies and,  7–​8, 10 end-​of-​life decision-​making process and,  7 legal liability immunity granted by,  7–​8 statutes enacted by,  10 CPR. See cardiopulmonary resuscitation

criminal homicide,  21 Crippen, D.,  244 critical interests,  276n7 end-​of-​life decisions and,  264 Jaworska on,  265 valuing and,  264–​65 cross-​cultural care, strategies for,  318–​19t CRPD. See Convention on the Rights of Persons with Disabilities CRS-​R. See JFK-​Johnson Coma Recovery Scale-​Revised Cruzan, Joyce,  30 Cruzan, Lester,  30 Cruzan, Nancy,  8, 14, 28 Cruzan lawsuit,  155 advance directive movement from,  31 evidentiary standard from,  31 liberty interests and,  36 culpability.  See legal liability culture communication influenced by,  209–​10 death and factors of,  291–​92, 295–​296, 298–​99 dimensions of,  418 of dying,  228, 236–​37 individuals influenced by,  293–​94 palliative care impacted by,  418 problematic,  292–​94 self-​actualization promoted by,  301 shifting of,  293 taken-​for-​granted character of,  293 understanding of,  418 visibility of,  293 Dallimore, D.,  203 Daniels, N.,  251 The Dartmouth Atlas of Health Care,  242–​43, 244 DCDD. See donation after circulatory determination of death death. See also actively hastening death; childhood death; dying; good death; hasten death acceptance of,  298, 302–​5 active intervention for,  21 allowing of,  132–​33, 144 anxiety of,  231–​32, 410

Index   453 burden of,  297 choice of,  401–​2 coming to terms with,  193–​200 confrontation of,  196–​97 control over,  300–​2, 199 declaration of,  300 de-​emphasizing of,  429 deep shadows surrounding,  195–​96 with dignity,  344, 374 drawn-​out forms of,  303 exclusion of,  429–​30 fear of,  196 global scale of,  416–​17 intent of,  19–​20 location of,  295 meaning of,  291 next phase of,  198–​200 nothing more to be done before,  302–​5 ownership of,  301–​2, 390 problematic,  291–​92 proximity to,  401 recognition of,  292 relocation during,  303–​4 responsibility of,  301–​2 silence regarding,  296–​97 social movement towards accepting of, 237 social silence surrounding,  295 as taboo,  297–​98, 411, 430 technology and,  297, 299–​300, 304 timing of,  304–​5 variations of,  298, 389–​90 waiting for,  302–​3 “Death and Dignity: A Case of Individualized Decision Making” (New England Journal of Medicine), 22 Death and Dying Group,  29 Death Café, 237, 238 death-​denying society,  194 Death Over Dinner,  237, 238 death-​postponing devices,  7 deathwatch, 303, 305 death with dignity,  344, 374 Death with Dignity Acts,  374–​75 death without weeping,  292 “Deciding to Forego Life-​Sustaining Treatment” report,  12, 30

decision architecture,  254 decision-​making. See also surrogate decision making current framework for,  281–​84 deliberative, not snap judgments for,  274 dementia and,  263–​64 during parental grief and loss,  60–​61 prenatal diagnosis and,  43–​44, 52 prognosis and,  184–​85 supported,  267, 274–​75 uncertainty and,  59 decision-​making capacity,  3 advance directives and,  15 cognitive disabilities and,  261–​68 court system determining,  11 in ICU,  68 of patients,  30 delirium, 395 dementia, 395 critical interest and,  264 decision-​making and,  263–​64 precedent autonomy and,  263 present preferences and,  263 valuing patients with,  263–​64 de Montaigne, Michel,  227, 238, 238n1 demoralization syndrome, depression and, 396, 399 denial communication influenced by,  210–​11 of death,  4, 296–​97, 299 parental grief and loss and,  57–​58 depression, 3 appreciation ability with,  399, 402 capacity and,  398–​399 death as desirable state with,  391, 400–​1 demoralization syndrome and,  396, 399 differentiation of,  395 feeling v. having,  392, 395 grief and,  395 life-​shortening decisions influenced by, 391 treatment for,  400 vignette on,  391–​92, 403–​4 despair, tension between hope and,  56–​57 Diagnostic and Statistical Manual (DSM), 392 Dijkhuis Committee,  348

454   Index disability. See also cognitive disabilities; noncognitive disabilities activists’ dismissive analysis for,  269 antipathy toward,  270–​7 1 autonomy and,  260 Bouvia on,  270 presumptions about,  260 sudden, 273 supported decision-​making and,  267 disabled population, end-​of-​life care preferences in, 3 discretionary treatment,  79–​80 disorders of consciousness (DOC),  156 active language paradigm for,  162 behavioral criteria diagnosis of,  162 diagnostic assessment of,  158–​59 diagnostic challenge of,  156–​57 differential diagnosis of,  157–​58 discharge diagnosis for,  161 evolution of,  158–​61 family intervention for,  162–​63 moral obligations for,  161 neuro-​palliative care for,  163–​66 recovery trajectory for,  160 timeframe of,  161 disproportionate treatment,  73 distributive justice approach, to futility,  76–​77 distrust, 69 DNR. See do-​not-​resuscitate orders DOC. See disorders of consciousness donation after circulatory determination of death (DCDD),  88, 96, 101 do-​not-​resuscitate (DNR) orders,  67 acceptance of,  14–​15 court system on,  15 family member disputes regarding,  15 futility and,  15 double effect doctrine,  25 application of,  115 continuous sedation and,  114–​15 as ethical justification,  115 legal basis of,  24, 115 moral culpability and,  24 of palliative care,  23–​24 Douglas, Mary,  235 Down’s syndrome. See trisomy  21 dramatis personae,  298

Dresser, Rebecca,  92 DSM. See Diagnostic and Statistical Manual dualism Catholic-​Christian perspective and,  334 end-​of-​life care and,  333 medical, technology contributing to,  334 durable power of attorney,  16 Dutch Hoge Raad (HR),  346 Dutch Physicians Union,  348 Dutch Society of Physicians for the Mentally Handicapped,  355–​56 Dutch Voluntary Euthanasia Society (NVVE),  346, 358–​59 Dworkin, Ronald,  264 dying. See also physician aid-​in dying advice books on speaking about,  229 in American hospitals,  294–​796 awareness of,  297–​98 awkwardness, embarrassment, and withdrawal regarding,  228–​29, 232–​36 care given during,  370 childhood fears rekindled with,  231 clinical approach to,  195 cost of,  242–​54 difficulties explained regarding,  230–​31 ethnographic study of,  294–​95 feelings of anticipation for,  233 hospice and palliative care for infants,  49 ideology of hopefulness,  231 individual emotions and behaviors regarding,  232–​33 process of,  295, 298 responsibility during,  298 shame regarding,  232–​36 shifting dilemmas of,  296–​305 as social category,  411 social discourse surrounding,  230 “Dying in America” (Institute of Medicine),  425 Earle, C. C.,  230 easeful death,  411 EC. See emergency contraception Economist Intelligence Unit,  417 Eisenberg, Nancy,  232 elderly population cost of dying among,  242–​54 end-​of-​life care preferences in,  3

Index   455 Medicare funds for,  244 spending by,  245 Elias, Norbert,  230, 235 emancipation, 343 Emanuel, E. J.,  246 Emanuel, L. L.,  246 embarrassment of bereavement,  228 of body functions,  234–​35 for dying,  228–​29 of dying body,  234 embodiment Christian notion of,  333 communal compassionate care and,  333–​34 end-​of-​life care and,  333–​34 emergency contraception (EC),  100 Emergency Medical Treatment and Active Labor Act (1986),  47 emotional issues, as futility dispute cause,  69 Enck, R. E.,  111 end-​of-​life care,  410. See also nutrition and hydration, at end of life advances in,  48–​50 African Americans and,  312, 315–​16 aggressive treatment in,  230 Catholic-​Christian perspective on,  330–​31, 334 communal dimension of,  336–​37 conscience clause and,  104–​5 conscientious objection and,  88 cost of, in US,  243–​45 cultural norm for,  252–​53 dualism and,  333 embodiment and,  333–​334 extremes models of,  332 financial costs of,  245 health literacy influencing,  315 Hispanics and,  315–​16 ICU and,  229 multidimensional approach to,  332 negative reactions to,  421–​22, 426 poor policymaking for,  419–​10 prognosis influencing,  171, 173t reframing care in,  330–​39 religion and,  330–​39 research on,  419–​20 SUPPORT on,  425

end-​of-​life care preferences good death and,  245–​48 variations of,  247 end-​of-​life controversies court system and,  7–​8, 10 legal uncertainty around,  9 media attention of,  8 passive vs. active hastening death as,  20–​21 end-​of-​life decision(s) active versus passive,  13 court system and,  7 experiential, critical interests and,  264 high-​stakes of,  400–​1 after learning of prognosis,  184–​85 legitimacy of, 12 from living will,  17 ordinary versus extraordinary,  13 passively versus actively hastening death and, 21 Presidential Commission on,  8 withholding versus withdrawing,  13 end-​of-​life ideation,  392 end-​of-​life law,  7, 11–​12, 18–​19 ERC. See Euthanasia Review Committee ESICM. See European Society of Intensive Care Ethical and Religious Directives for Catholic Health Care Services, 334, 336 ethical epistemic modesty,  90–​91 ethics committee,  10, 30, 372 ethnicity communication influenced by,  310–​11 cultural beliefs, values and preferences,  314–​15 cultural considerations with,  308–​9 future directions of,  322–​23 health care access influenced by,  51 health care professionals navigating encounters and,  317–​18 health care quality disparities and,  309–​10 health care system and,  322–​23 infant mortality influenced by,  51 pain management influenced by,  311 research and policy on,  323 European Association for Palliative Care, 111, 354

456   Index European Society of Intensive Care (ESICM), 79 euthanasia,  3, 24, 110, 118. See also physician-​ assisted suicide acceptable, 358 Alzheimer’s disease and,  363 Brongersma case on,  347–​48 Chabot case on,  346 for cognitive disabilities,  355–​56 court case on,  344–​47 definition of,  350 goals of,  362–​63 increase in,  357–​58 legal angles of,  345–​46 palliative care and,  356 Postma case of,  345–​46 for psychiatric patients,  347 PVS and,  356 Schoonheim case on,  346 suffering relieved by,  346–​48 Euthanasia Law,  345 Euthanasia Review Committee (ERC), 348, 353 Alzheimer’s patients and,  361 cases reported to,  358–​59 for Netherland PAD quality control,  352–​54 NVVE and,  358–​59 psychiatric patients and,  360–​61 reporting from,  353 request approval by,  360–​62 evidentiary standard Cruzan lawsuit influencing,  31 for life-​sustaining treatment,  30–​31 exercise of conscience,  90–​91 extramural appeal,  81 Facebook,  280, 287, 288n11 metric tool for,  282, 289n26 privacy of,  284, 289n34 for Schiavo,  282 Facing Death, 389 Fainsinger, R. L.,  110 families African American’s importance on,  312 brain injury response by,  155 childhood illness and death isolation for,  55 clinician’s support for,  50–​51

consciousness recognized by,  164 counseling for,  50 DOC intervention by,  162–​63 Hispanics importance on,  312 psychological burden in,  162 recovery expectations from,  164 family member authority of, 16 DNR order disputes among,  15 support of, 49 surrogate decision making role forced upon,  33–​34 Farley, Margaret,  335 fear, types of,  196 federal funding, abortion and,  87 feeding tube Cruzan and,  14 as medical treatment,  14 public opinion on,  30 for Quinlan, K.,  13–​14 special concerns raised on,  142 Supreme Court on,  14 withdrawal of, 30 fetal surgery,  43 fetal treatment, prenatal diagnosis compared to, 43 fetus, legal personhood lack and,  44 Final Exit Network,  376 Final Rule,  102 financial cost of aggressive treatment,  242, 245 of dying elderly population,  242–​54 of end-​of-​life care,  245 limitations of,  244 as quality of life consideration,  244 Finlay, I.,  203 Fins, Joseph,  154 fMRI. See functional magnetic resonance imaging force feeding,  144–​45 Foreman, Vicki,  58 forgoing life-​sustaining treatment acceptance of, 18 competent patients rights to,  30 conscientious objection and,  88 controversy of, 21 evidentiary standards for,  30–​31

Index   457 intent of,  19–​20 legal liability and,  19, 25 refusal right,  20 functional magnetic resonance imaging (fMRI), 162 fundraising, hope from,  57 futility,  17–​18 AACN on, 79 ACCP on, 79 AMA on, 77 American College of Physicians on,  78 Ancient Greece origins of,  66 ATS on,  77, 79 Baby K case influencing,  46–​47 concept of, 66 conflict resolution for,  79–​82 conscience-​based objection approach to, 76 debate history of,  65–​66 distributive justice approach to,  76–​77 DNR orders and,  15 ESICM on, 79 General Medical Council on,  78 imminent demise,  72 legal condition,  73 of life-​sustaining treatment,  18 major positions on,  71–​77 of mechanical ventilation,  46–​47 medical ineffectiveness approach to,  72–​73 medical power and,  2 physician power and,  65 physiological definitional approach to,  72 policy statements on,  77–​79 procedural process-​based approach to,  74–​75 qualitative futility approach to,  73–​74 quantitative definitional approach to,  73 SCCM on,  77, 79 surrogate selection approach,  76 treatment definition for,  74 treat ‘til transfer clause and,  76 of Wanglie case,  18 futility disputes causes of,  68–​7 1 cognitive issues as cause of,  69 communication as cause of,  75 distrust as cause of,  69

prevalence of, 68 psychological and emotional issues as cause of, 69 religion and miracles as cause of,  69–​70 treatment definition for,  74 understanding of,  71 Gade, G.,  248 Ganzini, L.,  400 Garnier, Jeanne,  411 GCS. See Glasgow Coma Scale GDP. See gross domestic product Geertz, Clifford,  237, 293 Gelpi, A. H.,  251 generalist palliative care (GPC) critique on,  434–​35 deskilling of,  430–​32 hospital presence required for,  427–​28 population health view of,  433–​35 referrals for,  432 SPC compared to,  426–​27, 430–​31 General Medical Council (2010),  78 George Soros’s Open Society Foundation,  425 Ghormley, C.,  399 Giacino, J. T.,  159 Glaser, Barney,  294, 297, 300–​2, 305 on good death,  295 on proper death,  298 Glasgow Coma Scale (GCS),  159, 356 Glasgow Outcomes Scales (GOS),  159 Goffman, Erving,  233 Golubchuk, Samuel,  69 good death,  292, 294, 410, 415 as dignified,  301 end-​of-​life care preferences and,  245–​248 expectations of,  206–​7 Glaser on,  295 in health care system,  246 Kaufman on,  295 notion of,  305, 389–​90 script for,  298–​300 Strauss on,  295 good stewardship promotion,  71 GOS. See Glasgow Outcomes Scales Gozalo, P. L.,  229 GPC. See generalist palliative care Greenberg, Jeff,  231

458   Index grief. See also parental grief and loss depression and,  395 of health care professionals,  203, 212–​15 identification of,  203–​4 loss types invoking,  203 profound,  58–​59 resolution for,  212–​15 A Grief Observed (Lewis),  228 Griffiths, Roland,  379 Grob, Charles,  379 gross domestic product (GDP),  243 A Guideline on Palliative Sedation, 349 Guidelines on Euthanasia and Reduced Consciousness, 349 Halifax, Joan,  232 Halpern, John,  378 happy slaves,  270–​7 1 Harvard, 433 hasten death depression and,  400–​1 desire for,  400–​1 posture and process for,  401–​2, 402t The Hastings Center,  29 Hastings Center Report, 162 Hatch, Jenny,  267, 275 Hawthorne, Rose,  412 Health Affairs, 248 Health and Human Services (HHS), US Department of,  102, 322 health care access LEP and,  311 race and ethnicity influencing,  51 systemic barriers to,  51 health care professionals conscientious objection by,  1 cross-​cultural encounters and,  317 cultural sensitivity lacking,  317 debriefing for,  213 grief experienced by,  203, 212–​15 hospice and palliative care services discussion,  320–​21 LEP and low health literacy care provided by, 321 stereotyping avoidance approach of,  319–​20 health care system code status required by,  15

cost reduction in,  254 end-​of-​life passage through,  20 ethnic diversity among,  322–​23 financial cost of,  244 good death and,  246 MCS and,  160–​61 mistrust in,  317 physician limitations within,  367–​68 quality disparities in,  309–​10 resource allocation of,  251 spending on,  251 health literacy,  314 end-​of-​life care influenced by,  315 health care professionals caring for patients with low,  321 health-​promoting palliative care,  420 The Hemlock Society,  373 Hemon, Alexander,  57, 59, 61 Hepler, C. D.,  90 HHS. See Health and Human Services, US Department of Hinds, P. S.,  61 Hippocrates, 66, 171 Hispanics advance care planning by,  312 aggressive treatment for,  247–​48, 316 cultural beliefs, values and preferences of,  314–​15 culture of,  309 end-​of-​life care and,  315–​16 family importance of,  312 health care system mistrust and,  317 hospice and palliative care services use,  313–​14 LEP of,  311 pain management of,  311 spiritual beliefs of,  316 Holm, S.,  120 hope from fundraising,  57 preservation of, 57 during prognosis communication,  179 tension between despair and,  56–​57 Horder, Thomas,  66 hospice acceptance of,  413 in affluent and poor countries,  413 African American’s use of,  313–​14

Index   459 birth of,  296, 411, 412 care focus in,  414 contested space of,  420–​22 criticisms on,  414 for dying infants,  49 health care professional discussion on,  320–​21 Hispanic’s use of,  313–​14 idea of,  412–​13 international patterns within,  416–​17 language adopted from,  442 limitations of, 50 modern movement of,  440 palliative care and,  409–​22 prognosis influencing,  171 religious, philanthropic and moral concerns of,  412–​13 shame association with,  234 spiritual beliefs on,  316 House of Delegates Heath Policy,  97, 106n6 How to Be a Friend to a Friend Who’s Sick (Pogrebin), 229 “How to Handle Patients with Terminal Illness,” 194 HR. See Dutch Hoge Raad humility,  90–​91 hydration. See nutrition and hydration Hyperalgesia, 48 hypodermoclysis,  130–​31 ICU. See intensive care unit I Don’t Know What to Say (Buckman),  229 imminent demise futility,  72 independent review role,  82 infants acute suffering for,  45 hospice for dying,  49 individualized care for,  50 medical advances for premature,  47 mortality of,  51, 292 nutrition and hydration withholding for dying, 49 pain perception of,  48 quality of life consideration for,  46, 48 suffering of,  45, 47 infant survival,  1–​2, 42 gains in,  44–​45

improvement of, 48 physiologic and neurologic damage from, 45 pulmonary insufficiency and,  44 underestimation of, 50 informed consent,  29, 36 Code of Medical Ethics on, 97 improvement of,  250–​51 prognosis communication and,  179 surrogate decision maker role and,  37 Institute of Medicine,  246, 425 The Institute of Medicine,  50 intensive care unit (ICU) decision-​making capacity in,  68 end-​of-​life care and,  229 futility in, 65 length of stay in,  44 quality of life in,  48 intent consequences of, 20 forgoing life-​sustaining treatment and,  19–​20 interdisciplinary hospital committee,  81 interests, experiential and critical dementia and,  264 valuing and,  264–​65 Internal Review Board (IRB),  154 intravenous hydration and nutrition,  129 invasive care, continuance of,  253 invidious discrimination, conscience-​based refusals due to,  99, 106n9, 106n10 IRB. See Internal Review Board isolation from childhood illness,  61 familial, 55 from MCS,  165–​66 within neuro-​palliative care,  165–​66 parental grief and loss and,  61 Jacobson v. Massachusetts (1905),  386 Jaworska, Agnieszka,  263, 265 Jecker, Nancy S.,  74 jejunostomy tube feeding,  128–​29 Jennett, B.,  157 JFK-​Johnson Coma Recovery Scale-​Revised (CRS-​R),  159 Journal of Clinical Oncology, 427

460   Index Journal of the American Medical Association, 22, 427 justice distributive approach, to futility,  76–​77 theories of, conflicting with autonomy,  251–​52 Juyi, Bai,  59 Kaszniak, Alfred,  232 Kaufman, Sharon,  294, 304–​5 on good death,  295, 298–​300 Keenan, James,  334, 337 Kevorkian, Jack,  22, 441 killing PASE and,  367 physicians oath to not,  368, 369–​70 prohibition on,  367 kindness of strangers,  2, 55–​56 Kissane, David,  234, 399 KNMG. See Koninklijke Nederlandse Maatschappij tot bevordering der Geneeskunst Kohn, Nina,  268 Koninklijke Nederlandse Maatschappij tot bevordering der Geneeskunst (KNMG), 344, 361 Kubelbeck, Amy,  58 Kübler-​Ross, E.,  195, 199, 343 Lammi, M. H.,  160 L’Association des Dames du Calvaire,  412 Laureys, Stephen,  157 Law on Euthanasia,  349 Lawton, Julia,  234 LEC. See Life Ending Clinic Leeuwarden Court,  347 legal condition futility,  73 legal liability from Baby Doe case,  46 causation influencing,  19 fear of,  7–​8 intent influencing,  19 life-​sustaining treatment and,  19 treatment refusal right and,  20 legislation abortions and,  87, 103 advance directives supported by,  17

LEP. See limited English proficiency Lessons for the Living (Kübler-​Ross),  343 lethal injection. See active euthanasia “Letter of Consolation to His Wife” (Plutarch), 227 Levy Criteria,  159 Lewis, C. S.,  228, 236 “Liberalism and Individually Scripted Ideas of the Good: Meeting the Challenge of Dependent Agency,”  265 liberty interests,  36 Life Ending Clinic (LEC),  358, 359, 363 life expectancy, estimation of,  119 life-​long cognitive disabilities,  262, 263 life preservation,  12, 87 life-​shortening decisions, depression influencing, 391 life-​sustaining treatment. See also forgoing life-​sustaining treatment for children with disabilities,  46 conscientious objection and,  87–​88 futility of, 18 informed consent and,  29 introduction of,  66–​67 legal requirement of,  17 limitation rejection on forgoing,  11–​12 medical ineffectiveness of,  72–​73 minimum goal of,  74 moral obligations for,  132 physician’s choices regarding,  18 withholding and withdrawing view on,  28–​29 life-​threatening illness, newborns with, aggressive treatment support for,  45 life worth living,  42, 45–​47 Lifton, Robert Jay,  231 limited English proficiency (LEP),  311, 321 Lindemann, Hilde,  266 Liverpool Care Pathway,  419 living-​dying interval,  59, 63 living will as advance directives,  15–​16 end-​of-​life decisions from,  17 The Loneliness of the Dying (Elias),  230 Major Depressive Disorder (MDD),  392, 393–​94t March of Dimes,  57

Index   461 MBPSL. See medical behavior that potentially shortens life McAfee, Larry,  272, 275 MCS. See minimally conscious state MDD. See Major Depressive Disorder mechanical ventilation, futility of,  46–​47 media, end-​of-​life controversies and,  8 Medicaid funds, PSDA influencing,  31–​32 medical behavior that potentially shortens life (MBPSL), 344 medical futility.  See futility medical ineffective treatment,  72–​73 medically provided nutrition and hydration (MPNH), 91 as adjunctive,  131 burden of,  132 conscience-​based approach to,  96 critical decisional junctures for,  131 ethics of,  126–​47 as life-​prolonging therapy,  131 as moral mistake,  137 moral obligations and,  132 as palliative in agonal phase,  131, 137–​39 as palliative in terminal phase,  131, 134–​37 medically provided nutrition and hydration treatment modalities,  127–​31 hypodermoclysis,  130–​31 intravenous hydration and nutrition,  129 jejunostomy tube feeding,  128–​29 NG tube feeding,  127–​28 PEG tube feeding,  128 PPN, 130 proctoclysis, 129 surgical gastrostomy tube feeding,  128–​29 TPN,  129–​30 medical power, futility and,  2 medical profession integrity protection for,  70 safeguarding integrity of,  12 Medical Student Association,  98 Medicare funds for elderly Americans,  244 expenditures of,  244–​45 PSDA influencing,  17, 31–​32 Medicare Hospice Benefit,  440 Meert, K. L.,  62 mental retardation,  265

mercy, 338 mercy killings.  See physician aid-​in dying Millennium Development Goals,  419 minimally conscious state (MCS) behavioral, 162 diagnostic challenge of,  156–​57 emergence from,  159 health care system and,  160–​61 isolation from,  165–​66 language response in,  158, 162 nonbehavioral, 162 pain management in,  165 PVS compared to,  157–​58 from TBI,  159 transition to,  159, 160 miracles, as futility dispute cause,  69–​70 model-​conscience protection policy and statute, 92 modernity, paradox of,  227–​30, 234 Monti, M. M.,  162 Moore, Lorrie,  57, 59, 61 moral agents,  93, 266, 276n6 moral beliefs,  88, 106n1, 367 moral culpability actively hastening death and,  23–​24 double effect and,  24 moral distress avoidance of, 71 among caregivers,  62–​63 childhood illness accompanied by,  60, 62 moral integrity conscience absolution and,  93 loss of, 90 PASE weakening,  372 protection of, 90 moral obligation for MCS,  161 MPNH and,  132 of patients,  131–​32 to provide treatment,  105 Morrison, Sean,  4, 427 Moyers, Bill,  390 MPNH. See medically provided nutrition and hydration Multi-​Society Task Force,  159 Munk, William,  411 MySpace, 284

462   Index nasogastric (NG) tube feeding,  127–​28 National Council of Commissioners on Uniform State Laws,  16–​17 National Hospice Study,  440 National Institute of Health State-​of-​the-​ Science Conference Statement,  415 National Institute of Standards and Technology, 285 National Priorities Partnership,  249 natural death,  417 The Nature of Suffering (Cassell),  225 Nederlands Artsen Verbond,  348 Neighbourhood Networks in Palliative Care, 413 neonatal care, palliative sedation and,  355 neonatal intensive care unit (NICU),  42 length of stay in,  44 outcomes of, 51 quality of life in,  48–​49 resource allocation for,  51 Netherlands cognitive disorders and euthanasia,  355–​56 coma patients and euthanasia,  356 continuous sedation guidelines and,  110–​11, 120 euthanasia goals of,  362–​63 neonatal care and palliative sedation in,  355 palliative care and terminal sedation in,  356–​55 SED and,  356–​57 Netherlands, PAD in,  3, 343–​63 for Alzheimer’s patients,  352 careful practice requirement and,  353–​54 climate of,  343 as doctor-​centered law,  351–​52 ERCs and,  360–​61 functionalization of,  359 guidelines for,  349 increase in,  350 LEC and,  359 legal development of,  344–​48 medical decisions with,  351t medical profession acceptance to integration of,  348–​49 new tensions surrounding,  357–​62 NVVE on,  358–​59

political acceptance and regulation for,  349–​51 political stalemate in,  350 quality control,  352–​54 reflections and conclusion with,  362–​63 regulations on,  344 request variations and,  350–​51, 353 research on,  353–​54 review committees for,  351–​54 social acceptance of,  344 special groups and issues with,  354–​57 Uit Vrije Wil group assisting with,  359–​60 neuroethics,  161–​62 neuroimaging,  161–​62 neuro-​palliative care,  156 clinical challenges in,  162–​65 for DOC,  163–​66 family systems and,  162–​63 isolation within,  165–​66 pain and symptom management in,  165 suffering and,  164 New England Journal of Medicine,  22, 225, 373, 427 New Jersey Supreme Court,  9, 11–​12, 28 New York Times,  92, 426, 433 NG. See nasogastric tube feeding NICU. See neonatal intensive care unit no-​exemption approach,  106n3 for clinicians,  91 conflict management models similar to, 92 to conscience-​based objection,  91–​92 for pharmacists,  92 noncognitive disabilities death decision by patients with,  268–​74 self-​governance and,  269 nontreatment decisions (NTDs),  350 nutrition and hydration, at end of life adjunctive therapeutic use of,  133 advance care planning regarding,  143 alternatives for,  136 during Alzheimer’s disease,  134–​35 as basic care,  126, 136–​37, 147 bowel obstruction and,  138 during cancer,  136 Christians beliefs of,  140–​41, 333 comfort feeding only and,  136

Index   463 dying infants withholding of,  49 force feeding and,  144–​45 as life-​prolonging therapy,  133–​34 Orthodox Jews beliefs of,  140 as palliative care in agonal phase of disease,  137–​39 as palliative care in terminal phase of disease,  134–​37 PAS and,  145–​46 patient’s voluntarily refusal for,  143–​46, 148n2 religious issues and,  139–​41 Roman Catholic Church beliefs on,  140–​41 special considerations for,  139–​46 symbolism of,  135 during terminal delirium,  138–​39 withholding versus withdrawing of,  133, 141–​43, 147n1 NVVE. See Dutch Voluntary Euthanasia Society Obama, Barack,  102 O’Connor, Sandra Day (Justice),  14, 110 Oe, Kenzaburo,  60, 61 On Civility in Children, 235 On Death and Dying: What the Dying Have to Teach Doctors, Nurses, Clergy and Their Own Families (Kübler-​Ross),  195, 343 online identity,  284 On Our Own Terms (Moyers),  390 Ontario Consent and Capacity Board Approach, 76 Open Society Institute’s Project on Death in America (2004),  228, 441 opioids, for continuous sedation,  113 ordinary versus extraordinary end-​of-​life decisions, 13 O’Rourke, K.,  333, 338 Orthodox Jews, nutrition and hydration beliefs of,  140 Our Lady’s Hospice for the Dying,  412 PAD. See physician-​assisted death pain management of African Americans and Hispanics,  311 in death,  109 ethnicity and race influencing,  311

in MCS,  165 in neuro-​palliative care,  165 in palliative care,  110 pain perception of infants,  48 in PVS,  157 palliative care. See also generalist palliative care; neuro-​palliative care; specialty palliative care acceptance of,  419 actively hastening,  18–​24 African American’s and Hispanic’s use of,  313–​14 benefits of,  249–​50, 253, 420 concerns regarding,  421 consumerist orientation in,  200, 416 contested space of,  420–​22 cost improvement method for,  248–​50 cost saving strategies for,  248 culture impact on,  418 default offering of,  253–​54 definition of,  414–​16, 421 development of,  417, 441 double effect of,  23–​24 for dying infants,  49 efficacy of,  419 enhancing access to,  441–​42 ethnicity and access to,  3 euthanasia and,  354 expansion of,  415 ghettoization of,  430–​32 goals of,  415–​16 health care professional discussion on,  320–​21 health-​promoting,  420 hospice and,  409–​22 as human rights issue,  420, 427 improvement movement for,  22 increased need for,  426 information disclosed by physicians regarding, 172 as integrated component of medical practice, 416 marketing approaches to,  429–​30, 444 national system for,  249 in Netherlands,  354–​55 paradigm shift of,  422 pediatric,  61–​62

464   Index palliative care (Cont.) pharmaceutical agents used in,  23 physician shortage for,  428 prognosis importance for,  170, 172–​75 public support for,  443 quality of care improvement method for,  248–​50 reframing of,  415 research on,  419–​20, 422, 434 specialty,  414–​15 symptomatologists in,  421 UCSF program development for,  432 WHO on,  415 Palliative Care Information and Counseling Act, 172, 186n1 palliative care marketing to clinicians,  443–​44 conclusion on,  445 enhancement from,  441–​42 introduction to,  439 modern hospice movement and,  440 PalliativeDoctors.org, 280 palliative medicine,  409, 420 palliative sedation,  111, 355 parental grief and loss decision-​making during,  60–​61 denial and,  57–​58 isolation and,  61 profound grief and,  58–​59 role change during,  61–​62 uncertainty and,  59–​60 as visceral process,  58 parent decision making,  45 partial parenteral nutrition (PPN),  130 PAS. See physician-​assisted suicide PASE. See physician-​assisted suicide and euthanasia passively hastening death,  20–​21 Pastrana, T.,  421 paternalism, soft, autonomy and,  250–​54 patient autonomy,  1, 3, 66 conscience-​based referrals impacting,  97–​98 continuous sedation and,  116–​17 in PAS,  331–​32 principle of, 97 respect of, 70

rise of, 67 Patient Protection and Affordable Care Act,  102–​3, 243, 252 patient rights negative to positive,  67–​68 rise of,  28–​32 surrogate decision maker role and,  37 surrogate decision making impacting,  32 patients. See also competent patient best interest of,  11 continuous sedation consent from,  114, 116–​17 death controlled by,  298 as final actor,  24 goals of,  209, 429–​30 moral disapproval of,  99 moral obligations of,  131–​32 respect and honor for,  37, 38 SPC care improvement of,  433–​34 suffering of,  20–​21 support before death,  197–​98 patient self-​determination,  97 Patient Self-​Determination Act (PSDA),  17, 31–​32 patient wishes evidentiary standard for,  30 prediction of, 37 probable compared to actual,  11 Pattison, E.,  195–​96 pediatric palliative care,  61–​62 PEG. See percutaneous endoscopic gastrostomy tube feeding “People Like That Are the Only People Here” (Moore), 57 percutaneous endoscopic gastrostomy (PEG) tube feeding,  128 permanent vegetative state (PVS),  140 autonomy and,  262 euthanasia and,  356 as functionally disconnected,  157 MCS compared to,  157–​58 pain response in,  157 prolonging life in,  333 person, concept of complexity of,  218 suffering and,  217–​18 A Personal Matter (Oe),  60, 61

Index   465 personhood agency and,  266 Baby K case influencing,  46–​47 PET. See positron emission tomography Pew Research Center (2011),  280 Pew Research Center (2013),  247 pharmaceutical agents, used in palliative care, 23 pharmacists conventional compromise by,  93 no-​exemption approach for,  92 Pharmacists for Life,  93 Pharmacists for Life International Model Conscience Clause,  92 physician aid-​in dying abuse of, 22 Constitutional rights and,  23 definition of,  415, 421 legalization of,  22–​23 as morally permissible,  383–​88 patient as final actor in,  24 pharmaco-​assisted therapy for,  379 political traction for,  22 public debate on,  22 regulation of, 22 for terminally ill persons,  23 physician-​assisted death (PAD),  3. See also Netherlands, PAD in SED and,  356–​57 physician-​assisted suicide (PAS),  103, 144 abuse potential,  384, 386 Callahan on,  154 case against,  383–​84 case for,  384–​85 conscientious objection for,  89, 348 definition of,  350, 377 disagreements over,  369–​70, 379 Dutch guideline on,  110–​11 guidelines and regulations for,  371 as killing,  383, 385 legalization of,  110–​11, 375 legal recognition of,  1 liberation from,  384 nutrition and hydration and,  145–​46 patient autonomy and,  331–​32 physician integrity impacted by,  383–​84, 385–​86

pursuit of happiness and,  384 refusal to provide,  89 right for,  387 values challenged by,  370 physician-​assisted suicide and euthanasia (PASE) autonomy and,  371 case against,  366–​79 coercion and,  374 debate over,  372, 376–​77 ethical points regarding,  369 ethics committee and,  372 as flawed, damaging, and dangerous,  372–​76 good intentions and,  368–​69 killing, 367 moral beliefs and,  367 moral integrity weakened by,  372 need for,  373 for non-​terminally ill patients,  375–​76 opposition to,  368–​69, 376–​77 principles and pragmatism and,  370–​72 public confusion on,  373–​74 rights for,  373 slippery slope of,  374–​76 suffering alleviated by,  371, 377–​79 support for,  368–​7 1 war and fighting influencing,  371–​72 physician power erosion of,  28–​32 futility and,  65 tectonic shift of,  32 treatment recommendation and,  35 physician protection, from Act 69,  105 physicians debriefing of,  213 emotional boundaries of,  203 good intentions of,  368–​69 grief work by,  212–​215 limitations of,  367–​68 oath to not kill by,  368, 369–​70 palliative care information disclosed by,  172 PAS impacting integrity of,  383–​84, 385–​86 patient’s death impacting,  202 power of,  301 prognostication inability for,  174–​75 refuse to provide,  105

466   Index physicians (Cont.) role-​specific feelings constrained by,  231 self-​care for,  213–​15 shortage of, providing palliative care,  428 suffering impacted by,  225 vulnerability of,  204 physiological futility,  72 “A Piece of My Mind: It’s Over Debbie” (Journal of the American Medical Association), 22 PiPS. See Prognosis in Palliative care Study Pittsburgh Post-​Gazette, 433 Plum, Fred,  155, 157 Plutarch, 227 Pogrebin, Letty Cottin,  229 positron emission tomography (PET),  157 Posner, J. B.,  158 Postma-​van Boven euthanasia case,  345–​46 posttraumatic stress disorder (PTSD), from surrogate decision making,  34 posture and process, for hastening death,  401–​2, 402t power of shame,  233 PPN. See partial parenteral nutrition precedent autonomy,  261, 275n1 dementia and,  263 life-​long cognitive disabilities and,  263 present preferences and,  263–​264 premature infant expectations of,  50–​51 increase incidence of,  50 long-​term problems of,  50 medical advances for caring for,  47 medical interventions for,  51 nonresuscitation of, 51 prenatal diagnosis decision making and,  43–​44, 52 fetal treatment lagging compared to,  43 political and legal efforts surrounding, 44 prenatal fetal screening,  43 present preferences dementia and,  263 precedent autonomy and,  263–​64 preservation of life,  12 Presidential Commission,  8, 12, 25

procedural process-​based approach, to futility,  74–​75 proctoclysis, 129 prognosis. See also communication, of prognosis accurate estimation obligation for,  173 chronic incurable illness burden and,  171 clinical act of,  171 in clinical research,  174 as clinical skill,  170 decision-​making and,  184–​85 end-​of-​life care influenced by,  171, 173t end-​of-​life decisions made after,  184–​85 estimation of,  175–​77 historical perspective on,  171–​72 hospice influenced by,  171 imprecision and uncertainty of,  177 inaccuracy of,  175–​76 life choices hinging on,  174, 174t overly optimistic estimation for,  176 in palliative care,  170, 172–​75 for quality of life,  173–​74 Prognosis in Palliative care Study (PiPS),  176 prognostication challenge of,  185–​86 norms of,  175 physician’s inability with,  174–​75 PiPS model for,  176 resources used for,  177, 178t proper death,  298 proscribed treatment,  79 The Protection of Conscience Project Model Protection of Conscience Act,  92 PSDA. See Patient Self-​Determination Act psychiatric patients,  347, 360–​61 psychological issues, as futility dispute cause, 69 psychological suffering, as continuous sedation indication,  120–​21 PTSD. See posttraumatic stress disorder pulmonary insufficiency, infant survival and, 44 PVS. See permanent vegetative state Pyszczynski, Tom,  231qualitative futility criticism on, 74 disproportionate treatment and,  73–​74 inappropriate treatment and,  74

Index   467 Quality of Death Index,  417–​18 quality of life considerations Baby Doe case influencing,  45–​46 financial costs as,  244 in ICU,  48 for infants,  46, 48 in NICU,  48–​49 prognosis for,  173–​74 quantitative futility,  73 Quill, Timothy,  373–​74 Quinlan, Joseph,  29, 30 Quinlan, Julia,  29, 30 Quinlan, Karen Ann,  8, 25 brain injury of,  155–​56 court case of,  9–​11, 19, 155–​56 feeding tube for,  13–​14 physician’s ethical objections to respirator withdrawal from,  28–​29 respirator support withdrawn from,  28 race communication influenced by,  310–​11 health care access influenced by,  51 health care quality disparities and,  309–​10 infant mortality influenced by,  51 pain management influenced by,  311 reasonable accommodation,  1 to conscience-​based objection,  94–​99 excessive threshold for,  96 guidelines for,  94 objection to,  101–​2 residency choice based on,  101–​2 refractory suffering, last resort and,  116 refusal right, for treatment,  20–​21, 269–​70 Bouvia and,  269–​70 Bowers and,  273 McAfee and,  272–​73 Regulatory Mode theory,  185 religion. See also Catholic-​Christian perspective; Christians; Orthodox Jews; Roman Catholic Church Catholic Health Care Services directives,  140–​41 death and,  299 end-​of-​life care and,  330–​39 as futility dispute cause,  69–​70 hospice concerns of,  412–​13

physician’s objection to abortion and,  87 surrogates’ beliefs,  69–​70 Renjilian, C. B.,  62 reproductive health, conscientious objection and, 88 Research Committee Medical Practice and Euthanasia, 350 respirator withdrawal, controversy of,  28 Rights Come to Mind: Brain Injury, Ethics and the Struggle for Consciousness (Fins),  154 right to die Bouvia and,  269–​70 Bowers and,  273 brain injury and,  155–​56 court cases on,  376–​77 deliberate, 274 generalization of,  156 McAfee and,  272–​73 PASE and,  373 support for,  274–​75 waiting period for,  274–​75 Robert Wood Johnson Foundation,  425, 441 The Rockefeller University,  155 Roe v. Wade (1973),  8, 87, 376 “The Role of the Physician in the Voluntary Termination of Life,”  349 Roman Catholic Church, nutrition and hydration beliefs of,  140–​41 Rousseau, P.,  112 Royal Dutch Medical Association,  111 Royal Society for the Advancement of Medicine, 344 Royal Victoria Hospital,  414 Rushton, Cynda,  232 Saloner, B.,  251 Saunders, Cicely,  49, 412, 440 Savulescu, Julian,  92, 101 SCCM. See Society of Critical Care Medicine SCEN. See checks Scheff, Thomas,  233 Scheper-​Hughes, Nancy,  292 Schiavo, Terri,  14, 282 Schiavo lawsuit,  14, 156, 166 Schiff, N. D.,  158 Schnakers, C.,  160 Schneiderman, Lawrence J.,  74

468   Index Schoonheim case on euthanasia,  346, 349 Schostak, Z.,  142 second medical opinion,  80 SED. See stopping eating and drinking sedation.  See continuous sedation; palliative sedation; terminal sedation “Sedation to Unconsciousness” (AMA),  89 self-​actualization,  301 self-​awareness, during death,  301 self-​care, for physicians,  213–​15 self-​conflict, suffering involving,  222–​23 self-​determined life closure,  390 self-​governance cognitive disabilities and,  261 noncognitive disabilities and,  269 Sen, A.,  251 shame, 238n2 regarding dying,  232–​36 hospice associated with,  234 power of,  233 surrounding dying body,  234 silver lining search, childhood death and,  63 Simon, Scott,  237 skin-​to-​skin,  49 Sneddon, Andrew,  266 SNS. See social network sites social marketing campaign of,  444 principles of,  442–​43 social media advance directives linked to,  280, 286–​87, 288n19 advertising on,  284 communication in,  282–​83 concerns on,  284–​87 date stamp of,  283 emotional tone lacking in,  282 on end-​of-​life decision making,  279–​87 information evaluation on,  283–​84 legal sufficiency of,  285 online identity and,  284, 289n32 participant’s age and,  284–​85 preponderance of evidence on,  285, 289n38 surrogate’s knowledge of,  286 thoughtfulness on,  283 social network sites (SNS),  279, 287n7, 287n9

communications within,  280 types of,  280 Society of Critical Care Medicine (SCCM) (1997), 77 Society of Critical Care Medicine (SCCM) (2015), 79 soft paternalism,  250–​54 Solomon, Sheldon,  231 Sophocles, 66 Sourkes, Barbara,  59 specialty palliative care (SPC),  425 co-​management by,  428 core roles of,  429 critique on,  434–​35 dying experience failure,  427–​29 funding for,  432 GPC compared to,  426–​27, 430–​31 growth of,  426, 427–​29, 428, 433 hospital presence required for,  427–​28 lower hospital costs and,  427 need for,  428 passing patients off to,  431–​32 patient care improvement and,  433–​34 referrals for,  431 standards promotion for,  433 training for,  435 Standards for Culturally and Linguistically Appropriate Services,  322 starving to death, feeding tube withdrawal seen as, 30 State Committee on Euthanasia,  350 state interest analysis, limitation of,  12 St. Augustine,  333 St. Christopher’s Hospice,  440 Steindal, S. H.,  110 stereotyping by clinician,  309 health care professional avoidance approach to,  319–​20 Stewart, Potter (Justice),  74 stopping eating and drinking (SED),  356–​5 7 Strauss, Anselm,  294, 297, 300–​2, 305 on good death,  295 on proper death,  298 Street, Annette,  234 A Streetcar Named Desire, 56

Index   469 Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, 12, 30 “The Study to Understand the Prognoses and Preferences for Outcomes and Risks of Treatment”(SUPPORT) on end-​of-​life care,  425 negative results of,  435, 441 substituted judgment,  34, 275n4, 281 suffering,  2–​3. See also refractory suffering actively hastening death and,  21–​22, 378–​79 alleviation of,  378 avoidance of, 70 of children,  61 compassionate care and,  335, 338–​39 constant features of,  217 continuous sedation for,  110 distinction of,  216–​17 euthanasia relieving,  346–​48 fighting-​back,  223 individual and uniqueness of,  221–​22 of infants,  45, 47 loneliness from,  219, 223 loss from,  223 as meaningful,  109, 223 neuro-​palliative care and,  164 objectivity of,  218 observation for,  218–​19 PASE alleviation of,  371, 377–​79 persistent, 378 person concept and,  217–​18 physician’s impact on,  225 Postma euthanasia case defining,  345 psychological,  120–​21 relief of, 22 scenarios on,  220–​21, 224–​25 self-​conflict and,  222–​23 stimulus for,  219, 221–​22 subjectivity of,  217, 218 treating and controlling,  109, 223–​24, 378 unbearable, 346 suicide. See also physician-​assisted suicide Aquinas’s argument on,  387–​88, 388n2 prevention of, 12 terminology of,  377 “suicide machine,”  22 Summa Theologiae (Aquinas),  387–​88

Sunstein, C. R.,  253 Superintendent of Belchertown State School v. Saikewicz (1977),  10 SUPPORT.  See “The Study to Understand the Prognoses and Preferences for Outcomes and Risks of Treatment” supported decision-​making disability and,  267 implementation of,  268 trial court on,  267–​68 supportive care,  430 Supreme Court, on feeding tubes,  14 surfactant therapy,  45 surgical gastrostomy tube feeding,  128–​29 surrogate decision maker role burden of,  32–​34 implications for,  36–​37 informed consent and,  37 language usage for,  37 liberty interests and,  36 moral and political basis of,  36 patient rights and,  37 patient wish prediction and,  37 reconceiving of,  34–​37 standard view pitfalls and,  34–​35 struggle with,  35 substituted judgment and,  34 understanding of,  38–​39 surrogate decision making “Beloved Brother” case regarding,  32–​33 ethics consultation and problems with,  27 family members forced into role of,  33–​34 nonrecommended treatment request and,  68–​69 patient right’s impacted by,  32 physician’s recommendation refusal and,  35 problems arising from,  28 PTSD risk from,  33–​34 Quinlan, Joseph and Julia’s, fight for,  29 reliance and need for,  27 standards for,  11–​12, 281 surrogates, 288n21 clinician’s resist for requests by,  70–​7 1 distrust by, 69 dubious motives of,  69 extramural appeal opportunity for,  81 process notice given to,  80

470   Index surrogates (Cont.) psychological and emotional issues of,  69 religious and miracle belief by,  69–​70 selection of, 76 social media information knowledge and,  286 transfer offered for,  81 survivor guilt,  231 symptom management in home,  246 in neuro-​palliative care,  165

suffering and,  109, 223–​24 withholding or withdrawing of,  13 trisomy 21 (Down’s syndrome) Baby Doe case and,  45–​46 cognitive disabilities and,  267–​68 cognitive impairment from,  46 Truth and Lies at the Sickbed (Buskes),  343 Tulsky, James,  432 Twitter, 279, 282 Twycross, Robert,  377

“Taking Control of Your Death by Stopping Eating and Drinking” (Chabot),  357 Tännsjö, T.,  112 TBI. See traumatic brain injury Teno, J. M.,  229 tension between despair and hope,  56–​57 terminal delirium, nutrition and hydration during,  138–​39 terminal illness, final stages of,  89 terminal sedation,  111, 354–​55 Battin on,  117 terror management theory,  231–​32 Texas Advance Directives Act (1999),  78 Thaler, R. H.,  253 therapeutic obstinacy,  78 therapeutic privilege,  97 “Thinking About the Good: Reconfiguring Liberal Metaphysics (Or Not) for People with Cognitive Disabilities,”  266 third party protection,  12 threshold percentage, for quantitative futility, 73 total parenteral nutrition (TPN),  129–​30 traumatic brain injury (TBI),  159 treatment. See also aggressive treatment for depression,  400 discretionary,  79–​80 disproportionate,  73–​74 limitations of,  303–​4 moral obligation to provide,  105 nonrecommended request,  68–​69 physician power recommendation,  35 proscribed, 79 refusal to provide,  88–​89 right to refusal for,  20–​21, 269–​73 side effect trade off for,  247

Uit Vrije Wil group,  359–​60 unbearable suffering,  346 uncertainty advance directives,  16 clinical amplifying moral,  59 living-​dying interval and,  59 parental grief and loss and,  59–​60 of prognosis,  177 prognosis communication disclosure for, 182, 183t United Nations,  266 unresponsive wakefulness syndrome (UWS), 356 Vacco v. Quill,  115, 373–​74, 376 valuing critical interests and,  264–​65 dementia patients and,  263–​64 van Delden, J. J.,  112 Veatch, Robert,  29 Veterans Administration,  427 Wald, Florence,  440 Wallis, Terry,  159 Wanglie case, futility of,  18 Washington v. Glucksberg, 110, 376 Weisman, Avery,  195–​96 Weldon Amendment,  102 WHO. See World Health Organization Winslow, Jean-​Jacques,  291–​92 Wired Magazine, 284 withholding versus withdrawing end-​of-​life decisions, 13 nutrition and hydration as,  141–​43 as passively hastening death,  21 World Health Organization (WHO),  415