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

Table of contents :
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Page 11
Page 12
Page 13
Page 14
Page 15
Page 16
Page 17
Page 18
Page 19
Page 20
Page 21
Page 22
Page 23
Page 24
Page 25
Page 26
Page 27
Page 28
Page 29
Page 30
Page 31
Page 32
Page 33
Page 34
Page 35
Page 36
Page 37
Page 38
Page 39
Page 40
Page 41
Page 42
Page 43
Page 44
Page 45
Page 46
Page 47
Page 48
Page 49
Page 50
Page 51
Page 52
Page 53
Page 54
Page 55
Page 56
Page 57
Page 58
Page 59
Page 60
Page 61
Page 62
Page 63
Page 64
Page 65
Page 66
Page 67
Page 68
Page 69
Page 70
Page 71
Page 72
Page 73
Page 74
Page 75
Page 76
Page 77
Page 78
Page 79
Page 80
Page 81
Page 82
Page 83
Page 84
Page 85
Page 86
Page 87
Page 88
Page 89
Page 90
Page 91
Page 92
Page 93
Page 94
Page 95
Page 96
Page 97
Page 98
Page 99
Page 100
Page 101
Page 102
Page 103
Page 104
Page 105
Page 106
Page 107
Page 108
Page 109
Page 110
Page 111
Page 112
Page 113
Page 114
Page 115
Page 116
Page 117
Page 118
Page 119
Page 120
Page 121
Page 122
Page 123
Page 124
Page 125
Page 126
Page 127
Page 128
Page 129
Page 130
Page 131
Page 132
Page 133
Page 134
Page 135
Page 136
Page 137
Page 138
Page 139
Page 140
Page 141
Page 142
Page 143
Page 144
Page 145
Page 146
Page 147
Page 148
Page 149
Page 150
Page 151
Page 152
Page 153
Page 154
Page 155
Page 156
Page 157
Page 158
Page 159
Page 160
Page 161
Page 162
Page 163
Page 164
Page 165
Page 166
Page 167
Page 168
Page 169
Page 170
Page 171
Page 172
Page 173
Page 174
Page 175
Page 176
Page 177
Page 178
Page 179
Page 180
Page 181
Page 182
Page 183
Page 184
Page 185
Page 186
Page 187
Page 188
Page 189
Page 190
Page 191
Page 192
Page 193
Page 194
Page 195
Page 196
Page 197
Page 198
Page 199
Page 200
Page 201
Page 202
Page 203
Page 204
Page 205
Page 206
Page 207
Page 208
Page 209
Page 210
Page 211
Page 212
Page 213
Page 214
Page 215
Page 216
Page 217
Page 218
Page 219
Page 220
Page 221
Page 222
Page 223
Page 224
Page 225
Page 226
Page 227
Page 228
Page 229
Page 230
Page 231
Page 232
Page 233
Page 234
Page 235
Page 236
Page 237
Page 238
Page 239
Page 240
Page 241
Page 242
Page 243
Page 244
Page 245
Page 246
Page 247
Page 248
Page 249
Page 250
Page 251
Page 252
Page 253
Page 254
Page 255
Page 256
Page 257
Page 258
Page 259
Page 260
Page 261
Page 262
Page 263
Page 264
Page 265
Page 266
Page 267
Page 268
Page 269
Page 270
Page 271
Page 272
Page 273
Page 274
Page 275
Page 276
Page 277
Page 278
Page 279
Page 280
Page 281
Page 282
Page 283
Page 284
Page 285
Page 286
Page 287
Page 288
Page 289
Page 290
Page 291
Page 292
Page 293
Page 294
Page 295
Page 296
Page 297
Page 298
Page 299
Page 300
Page 301
Page 302
Page 303
Page 304
Page 305
Page 306
Page 307
Page 308
Page 309
Page 310
Page 311
Page 312
Page 313
Page 314
Page 315
Page 316
Page 317
Page 318
Page 319
Page 320
Page 321
Page 322
Page 323
Page 324
Page 325
Page 326
Page 327
Page 328
Page 329
Page 330
Page 331
Page 332
Page 333
Page 334
Page 335
Page 336
Page 337
Page 338
Page 339
Page 340
Page 341
Page 342
Page 343
Page 344
Page 345
Page 346
Page 347
Page 348
Page 349
Page 350
Page 351
Page 352
Page 353
Page 354
Page 355
Page 356
Page 357
Page 358
Page 359
Page 360
Page 361
Page 362
Page 363
Page 364
Page 365
Page 366
Page 367
Page 368
Page 369
Page 370
Page 371
Page 372
Page 373
Page 374
Page 375
Page 376
Page 377
Page 378
Page 379
Page 380
Page 381
Page 382
Page 383
Page 384
Page 385
Page 386
Page 387
Page 388
Page 389
Page 390
Page 391
Page 392
Page 393
Page 394
Page 395
Page 396
Page 397
Page 398
Page 399
Page 400
Page 401
Page 402
Page 403
Page 404
Page 405
Page 406
Page 407
Page 408
Page 409
Page 410
Page 411
Page 412
Page 413
Page 414
Page 415
Page 416
Page 417
Page 418
Page 419
Page 420
Page 421
Page 422
Page 423
Page 424
Page 425
Page 426
Page 427
Page 428
Page 429
Page 430
Page 431
Page 432
Page 433
Page 434
Page 435
Page 436
Page 437
Page 438
Page 439
Page 440
Page 441
Page 442
Page 443
Page 444
Page 445
Page 446
Page 447
Page 448
Page 449
Page 450
Page 451
Page 452
Page 453
Page 454
Page 455
Page 456
Page 457
Page 458
Page 459
Page 460
Page 461
Page 462
Page 463
Page 464
Page 465
Page 466
Page 467
Page 468
Page 469
Page 470
Page 471
Page 472
Page 473
Page 474
Page 475
Page 476
Page 477
Page 478
Page 479
Page 480
Page 481
Page 482
Page 483
Page 484
Page 485
Page 486
Page 487
Page 488
Page 489
Page 490
Page 491
Page 492
Page 493
Page 494
Page 495
Page 496
Page 497
Page 498
Page 499
Page 500
Page 501
Page 502
Page 503
Page 504
Page 505
Page 506
Page 507
Page 508
Page 509
Page 510
Page 511
Page 512
Page 513
Page 514
Page 515
Page 516
Page 517
Page 518
Page 519
Page 520
Page 521
Page 522
Page 523
Page 524
Page 525
Page 526
Page 527
Page 528
Page 529
Page 530
Page 531
Page 532
Page 533
Page 534
Page 535
Page 536
Page 537
Page 538
Page 539
Page 540
Page 541
Page 542
Page 543
Page 544
Page 545
Page 546
Page 547
Page 548
Page 549
Page 550
Page 551
Page 552
Page 553
Page 554
Page 555
Page 556
Page 557
Page 558
Page 559
Page 560
Page 561
Page 562
Page 563
Page 564
Page 565
Page 566
Page 567
Page 568
Page 569
Page 570
Page 571
Page 572
Page 573

Citation preview

OXFORD

THIRD EDITION

BIOMEDICA EllICS A CANADIAN FOCUS

Roma

111

EDITED BY

FISHER RUSSELL BROWNE r BURKHOLDER

1.I

CCM ©40

THIRD EDITION

5\A I L_

A CANADIAN FOCUS

EDITED BY

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

OXFORD UNIVERSITY PRESS

11 III I

I I ill

W

OXFORD UNIVERSITY PRESS

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries. Published in Canada by Oxford University Press 8 Sampson Mews, Suite 204, Don Mills, Ontario M3C OHS Canada wwwoupcanada.com Copyright 0 Oxford University Press Canada 2018 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First Edition published in 2009 Second Edition published in 2013 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Permissions Department at the address above or through the following url: www.oupcanada.com/permission/permission_request.php Every effort has been made to determine and contact copyright holders. In the case of any omissions, the publisher will be pleased to make suitable acknowledgement in future editions.

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

174.20971 C2017-906107-0

C2017-906106-2

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

1111111

Contents Preface xiii

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

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

A Relational Approach to Autonomy in Health Care 35 Susan Sherwin

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

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

Transparency: Informed Consent in Primary Care 51 Howard Brody

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

■o i

®

LI

I I JII

III Wil

Contents

Informed Consent and Public Health 60 Onora O'Neill

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

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

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

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

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

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

Ethical Relativism in a Multicultural Society 83 Ruth Macklin

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

3

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

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

Telling Patients the Truth 110 Robert Scott Stewart

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

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

Contents

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

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

11

II

I

ill

111111

Contents

5

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

Preconception Gender Selection 226 John A. Robertson

Preconception Arrangements 233 Royal Commission on New Reproductive Technologies

5.3 Stopping Reproduction 238 Why Abortion Is Immoral 238 Don Marquis

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

A Defence of Abortion 253 Judith Ja►vis Thomson

A Third Way 262 L.W. Sumner

Abortion through a Feminist Ethics Lens 268 Susan Sherwin

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

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

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

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

Voluntary Active Euthanasia 297 Dan W. Brock

Contents ® Active and Passive Euthanasia 309 James Rachels

When Self-Determination Runs Amok 312 Daniel Callahan

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

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

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

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

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

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

7

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

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

Justice and the High Cost of Health 372 Ronald Dworkin

Why UndervaluingnStatisticar People Costs Lives 377 Tony Hope

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

O

Contents

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

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

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

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

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

Free Riding and Organ Donation 431 Walter Glannon

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

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

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

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

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

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

Contents

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

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

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

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

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

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

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

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

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

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

Glossary 553 Index 557

G

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

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

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

-1 :

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

,=-

un

II

I

:,111

111111

1 Morality and Moral Decision-Making

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

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

1.2 The Case of Hassan Rasouli 0

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

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

cultural relativism what is

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

11 11 I

O

W

II

all

111141

1 Morality and Moral Decision-Making

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

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

13 Taking Morality Seriously 0

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

I I In

I0

111111

1 Morality and Moral Decision-Making

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

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

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

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

13 Taking Morality Seriously 10

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

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

metaethics the inquiry into the nature and status of morality.

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

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

consequentialism rightness is a function of promoting good consequences.

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

Utilitarianism

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

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

In I

I I UM

1111.0

® 1 Morality and Moral Decision-Making

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

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

beneficence prevent harm, reduce harm, and do good.

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

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

1.4 Normative Ethical Theories ®

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

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

paternalism interfering with

someone for their own good.

uii I

10

LI

II

al

uu

1 Morality and Moral Decision-Making

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

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

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

1.4 Normative Ethical Theories

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

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

®

111

8

H i1

11111

1 Morality and Moral Decision-Making

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

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

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

1.4 Normative Ethical Theories

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

8

on

8

1

I I :111

111111

1 Morality and Moral Decision-Making

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

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

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

1.4 Normative Ethical Theories 8

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

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

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

I I III

G

111111

1 Morality and Moral Decision-Making

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

1.4 Normative Ethical Theories e

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

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

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

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

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

YY I

18

YI

I I .111

our

1 Morality and Moral Decision-Making

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

Ranking Principles

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

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

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

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

1.5 Moral Reasoning in Biomedical Ethics

19

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

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

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

I I HI

I

MIA

1 Morality and Moral Decision-Making

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

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

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

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

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

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

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

11 111 i

e

m

II A

111111

1 Morality and Moral Decision-Making

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

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

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

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

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

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

nu I

II

111

111111

® 2 Medical Decision-Making Patient

N Decision-making capacity

Lacks the characteristics necessary to make decisions for him/herself

Voluntary decision

Full information

Patient decides

Family + Physicians decide

N Substituted judgment

Best-interest judgment

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

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

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

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

Decision-Making by Patient S

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

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

voluntary decision

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

II All

111111

2 Medical Decision-Making

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

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

Decision-Making by Patient ®

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

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

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

II

WlI

1114W

2 Medical Decision-Making

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

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

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

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

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

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

Decision-Making by Substitute Decision-Makers

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

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

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

YY 1

dl

II hill

1111.1.1

2 Medical Decision-Making

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

Decision-Making by Substitute Decision-Makers 31

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

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

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

ii

di]

wu

C

I 2 Medical Decision-Making

2.2 Determining Decision-Making Capacity

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

I. Different Standards of Competence

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

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

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

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

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

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

Buchanan/Brock: Standards of Competence

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

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

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

33

SW 1

s

II di

91

1111:u

2 Medical Decision-Making

Table 1 Decision-Making Competence and Patient Well-Being

The patient's treatment choice

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

Patient consents to lumbar puncture for presumed meningitis

Level of decision-making competence required

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

Net balance substantially better than for possible alternatives

Low/minimal

Principally the benefit/risk assessment made by others

Patient chooses lumpectomy for breast cancer

Net balance roughly comparable to that of other alternatives

Moderate/median

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

Patient refuses surgery for simple appendectomy

Net balance substantially worse than for another alternative or alternatives

High/maximal

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

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

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

Buchanan/Brock: Standards of Competence C

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

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

Notes 1.

2. 3.

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

A Relational Approach to Autonomy in Health Care Susan Sherwin

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

4.

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

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

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

II SI

ww

2 Medical Decision-Making

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

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

Sherwin: A Relational Approach to Autonomy in Health Care

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

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

37

II w

I11111

2 Medical Decision-Making

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

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

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

Sherwin: A Relational Approach to Autonomy in Health Care

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

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

39

II aIII

40

111111

2 Medical Decision-Making

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

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

Sherwin: A Relational Approach to Autonomy in Health Care 41

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

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

II 111

e

11111I

2 Medical Decision-Making

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

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

Sherwin: A Relational Approach to Autonomy in Health Care

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

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

II III

O

uuJJ

2 Medical Decision-Making

Notes 1.

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

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

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

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

Sherwin: A Relational Approach to Autonomy in Health Care 0

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

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

2.3 Informed Consent

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

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

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

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

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

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

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

®

2 Medical Decision-Making

ESTABLISH

DETERMINE

Voluntary atmosphere

Assess mental

Establish relationship by getting to know patient

competency to determine if patient can give consent

Legal capacity

DISCLOSE

DEFINE

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

With the patient

-70

permissible scope of the procedure to be performed

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

VALID CONSENT

Figure 1 The Consent Process

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

Candid Responses to Patient Inquiries

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

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

Rozovsky: The Nuts and Bolts of Obtaining Consent to Treatment

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

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

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

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

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

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

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

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

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

47

II

rill

111111

2 Medical Decision-Making

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

Faden/Beauchamp: The Concept of Informed Consent

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

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

49

■ II

I

I I Ell

1.11

111111

2 Medical Decision-Making

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

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

Notes 1.

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

2.

3.

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

Brody: Transparency: Informed Consent in Primary Care

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

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

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

8

II

41I1

1111;k1

C

I 2 Medical Decision-Making

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

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

Brody: Transparency: Informed Consent in Primary Care e

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

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

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

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

I I Li

54

.111111

2 Medical Decision-Making

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

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

Brody: Transparency: Informed Consent in Primary Care S

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

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

Notes 1. 2.

3.

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

4.

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

II III

G

I

IIan

411

1111.1.1

2 Medical Decision-Making 5.

6.

7.

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

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

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

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

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

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

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

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

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

(Cree): I'm not sure.

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

(Cree): Sometimes dark.

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

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

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

(Nods, but makes no verbal response.)

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

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

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

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

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

INTERPRETER

x-rayed the bowel.

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

Yes.

DOCTOR: And

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

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

II II I

II

111

111111

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

(Nods but makes no verbal response.)

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

INTERPRETER (English): Will

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

(Cree): I don't know.

INTERPRETER: You

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

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

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

DOCTOR:

there be complications?

DOCTOR: There

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

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

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

don't know.

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

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

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

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

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

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

References

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

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

I I UN

60

1111111

2 Medical Decision-Making

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

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

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

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

O'Neill: Informed Consent and Public Health 61

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

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

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

I I al

II1La

2 Medical Decision-Making

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

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

O'Neill: Informed Consent and Public Health

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

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

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

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

2.4 Substituted Judgments

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

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

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

II

i311

111111

2 Medical Decision-Making

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

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

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

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

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

• NI II

II

illl

111Li

2 Medical Decision-Making

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

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

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

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

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

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

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

II

.11

0

Jill

III Vil

2 Medical Decision-Making

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

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

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

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

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

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

II 1111

111111

2 Medical Decision-Making

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

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

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

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

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

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

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

71

II

8

IN

111111

2 Medical Decision-Making

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

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

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

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

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

8

uu

8

I

Al

2 Medical Decision-Making

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

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

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

62. 63. 64.

65. 66. 67.

68.

69. 70.

71.

72.

73. 74.

75. 76.

77.

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

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

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

89.

90. 91.

92. 93.

94.

95.

96.

97. 98.

99.

100.

101. 102.

103.

104. 105.

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

II

1111

1111ild

2 Medical Decision-Making

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

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

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

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

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

2.5 Best-Interest Judgments

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

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

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

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

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

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

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

77

II 111

■ ui I

111111

2 Medical Decision-Making

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

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

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

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

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

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

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

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

Adolescents

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

•

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

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

•

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

The Case

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

Notes 1.

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

2. 3.

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

I I ;ill

111111

C

I 2 Medical Decision-Making

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

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

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

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

Christine Harrison; Canadian Paediatric Society, Bioethics Committee

[...] Withholding or Withdrawing Treatment

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

• • •

•

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

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

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

Harrison/Canadian Paediatric Society: Position Statement

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

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

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

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

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

Withholding or Withdrawing LifeSustaining Treatment in Children*

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

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

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

•

II

LI

111111

2 Medical Decision-Making

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

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

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

2.

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

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

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

3.

4.

5.

6.

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

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

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

Macklin: Ethical Relativism in a Multicultural Society

Ethical Relativism in a Multicultural Society Ruth Macklin

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

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

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

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

83

II

1111

1111;111

2 Medical Decision-Making

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

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

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

Macklin: Ethical Relativism in a Multicultural Society

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

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

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

85

H kill

86

111111

2 Medical Decision-Making

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

Macklin: Ethical Relativism in a Multicultural Society

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

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

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

II

;III

IIIIi

2 Medical Decision-Making

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

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

Macklin: Ethical Relativism in a Multicultural Society

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

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

II

311

II 1111

2 Medical Decision-Making

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

Macklin: Ethical Relativism in a Multicultural Society

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

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

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

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

2.6 Cases

Case 1

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

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

91

II

0

'ail

111111

2 Medical Decision-Making

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

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

Notes 1.

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

Case 2

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

2.

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

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

Cases

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

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

Notes 1.

CBC News, "Quebec Parents Refuse Cancer Treatment

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

Case 3

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

2.

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

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

93

II III

94

II

II

ei

1111

.11111J

2 Medical Decision-Making

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

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

Notes 1.

2.

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

Case 4

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

3. 4.

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

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

Cases

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

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

Notes 1. 2.

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

Case 5

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

3. 4.

Ibid. CBC News, 26

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

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

Note 1.

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

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

95

II

UM

111111

2 Medical Decision-Making

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

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

Suggested Further Reading

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

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

111

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

Access to Information Therapeutic Privilege

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

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

Access to Information

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

•

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

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

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

LA.

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

C /

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

YU I

ai

II 'al

iuw

3 Management of Medical Information

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

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

Confidentiality

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

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

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

II

ill

11111,1

3 Management of Medical Information

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

Confidentiality

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

Table 3.1 Data with Identifying Details Name

Birth Date

Sex

Visit Date

Postal Code

Medical Issue

Sean Phillips

21 Mar 1995

Male

14 Apr 2016

V6T1Z1

Fever

May Ong

10 Nov1994

Female

15 Apr 2016

V6Y2Z6

Chest pain

John Scatopolus

17 Apr 1994

Male

15 Apr 2016

V6T1Z1

Wheezing

Lisa Friedrich

01 Dec 1993

Female

16 Apr 2016

V6Y2Z6

Aching joints

Table 3.2 Data with Identifying Details Removed Name

Birth Date

Sex

Postal Code FSA

N1

21 Mar 1995

Male

V6T

Fever

N2

10 Nov 1994

Female

V6Y

Chest pain

N3

17 Apr 1994

Male

V6T

Wheezing

N4

01 Dec 1993

Female

V6Y

Aching joints

Medical Issue

II all

I

1111.1J

3 Management of Medical Information

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

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

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

Genetic Information

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

Notes 1.

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

8.

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

I I IA

106

111111

3 Management of Medical Information

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

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

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

Kant: On the Supposed Right to Lie from Benevolent Motives

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

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

Reasons for Telling the Truth

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

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

In General Relationships

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

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

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

II

111413

3 Management of Medical Information

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

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

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

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

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

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

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

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

II

1111i1

3 Management of Medical Information

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

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

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

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

4.

5. 6.

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

p; Telling Patients the Truth Robert Scott Stewart Introduction

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

Respect for Persons

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

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

Stewart: Telling Patients the Truth

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

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

I I !III

llllll

3 Management of Medical Information

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

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

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

Stewart: Telling Patients the Truth

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

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

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

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

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

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

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

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

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

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

LE

114

I I ILII

111111

3 Management of Medical Information

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

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

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

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

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

I I al

111111

3 Management of Medical Information

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

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

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

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

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

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

Conclusion

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

References 1. 2. 3.

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

4. 5.

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

3.3 Confidentiality of Information

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

The following fictionalized case is based on an actual incident.

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

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

II

111111

3 Management of Medical Information

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

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

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

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

Kipnis: A Defense of Unqualified Medical Confidentiality

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

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

II

al

III Lid

3 Management of Medical Information

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

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

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

Kipnis: A Defense of Unqualified Medical Confidentiality

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

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

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

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

11 I

II

3

111111

3 Management of Medical Information

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

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

Kipnis: A Defense of Unqualified Medical Confidentiality 123

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

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

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

II III I

ai

II

IA

111111

3 Management of Medical Information

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

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

Kipnis: A Defense of Unqualified Medical Confidentiality

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

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

II

al

liIIlI

3 Management of Medical Information

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

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

Kipnis: A Defense of Unqualified Medical Confidentiality

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

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

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

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

II

,111

1111.111

3 Management of Medical Information

Breaching Confidentiality Dave Unger

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

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

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

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

Unger: Breaching Confidentiality

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

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

uW I

II

11

111111

3 Management of Medical Information

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

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

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

Unger: Breaching Confidentiality

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

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

II ;A

1111.11.1

3 Management of Medical Information

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

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

Unger: Breaching Confidentiality

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

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

Note

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

3.

4.

5.

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

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

II

Lill

IIIIJJ

3 Management of Medical Information

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

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

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

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

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

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

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

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

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

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

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

.11111.1

3 Management of Medical Information

The European Union takes a more comprehensive general approach to privacy; Article 8 of the European Convention on Human Rights includes the right to data protection. This right is embodied in the 1995 Directive 95/46/EC on the Protection of Individuals with Regard to the Processing of Personal Data and on the Free Movement of Such Data.12 Member states implement directives differently, but the EU Data Protection Regulation establishes a single set of rules for data protection across the EU; the final texts are expected to be adopted by the European Parliament at the beginning 2016 and the new rules to become applicable two years after.13,14 Despite their differences, both the United States and the EU construe privacy as control and protection of data rather than other conceptions of privacy15 Both the United States and the EU also make special note of health information, and both rely on stripping data of content presumed to identify the individual represented by the data. As Paul Ohm points out: "In addition to HIPAA and the EU Data Protection Directive, almost every single privacy statute and regulation ever written in the U.S. and the EU embraces—implicitly or explicitly, pervasively or only incidentally—the assumption that anonymization protects privacy, most often by extending safe harbors from penalty to those who anonymize their data."16 As these safe harbors stipulate, neither HIPAA nor the EU Data Protection Directive applies after data are de-identified. However, relying on de-identification contributes to what is considered an inadequate and problematic legal framework for data protection.17 Addressing concerns over de-identification "would require a significant shift in approach towards data protection across Europe."18 Similar deficiencies plague the United States.'9,2° Privacy protection, then, depends not merely on de-identification but on expectations, transparency, and how data are used. De-identification, or anonymization, presumes that it is possible to identify and enumerate the kinds of data that might contribute to privacy risks and to specify how to prevent harms,21 that such a list is static and sufficient in all contexts,22 and that there are no privacy harms if the individual is not identified, even though individuals may object to uses of their personal data even if they themselves are anonymous.23 Furthermore, HIPAA permits secondary uses of data for research, public health, law enforcement, judicial proceedings, and other "public interest and benefit activities," without individual authorization, thereby assuming that "public interest" is clearly understood.24,25 All are questionable assumptions.

Duty of Confidentiality

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

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

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

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

Patient Benefits and Harms

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

Transparency and Consent

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

3 Management of Medical Information

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

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

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

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

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

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

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

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

■ II

I

Al

II

dill

111141

3 Management of Medical Information

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

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

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

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

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

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

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

uw I

ai

I1..11

IIILIJ

3 Management of Medical Information

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

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

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

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

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

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

12.

13.

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

8.

9.

10.

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

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

14.

15. 16.

17.

18. 19.

20. 21. 22. 23. 24.

25. 26.

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

mu I

H 11

MILLI

3 Management of Medical Information

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

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

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

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

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

56.

57.

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

58. 59.

60.

61. 62.

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

43. 44.

45.

46.

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

63.

64.

65.

66. 67.

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

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

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

Enhancing Privacy, Improving Health through Research.

74.

75.

76.

77.

78.

79.

80.

81. 82. 83. 84.

85.

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

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

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

3 Management of Medical Information

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

Simple Traits, to Complex Traits, to Personalized Medicine.

111.

112. 113.

114.

115. 116. 117.

118.

119.

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

120.

121.

122. 123. 124. 125.

126.

127.

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

135.

136. 137. 138. 139. 140.

141. 142. 143. 144. 145.

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

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

Murray: Genetic Exceptionalism and "Future Diaries" 147

3.4 Genetic Information

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

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

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

lb I

3 Management of Medical Information

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

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

Murray: Genetic Exceptionalism and "Future Diaries"

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

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

II

Al

111111

3 Management of Medical Information

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

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

Murray: Genetic Exceptional ism and "Future Diaries"

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

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

II

1111.1.1

3 Management of Medical Information

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

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

Murray: Genetic Exceptionalism and "Future Diaries"

Notes 1. Task Force Report: Genetic Information and Insurance

2. 3.

4.

5. 6. 7.

8. 9.

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

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

3.5 Cases

Case 1

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

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

■m I

II

II

IIILL1

3 Management of Medical Information

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

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

Notes 1. 2. 3. 4. 5.

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

Case 2

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

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

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

Cases

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

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

Notes 1. 2. 3. 4.

5.

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

Case 3

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

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

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

11 111

1

II 1,11

II

11141

3 Management of Medical Information

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

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

Notes 1. 2.

3.

4. 5.

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

Case 4

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

6.

7. 8.

9.

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

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

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

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

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

Notes 1. 2.

3. 4.

5.

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

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

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

II

al

111111

3 Management of Medical Information

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

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

-

cro.lessona0 Ethilcs

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

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

nni

II

11

Al

11110

4 Professional Ethics

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

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

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

Futility

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

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

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

II 131

162

nurI

4 Professional Ethics

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

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

Gatekeeping 163

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

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

II

164

al

1111111

4 Professional Ethics

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

Conscientious Objection

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

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

Conscientious Objection

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

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

4.2 The Nature and Limits of Professional Autonomy and Professional Responsibility

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

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

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

II LE

166

111111

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

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

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

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

Pellegrino: Patient and Physician Autonomy

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

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

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

II

11111

Mid

4 Professional Ethics

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

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

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

Pellegrino: Patient and Physician Autonomy

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

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

Hal

1111111

4 Professional Ethics

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

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

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

Pellegrino: Patient and Physician Autonomy

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

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

I I AI

MLA

4 Professional Ethics

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

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

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

Pellegrino: Patient and Physician Autonomy

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

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

I

174

II

111111

4 Professional Ethics

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

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

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

Pellegrino: Patient and Physician Autonomy

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

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

uu

I

II Jl

UI

1111113

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

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

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

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

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

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

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

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

Truog/Brett/Frader: The Problem with Futility

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

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

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

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

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

I .11

1111.111

4 Professional Ethics

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

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

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

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

Truog/Brett/Frader: The Problem with Futility

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

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

179

I I 31

111111

4 Professional Ethics

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

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

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

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

Truog/Brett/Frader: The Problem with Futility

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

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

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

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

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

I I DII

111111

182 4 Professional Ethics

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

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

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

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

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

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

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

.1 .11

1

1.11

II ill

muu

4 Professional Ethics

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

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

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

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

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

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

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

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

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

186

4 Professional Ethics

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

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

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

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

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

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

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

II

MILLI

4 Professional Ethics

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

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

References

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

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

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

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

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

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

Canadian Medical Association: Caring in a Crisis

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

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

uW

II

I

;ill

III W

4 Professional Ethics

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

•

•

During a Pandemic

• • •

• • •

•

• Prior to a Pandemic

•

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

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

•

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

Canadian Medical Association: Caring in a Crisis 191

•

•

•

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

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

After a Pandemic

•

•

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

C. How Are Physician Obligations and Reciprocal Obligations Related?

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

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

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

Conclusion

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

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

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

MI

192

I

I I Al

II I La

4 Professional Ethics

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

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

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

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

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

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

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

II I

I I II

111111

4 Professional Ethics

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

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

It could be objected that the argument advanced in this paper is too simplistic as it appears to target only unfettered conscientious accommodation stances, when the focus in today's policy debates is about fettered conscientious objection accommodation. It may be argued that the question really is: where and how could society draw reasonable lines that take into account societal interests as well as those of individual objectors? The preceding section explains to some extent why we have taken this stance. The courts in various jurisdictions have already, rightly so, conceded that it is a moot point trying to establish the truth or even plausibility of the views purportedly held by conscientious objectors. That is not an insignificant point, because this concession opens the door to any number of more or less arbitrary and random conscientious objection claims. For policy-makers aiming to establish a functioning healthcare system with predictable service delivery and

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

guaranteed service levels to the people who finance the system, this constitutes an insurmountable problem. It is nigh impossible to predict which healthcare professional, in which part of the system, will demand accommodation for which kinds of purported or real convictions. It is also evidently impossible to verify whether objecting healthcare professionals even hold the views they profess to hold. Such claims may merely be a convenient way out of the provision of inconvenient healthcare services. In light of this, it seems to us that whatever attempt at a principled fettered accommodation is made, it is by necessity resulting in arbitrary outcomes. Regardless of the accommodations that we make today, these can and likely will be challenged time and again by new generations of conscientious objectors. Today it might be abortion and assisted dying, tomorrow it might be the use of the tools of personalised medicine or something else altogether. Limiting conscientious objection accommodation to defensible claims seems impossible to us, unless we overcome the two problems mentioned: demonstrate the truth of the foundations of the conscientious objection and demonstrate evidence that objectors actually genuinely hold the views they claim to hold. Failing that, as we will show, the inevitably ensuing arbitrary accommodation demands will have harmful real-world consequences as far as healthcare outcomes and patient access to care is concerned. A Few Canadian Examples

Lobbyists for doctors usually defend the view that doctors must never be required to provide services that they object to on conscience grounds. Their focus is on ensuring that conscientious objectors do not have to oblige patients asking for such healthcare services or even have to assist them in finding a healthcare provider who will oblige them. A good example of this is Dr Jeff Blackmer, the Director of the Canadian Medical Association's Ethics Office. Dr Blackmer writes on behalf of the Canadian Medical Association that medical doctors are neither obliged to provide abortion services, nor are they obliged to transfer patients on to doctors they know will provide abortions to women seeking one.28 Women are legally entitled to access abortions in Canada if they so wish, and typically abortions are fully funded by provincial public healthcare systems. Canadian doctors are the only professionals legally entitled to provide such procedures. They are monopoly providers. Dr Blackmer's considered view on this subject matter is that pregnant women who wish to have access to a medical procedure that they are legally entitled to, and that is fully funded by the state, should

have to depend on the goodwill of volunteering doctors who also happen to be the only licensed providers of this procedure in the country. A second example: At the time of writing, Canadian legislators grapple with the question of how to implement a Supreme Court of Canada judgement that effectively decriminalises providing assistance in dying to competent patients suffering from an irreversible chronic illness that renders their lives not worth living in their own considered view.2° Dr Cindy Ford, the current president of the Canadian Medical Association, mirrors Dr Blackmer's stance on the obligations of conscientious objectors. In a statement made to the Canadian parliamentary committee tasked with investigating possible legislative options on assisted dying, Dr Ford reportedly stated that it is unreasonable to require conscientious objectors to even refer such patients to a clinician who they know will be prepared to render assistance in dying.3° Dr Blackmer agrees with her, while at the same time trying to assure the Canadian parliamentary committee members of reliable professional service delivery; as he says, "I can sit here . . . and guarantee that from simply a number perspective, access will not be an issue."31 Dr Blackmer's argues that given that 30% of Canadian doctors are willing to provide assistance in dying, there should not be a problem for patients wanting to access such services. If it was just a numbers game, Dr Blackmer would certainly be correct. Given the relatively small number of eligible patients likely to ask for assistance in dying, there should be a sufficient number of doctors willing to provide assistance to such patients. However, as so often, the devil is in the detail. The problem is— we suspect this is the reason why Dr Blackmer does not actually offer unconditionally guaranteed access on behalf of his association—that this likely will not help seriously sick and vulnerable patients who happen to live in remote areas of Canada where only few doctors happen to practise and where all of them might be opposed to assisted dying on conscience grounds. It is irrelevant then that elsewhere in the country plenty of doctors would be willing to provide assistance in dying on request if a particular patient is unable to travel to them. To these patients, Dr Blackmer's "guarantee" is worth very little. We should note here that the Canadian Medical Associations stance is not unusually radical among its peers. The British Medical Association, too, produced a guideline stipulating that "whenever possible physicians who are conscientiously opposed to forgoing treatment should be permitted to hand over care of the patient to a colleague," but they are not obliged to do so.32 The entitlement mentality that drives these

195

II

196

,d11

1111.1.6/

4 Professional Ethics

associations' stances is unsurprisingly supported by the doctors they represent.33 A further example of the potential impact of conscientious objection on patients' access to care in Canada can be seen in the recent application filed by various religious physicians and associations representing them in the Ontario Superior Court. The application challenges the College of Physicians and Surgeons of Ontario (cPSo) Policy Statement #2-15: Professional Obligations and Human Rights, which among other things requires that physicians "provide a referral to another appropriate health-care provider for the elements of care the physician is unable to manage directly."34 The applicants, including the Christian Medical and Dental Society of Canada and the Canadian Federation of Catholic Physicians' Societies, described the referral provisions of the CPSO policy statement as "unconscionable" and a violation of their freedom of conscience and/or freedom of religion.35 This example is hardly unique.36 These kinds of hardline positions disregard the needs of patients. Statements and actions of this kind strongly suggest that Professor Charo was not attacking strawmen in her article. Respect for Private Conscience Choices Is Already Limited

Regardless of the views espoused by representatives of the Canadian Medical Association and the religious doctors' groups mentioned, there are already limits placed on the conscience choices of Canadians in various settings. The Supreme Court of Canada stated that "freedom of conscience and religion protects the right to entertain beliefs, to declare them openly and to manifest them, while at the same time guaranteeing that no person can be compelled to adhere directly or indirectly to a particular religion or to act in a manner contrary to his or her beliefs."3' In that same case, the Court went on to say that the "state's duty to protect every person's freedom of conscience and religion means that it may not use its powers in such a way as to promote the participation of certain believers or nonbelievers in public life to the detriment of others."38 These principles are seen as cornerstones of maintaining a free and democratic society.39 This is not to say that physicians and other medical professionals should be denied the right to hold religious or conscience beliefs. In liberal democracies, a diversity of beliefs and opinions is welcomed, and individuals are free to advocate for societal acceptance of their particular world view. It is also trite to say that the law in liberal democratic societies may protect a particular social or moral position espoused by a particular

religion; however, it should not do so on the basis that it is a religious position but on the basis that "in reason its merits commend themselves."4° Therefore, in order to maintain a defensible balance between competing beliefs in a liberal and democratic society, such societies have generally considered "the practice of religion and the choices it implies to relate more to individuals' private lives or to voluntary associations."' To this end, it has been recognised that within the public realm an individual's freedom of religion and conscience may be legitimately burdened. In another case, the Supreme Court of Canada explicitly acknowledged that legislative or administrative actions may increase the cost of practising or otherwise manifesting one's religious beliefs.42 Further, such legislative or administrative burdens may be justified where they prevent conduct that would potentially cause harm to or cause interference with the rights of others.43 The reasoning of the Court supports the view that the conscientious belief of a physician may be legitimately halted at the point that it interferes with the rights of patients to access services they are entitled to receive. That interference, however, is precisely what happens each time a conscientious objector is accommodated and the patients are unable to access the required service elsewhere. Respect for Private Conscience Choices Will Result in Avoidable Suboptimal Access to Healthcare

Any society that grants medical professionals a conscientious objection-based opt-out will have to accept suboptimal health outcomes with regard to procedures that are considered at that point in time part and parcel of good professional healthcare practice in those societies. Why would toleration of conscientious objection lead invariably to suboptimal health outcomes? Take the Canadian assisted dying example. It is apparent that the representatives of Canadian doctors are unconcerned about the distress they would inflict on—often—dying patients in rural areas who cannot access easily a doctor willing to provide assistance in dying. In many cases, doctors' objection could be more than an inconvenience., it could be an insurmountable barrier to access. A case in point is the small Canadian province of Prince Edward Island (also known as PEI). Women in that province, at the time of writing, are required to leave the province if they wish to have an abortion because healthcare providers are not offering this service in the province. Empirical evidence elsewhere suggests strongly that "travel is a barrier to accessing legal abortion."44 The situation in PEI is not unique in the world, far from it. Minerva reported recently that some 70% of Italian gynaecologists conscientiously object to performing abortions,

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

which is arguably one of the reasons for staggeringly high backstreet abortion rates in that country.45 Patient interests come clearly last here. As we have seen, the representatives of the monopoly providers of these healthcare services in Canada are quite content with that outcome, as long as the association's members' consciences—or, more precisely, their untestable claims about the content of their consciences—are not burdened or otherwise inconvenienced. This attitude is quite the opposite of what it means to be a professional, where a promise is made to serve the public good and to serve patient interests first and foremost. Apparently such central features of what it means to be a professional can readily be sacrificed when arbitrary, untestable individual conscience claims are mounted by professionals. Respect for Private Conscience Choices Will Result in Avoidable Inequitable Workloads for Doctors

It seems also uncontroversially true to us that toleration of conscientious objectors will have a detrimental impact on healthcare practice in other respects. Because patients can expect to be shepherded around among different healthcare professionals choosing "conscientiously" to provide different levels of healthcare, there will be an increasing level of upheaval visited upon healthcare systems. Worse, if the Canadian Medical Association had its way, patients would not even be shepherded through the system, it would be up to them to find doctors obliging them for particular ailments that might be of concern to doctors with any number of idiosyncratic private views of the universe and on what may or may not be ethically acceptable. An important consequence of such a situation would be that doctors who are willing to provide services conscientiously refused by some or many of their colleagues would have to carry an inequitable load of such work. We cannot think of a good reason for why they ought to accept such an inequitable burden and, indeed, why a just society should design a health delivery system that burdens such doctors unfairly. Respect for Private Conscience Choices Will Result in Unpredictable and Unfair Service Delivery

Because of the unpredictable nature of what it is that conscientious objectors will object to, patients can never be quite certain about the kinds of services that they will be able to receive from a particular doctor, even their long-time family doctor. That is the opposite of what can reasonably be expected of a profession and professionals. Uniform service standards are arguably one of the hallmarks of what constitutes a

profession. Another aspect of this has to do with the unfairness of this situation vis-à-vis the patient. Patients, as Ruth Purtilo rightly notes, "can expect to be treated fairly. Persons seeking treatment should not be given advantage on the basis of arbitrary favouritism or be left out on the basis of arbitrary dislike."46 Given the intractability of conscience claims, it is not unwarranted to characterise them as essentially arbitrary dislikes. They might not be arbitrary in the eyes of the objector, but we cannot even be certain of that, given our inability to test the objector's conscience claims. They are arbitrary with regard to what kinds of services particular patients can or cannot expect from particular professionals belonging to the same profession in the same jurisdiction. The nature of professions is that they provide privileged services to the public. They are quite unlike political parties, churches, animal welfare, or environmental organisations. While the latter cater to particular sectarian audiences, professionals must cater uniformly to everyone within the scope of professional practice. The services professionals provide to the public must not be subject to their private normative judgements about individual needs of members of the public, as long as what these members of the public require falls within the scope of professional practice. After all, patients do not seek out doctors for their individual convictions but to receive professional services that they are trained to provide. There are other sounds reasons for refusing to tolerate conscientious objectors in medicine. They have to do with the voluntary choices made by those who decided to join the profession. Choosing to Join a Profession Is a Voluntary Activity Undertaken by an Autonomous Adult

It seems to us that the argument from voluntariness is important in this context. Nobody is forced to join a particular profession, medicine included. Pace Christopher Cowley's argument that the practice of medicine for some is a "calling"—one can still choose not to answer the "call."47 The decision to join the medical profession is quite distinct to, say, joining the military as a conscript. Medical schools the world all over reject most applicants, not because the applicants are unsuitable to become doctors but because the number of places they have available is always much smaller than the number of people who apply. Anyone joining a profession knows, or should know, that it is ultimately up to society to determine the scope of professional practice. Society can do this through professional bodies, but it can also choose to do so by other regulatory means. Having a monopoly on the provision

II

.41

111111

4 Professional Ethics

of services within that scope of practice is not something that comes cost-neutral to those who choose voluntarily to join a particular profession. Indeed, given the high-stakes competition to join the medical profession in the first place, it is reasonable to suggest that doctors refusing to provide professional services that are within the scope of practice should be replaced by someone who is willing to undertake the work. If at any given time a doctor is unable to continue practising due to their—ultimately arbitrary—conscience views, nothing would stop them from leaving the profession and taking up a different vocation. This happens across industries and professions very frequently. Professionals can be expected to take responsibility for the voluntary choices they make. Does the Nature of a Particular Procedure Matter?

It has been suggested that not all conscientious objections fall into the same category. The reason why it would not be acceptable for the female Muslim doctor in our example to refuse to see a male patient had to do with that being in violation of the core values in medicine. Assisted dying or abortion, on the other hand, would then be seen if not in violation of traditional medical values but at least being sufficiently controversial to justify the accommodation of conscientious objectors. Wicclair argues for one possible threshold that a conscientious objection must meet in order to be prima facie worthy of accommodation. He proposes that a conscientious objection "has significant moral weight only if the core ethical values on which it is based correspond to one or more core values in medicine."48 Wicclair is, therefore, supportive of conscientious objection to abortion or assisted dying because to his mind neither of these medical procedures corresponds uncontroversially to these core values. We think that he is mistaken about what he considers to be the core values of medicine, which would be applicable in these two instances, but we do not wish to debate the merits of this claim here. Rather, his argument also fails because despite his protestations to the contrary it is essentially an argument from tradition. What he calls core values could be more aptly described as traditional values. However, it is in the nature of such values that they are changing. Medicine is currently arguably undergoing a paradigm shift from maintenance of human life at nearly all cost to a quality-oflife focused approach.49 The traditional core values of medical practice are in flux, and that is one reason for why abortion is available in most liberal Western democracies, and that is the reason why the number of jurisdictions that are decriminalising assisted dying is steadily increasing.

To highlight just one example of this, again from Canada, the past president of the Canadian Medical Association, Dr Chris Simpson, on behalf of the association, referred to assisted dying as a "therapeutic service," a mere 15 years after Wicclair suggested that core professional values would make such a description impossible.8° It is unlikely that the association would refer to assisted dying as a "therapeutic service" if it thought that the provision of this service would violate Wicclair's claimed core values of the profession. If it is a therapeutic service, as we agree it is, Canadian doctors wanting to go the conscientious objector route to avoid providing this service to eligible patients could no longer avail themselves of Wicclair's analysis, because the "moral integrity of the medical profession" would not be served by accommodating individual doctors who refuse to provide therapeutic services. It is important to recognise that medical practice is also a cultural practice that changes over time. No democratic society and no medical profession should leave the scope of the provision of professional medical services to the vagaries of its members' personal ideological convictions. What about Our Societal Ability to Recruit People to Join the Medical Profession?

An argument could be mounted that if we prevented doctors from excusing themselves from providing professional services that they happen to feel strongly about, we might find it difficult to find a sufficient number of people willing to become doctors. This seems far-fetched to us, but it is conceivable that some people might reconsider joining the medical profession if the accommodation of their feelings with regard to particular professional services is not guaranteed. We indicated already that the profession and society would likely be better off if such people chose not to join the profession, seeing that they think that their idiosyncratic views of the universe should be of greater significance than the patients whose interests they promised to further first and foremost as professionals. The concern about recruitment problems does not appear to be based on real-world evidence to begin with. A recent survey reports that countries such as Sweden, Finland, and Iceland, which do not permit their doctors to opt out of the provision of abortion service on grounds of conscience, have not experienced harmful consequences as far as these countries' healthcare systems' ability to train a sufficient number of doctors is concerned." A somewhat related argument suggests that by preventing doctors from refusing service based on religious or conscience grounds, we would run the

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

risk of losing the participation of those kinds of individuals in the profession. However, that argument does not hold. Those who object to particular procedures could choose specialities that would not require that they violate their conscience; for example, they could opt for dermatology instead of gynaecology if they are opposed to abortion. A self-imposed limit on one's choice of speciality is not inherently unfair. Individuals with certain physical limitations can also be excluded from certain specialities in the medical profession. That does not preclude them from joining the medical profession. Does It Matter When Professionals Joined a Profession?

It could be argued that it matters when conscientious objectors joined the profession. The idea here is that they agreed to join a profession that at that time had a particular scope of practice that these professionals agreed to work in. In a way, this scope of practice constitutes a kind of contract between society and its professionals. If a society wishes to change a profession's scope of practice, it should accommodate conscientious objectors among those who joined the profession prior to a given controversial change, for example, by grandmothering them in. Some of the reasons we mentioned above mitigate against granting objectors a blanket right to such an accommodation. There is another reason that suggests that this argument is not particularly strong. It is surprising that doctors seem to think in large numbers that society provided them with a monopoly on the provision of medical services but that society at no time would add or subtract from the range of services it considers scope of practice of that profession. This is puzzling given the nature of medicine as an everevolving scientific and cultural pursuit. Most of us who do not work as medical professionals have encountered employers who wish to change our scope of practice. Many academics the world all over are suffering under their employers' desire and pressure to introduce online learning programmes, despite our objections on pedagogical and other grounds. However, nobody would think that our employers are not well within their rights to ask us to contribute to such programmes. Unlike doctors, we are not even entrusted with typically well-remunerated service delivery monopolies. Having said this, we do think that we should try at least to accommodate conscientious objectors who have practised for a long term under a different kind of contract than doctors who join the profession after particular significant changes to the scope

of practice have been made. However, that is a far cry from suggesting that doctors have a right to such accommodation. Could Not the State Be the Guarantor of Access?

Proponents of a fettered accommodation system might argue that we should try to accommodate conscientiously objecting healthcare professionals regardless of the arguments presented by us so far. They could argue that, for instance, institutions other than the medical profession ought to guarantee reliable access in a healthcare system that is proactively being subverted by accommodation-demanding conscientious objectors. Why should not the health ministries in Canada's provinces organise reliable access systems based on doctors who are not conscientiously objecting? We know from the mentioned experiences of pregnant women seeking an abortion in the province of PEI that this might not be feasible in parts of the country. The odds are fairly high that the more rural an assistance-seeking patient is located, the more difficult it could become to guarantee access. We have yet to see a persuasive reason, given the arguments presented so far, why any healthcare system should burden itself, and ultimately patients, with these sorts of logistical problems when there is an obvious, more efficient alternative: saying no to the conscience-based accommodation requests of healthcare professionals. A profession that is unable to guarantee reliable access to its services for no other reasons than its desire to accommodate the private ideological convictions of some or many of its monopoly provider members is failing in its mission. Conclusion

Medical professionals have no moral claim in liberal democratic societies to the accommodation of their individual conscientious objections. To accommodate such objections would subvert some of the very reasons for why the medical profession was created in the first place. To accommodate them would also permit such medical professionals to abuse the monopoly privileges that society endowed their profession with. Medical professionals practising medicine in the 21st century would be well advised to accept professional service delivery, as defined by the scope of professional practice, as one inevitable corollary of their voluntary decision to join the profession. Forcing patients to live by the conscientious objectors' values constitutes an unacceptable infringement on the rights of patients.

MU I

4 Professional Ethics

References 1. Smith VP. Conscientious objection in medicine: doctors' Amselem: Syndicat Northcrest v. Amselem, [2004] 2 SCR freedom of conscience. BMJ 2006;332:425. 551, 2004 SCC 47. 2. Braun H. Hermann Stohr (1898-1940). In: Hummel 24. Smith: Employment Division, Department of Human K-J, Strohm C, eds. Zeugen einer besseren Welt. ChristResources of Oregon v. Smith, [1990] 494 US 872. liche Martyrer des 20. Jahrhunderts. Evangelische Ver- 25. Charo RA. The celestial fire of conscience-refusing to deliver medical care. N Engl J Med 2005;352:2471-3. lagsanstalt. Leipzig: Butzon & Bercker, 2000:87-105. 3. Michalczykl]. Medicine and ethics in the Third Reich: historical 26. Cantor], Baum K. The limits of conscientious objection: and contemporary issues. NewYork: Sheed and Ward, 1994:8. may pharmacists refuse to fill prescriptions for emer4. Weinstock D. Conscientious refusal and healthcare gency contraception? N Engl J Med 2004;351:2008-12. professionals: does religion make a difference? Bioethics 27. Lynch HE Conflicts of conscience in health care: an institutional compromise. Cambridge MA: MIT Press, 2008. 2014;28:8-15. 5. Leiter B. Why tolerate religion? Princeton: Princeton 28. Blackmer J. Clarification of the CMA's position concernUniversity Press, 2013. ing induced abortion. CMAJ 2007;176:1310. 6. Laycock D. Regulatory exemptions of religious behavior 29. Carter: Carter v. Canada (Attorney General), [2015] 1 SCR 331, 2015 SCC 5. and the original understanding of the establishment clause. Notre Dame Law Review 2006;81:1739-842, at 1839 for 30. Zimonjik P. Doctors with moral objections to assisted dying should be able to opt out: committee hears. CBC the USA. Strasbourg Consortium for European Court of Human Rights cases at http://www.strasbourgconsortium. January 27, 2016. http://www.cbc.ca/news/politics/assisted-dying-committee-evidence-hearings-1.3422352 org/. We owe these references to Leiter B 2013, n4. (accessed 28 Jan 2016). 7. Sulmasy DP. What is conscience, and why is respect for 31. Zimonjik P 2016. it so important? Theo Med Bioeth 2008;29:135-49. 32. As quoted in Wicclair 2000, 218. 8. Sulmasy DP 2008, at 139. 9. West-Oram P, Buyx A. Conscientious objection in 33. Lawrence RE, Curlin FA. Physicians' beliefs about conscience in medicine: a national survey. Acad Med health care provision: a new dimension. Bioethics 2015. 2009;84:1276-82. dot 10.1111/bioe.12236. 10. Wicclair MR. Conscientious objection in medicine. Bio- 34. College of Physicians and Surgeons of Ontario (CPSO). Policy Statement #2-15: Professional Obligations and Human ethics 2000;14:205-27. Rights. March 2015. http://www.cpso.on.ca/CPSO/media/ 11. Pellegrino E. The physician's conscience, conscience documents/Policies/Policy-Items/Human-Rights clauses, and religious belief: a Catholic perspective. Fordham Urban Law J 2002;30:221-44. .pdf?ext=.pdf (accessed 31 Jan 2016). 12. E.g., Pellegrino E 2002. 35. End-of-Life Law and Policy in Canada (EOL). De13. E.g., Weinstock D 2014. velopments to watch-Christian Medical and Dental 14. LaFollette E, LaFollette H. Private conscience, public Society of Canada v. College of Physicians and Surgeons of Ontario. 2015. http://eol.law.dal.ca/wp-content/ acts. J Med Ethics 2007;33:249-54. uploads/2015/06/2015_03_20_Notice_of Application-3 15. Savulescu J. Conscientious objection in medicine. BMJ .pdf (accessed 31 Jan 2016). 2006;332:294-7. 16. Schuklenk U. Conscientious objection in medicine: pri- 36. Laycock D. How the Little Sisters of the Poor case puts religious liberty at risk. Washington Post 2015, March 20. vate ideological convictions must not supersede public https://www.washingtonpost.com/opinions/how-theservic