Beyond Regulations: Ethics in Human Subjects Research (Studies in Social Medicine)

Beyond Regulations

STUDIES IN SOCIAL MEDlClNE Allan M. Brandt and Larry R. Churchill, editors

O 1999 The University of North Carolina Press All rights reserved Manufactured in the United States of America The paper in this book meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources. Library of Congress Cataloging-in-Publication Data Beyond regulations: ethics in human subjects research 1 edited by Nancy M. P. King, Gail E. Henderson, Jane Stein. p. cm.-(Studies in social medicine) Includes bibli~~raphical references and index. ISBN 0-8078-2468-2 (cloth: alk. paper).-~s~N 0-8078-4770-4 (pbk.: alk. paper) I. Human experimentation in medicine-Moral and ethical aspects-Case studies. I. King, Nancy M. P. 11. Henderson, Gail, 1949- . 111. Stein, Jane. IV. Series. ~ 8 5 3 . ~ 8 I999 ~3 174l.28-DCZI

98-29189 CIP

Contents

Foreword ix ALLAN M . BRANDT AND LARRY R. CHURCHILL

Acknowledgments

xi

Introduction. Relationships in Research: A New Paradigm

I

NANCY M , P. KING, GAIL E. HENDERSON, AND JANESTEIN

KEYNOTE ESSAY

lntroduction

21

Is Ethics Universal?: Gender, Science, and Culture in Reproductive Health Research 23 RUTH MACKLIN

CASE 1. CONTRACTS AND COVENANTS

lntroduction

47

Contract and Covenant in Curacao: Reciprocal Relationships in Scholarly Research 49 ALAN F. BENJAMIN

I. Contract and Covenant in Ethnographic Research RENBE C . FOX

COMMENTARY

COMMENTARY

2.

The Gaze of Scholars and Subjects: Roles, and Obligations in Ethnographic Research 72

Relationships,

S U E E. ESTROFF

CASE 2. COMMUNITY-BASED HIV RESEARCH

lntroduction

83

67

VI CONTENTS

Community Assessment and Perceptions: Preparation for HIVVaccine Efficacy Trials 85 LYNN BLANCHARD

Community Advisory Board-Investigator Relationships in Community-Based HIVIAIDS Research 94

COMMENTARY I.

RONALD P. STRAUSS

Research Partnerships and People "at Risk": HIV Vaccine Efficacy Trials and African American Communities 102

COMMENTARY 2.

KEITH A. W A I L 0 0

CASE 3. CORPORATE SPONSORSHIP OF RESEARCH

lntroduction

111

Truth-in-funding: Studying the Infant-feeding Controversy with Industry Support 113 BARRY M. P O P K I N

Bias and Conflicts of Interest in Science: Controversial Industry Funding of Infant-feeding Studies 123

COMMENTARY I.

LORETTA M . KOPELMAN

Context and Community: Assessing the Ethics of Industry-funded Research 128

COMMENTARY 2.

ALLAN M. BRANDT AND LARA FREIDENFELDS

CASE 4. RISK AND TRUST IN ABORTION RESEARCH

lntroduction

137

Research on lnduced Abortion in Argentina: Avoiding Self-Incrimination 140 JORGE BALAN AND SILVINA RAMOS Research on lnduced Abortion in Argentina: Avoiding Self-Incrimination 147

COMMENTARY I.

C Y N T H I A WASZAK

COMMENTARY 2.

The Contexts of Social Research 153

BARBARA ENTWISLE

CASE 5. STUDYING MALTREATMENT IN FAMILIES

lntroduction

161

CONTENTS VII

Maltreatment in Families: A Research Dilemma

163

DESMOND K. RUNYAN

Child Abuse Research: Can Ethical Standards Be the Same in Developed and Developing Countries? 171

COMMENTARY I.

MARCELA ARACENA ALVAREZ

Research in Distressed Families: How Should Societies Make Judgments about Parents and Children? 180

COMMENTARY 2.

NANCY M . P. K I N G

CASE 6. WHOSE CONSENT?

Introduction

189

Can Community Consultation Substitute for lnformed Consent in Emergency Medicine Research? 191 ERNEST N. KRAYBILL A N D B. SUSAN BAUER

COMMENTARY

I. Can Community Consultation Substitute for lnformed

Consent in Emergency Medicine Research: A Response 199 J . PAT B R O W D E R

COMMENTARY 2.

Medical Research: Using a New P a r a d i p Where the 204

Old Might Do

NANCY M. P. K I N G

Conclusion. Regulations and Relationships: Toward a New Synthesis 213 NANCY M . P. KING, GAIL E. HENDERSON, A N D JANE STEIN

APPENDIX A.

Nuremberg Code

APPENDIX B. Federal Common

225

Rule

227

APPENDIX C.

ClOMS Epidemiological Research Guidelines

APPENDIX D.

ClOMS Biomedical Research Guidelines

239

247

Contract between Alan Benjamin and Congregation MikvC IsraeLEmanuel 252

APPENDIX E.

APPENDIX F.

Letter from Alan Benjamin to the Congregation

References 258 Contributors 272 Index

275

zyy

Foreword

This book is the initial volume in a new series, Studies in Social Medicine, and provides the occasion to describe the aims and ambitions of the series as well as to indicate why this volume helps to define it. For several decades some humanities and social science disciplineshave brought their methods and perspectives to bear upon medicine and health care. Medical sociology and anthropology, health economics, the history of medicine and science, and bioethics, for example, are well-established fields of study and areas of graduate specialization. We seek in this series to move beyond these wellestablished areas of scholarship and explore in more detail the powerful and pervasive ways that medicine shapes and is shaped by social forces. Such work requires a different and more challenging intellectual engagement; it demands more than simply using medicine and health care as the "site" of one's work, "applying" principles and methodological tools honed in other settings to the arena of medicine and the life sciences. Rather, this series is established in the belief that medicine and society interpenetrate and interpret each other, that the shaping is mutual and the lines of influence run both ways. Studies in Social Medicine aims to publish a wide-ranging collection of books that exhibit a fresh understanding of medicine as a social and cultural force and not simply a biological or scientific force with social implications. Hence, not all works in medical ethics, medical history, or sociology will belong in this series, but only those that seek to view society and medicine stereoscopically, combining multiple viewpoints and thereby aiming for clarity and insight at a greater depth. We inaugurate this series in the conviction that such new perspectives and approaches are necessary to rightly understand the role of medicine in American culture and history. Through this series we also hope to perpetuate and enhance interdisciplinary work in the humanities, social sciences, and health sciences. Much of the knowledge required to properly educate health professionals issues from the crossing of professional and disciplinary boundaries, and this will be increasingly true in the years ahead. Near the beginning of the twentieth century, Alfred North Whitehead warned of the hazards created when power is entrusted to those who are nar-

X

FOREWORD

rawly trained and lack broad understanding and conceptual agility. As this century comes to a close, it is even more evident that health professionals need to be literate in scholarship that places their work and its significance within a social matrix. Studies in Social Medicine is part of this effort in professional literacy Beyond Regulations: Ethics in Human Subjects Research is an exemplary volume for these ambitions. King, Henderson, and Stein present here a casebook with commentaries, favoring practical engagement with concrete ethical problems that researchers have encountered over the more typical deductive and theoretical approaches. Although a number of books are available on the ethics of research with human subjects, there is little systematic treatment of this area from a cross-cultural perspective. Certainly there is no sustained examination of the normative dimensions of relationships created in the process of social and medical research. The problem areas explored here are particular and situated but common enough for researchers to readily identify. This approach should make the volume useful for graduate education in the social and health sciences as well as for students of ethics. Those who have followed the recent debates on the ethics of federally financed AZT trials in Africa, or the HIV vaccine trials scheduled to begin in Thailand in 1998 and in Brazil and Uganda in 1999, will find this volume especially relevant. In addition to these pedagogical aims, King, Henderson, and Stein also challenge the reader theoretically in the introductory and concluding essays. It would be easy to assume that this volume, like most other ethics casebooks, is simply a skillful application of known principles to new and more complex relationships. Contrary to such an assumption, the editors argue that a relational paradigm can provide an important critique and complement for the principlist approach that underwrites federal regulations and legal requirements. Assessing the various contributions from both principlist and relational approaches for various types of research constitutes an important ongoing agenda. Thus, this volume is not only a contribution toward a more coherent, inclusive, and balanced approach to the ethics of research, but also a contribution to ethical theory more generally. Alhn M. Brandt Larry R. Churchill

Acknowledgments

This volume is based on material prepared for a conference held in Chapel Hill, North Carolina, on November 6-8,1995. "From 'Regs' to Relationships: Reexamining Research Ethics" was supported by a grant from the University of North Carolina, University Center for International Studies. In addition to the editors of this volume, the following people served on the year-long organizing committee, which gave shape to both the conference and this book: Paul Ilecki, Kate Macintyre, Kevin Moore, Paul Price, Emily Stern, and Matthew Tiedeman. We thank the graduate student rapporteurs, whose notes taken during the conference significantly assisted us in preparing this book: Amy Cunningham, Christina Fowler, James Dougherty, and Kate Macintyre. Secretarial assistance was ably provided by Jackie Jones of the Department of Social Medicine. At a research-oriented institution such as the University of North Carolina at Chapel Hill, the ethics of human subjects research is of particular concern. Appropriately, the conference was a pan-university event, not only reflecting the relevance of the topic and our self-consciously interdisciplinary approach, but also evidenced by the financial support offered by offices from the entire university. The following departments, centers, and organizations contributed funds to support the conference, and thus helped support the development of this book: University Center for International Studies; Office of the Vice Chancellor for Graduate Studies and Research; Cecil G. Sheps Center for Health Services Research; Department of Social Medicine; Center for Health Promotion and Disease Prevention; Carolina Population Center; Department of Surgery; Department of Epidemiology; Burroughs-Wellcome Company; Office of the Vice Chancellor for Health Ailairs; Department of Gastrointestinal Biology and Disease; Department of Psychiatry; Department of Obstetrics and Gynecology; Office of International Affairs, School of Medicine; Department of Pharmacology; Protein EngineeringIGenetics Training Program; Department of Microbiology and Immunology; and Department of Pathology. Our greatest debt, however, is owed to the conference participants, an audience that was clearly as expert as the speakers. Fifty people, including faculty and other interested researchers from Chapel Hill and elsewhere, were invited to at-

XI1 ACKNOWLEDGMENTS

tend because they were dealing with similar issues in their own research. Over two and a half days, they worked intensively with the organizers, case presenters, and commentators, in large and small group settings, shaping a truly interactive conference and continually pushing our thinking forward. This book is the result of all those efforts. We hope that it will stimulate many such discussions.

Beyond Regulations

2

INTRODUCTION

Across a broad range of disciplines and types of human subjects researchin biomedicine, the social sciences, and the humanities-researchers, researchers in training, and their teachers, as well as those charged with evaluating and overseeing research, are increasingly struggling with questions about the ethics of relationships between researchers and their subjects, funders, institutions, and colleagues. All those who are-or should be-asking these questions are the audience for this book. To adequately address the ethics of human subjects research, we must develop a shared set of ethical concepts, a common vocabulary, and a relationship-focused, richly contextual perspective. The first goal of this book is to contribute to this process through close examination of cases designed to elicit common issues and problems from a variety of research settings and scholarly perspectives. Based on this goal, the book's second objective is to develop a new, relationship-based paradigm for research ethics. Such a paradigm should enable scholars, researchers, and others involved in research design, evaluation, or approval to facilitate human subjects research that is sensitive and responsible as well as scientifically fruitful. The book's six cases were selected because they enable this audience to consider research as embodied in a set of relationships between researchers, subjects, communities, funders, institutions, and governments. Examining these relationships illuminates ethical dilemmas in a nuanced and realistic way, for all academic disciplines that engage in research with human subjects-including social science, public health, and clinical medicine. The relationships perspective is relevant to the broadest range of research questions and methods. The cases we include here feature a variety of methods and disciplines: ethnography, survey and interview methods, community-based research, longitudinal multi-method studies, and randomized controlled clinical trials. Most of the cases and commentators come from the University of North Carolina, but it should be obvious that this is not because U N C researchers and scholars are more ethical than anyone else. Instead, it is testimony to the pervasiveness of the issues that so many of them could be addressed in cases that were, for us, close to home. Similarly rich and provocative cases can surely be found in abundance at most research universities and elsewhere. In the first case, an anthropologist proposes to do ethnographic research in a small Jewish community in Cura~ao.In a break with traditional researcher-subject relations, community representatives require that the researcher sign a contract giving them complete veto power over any publications before they will give their approval to the research. What are the implications of such a contract? Does it undermine academic freedom? The second case is also about community, but in this case it is a community imagined by the National Institutes of Health and the Centers for Disease Control-persons at risk for HIVIAIDS. The goal ap-

RELATIONSHIPS IN RESEARCH 3

pears to be for university research teams to pave the way for successful HIV vaccine trials by studying and gaining the cooperation of at-risk populations. In this case, African Americans in Durham, North Carolina, may not constitute a single self-identified community, and we therefore wonder who should or can speak for them. In the third case, a nutrition economist accepted a large !gift from the Nest16 company in the early 1980s to support an ambitious research plan for studying infant feeding practices in the Philippines. The influence of the funder-any funder- on both the research process and the reporting of results is at issue, as are the effects of the decision of a single researcher on the rights and interests of the larger research community. In the fourth case, a research project studying illegal abortions in Argentina raises difficult logistical and ethical dilemmas. Because these investigators are not subject to the same institutional review requirements as exist in the United States, resolution of these dilemmas falls to them alone. Their justification for not informing potential subjects of the nature of the research is based upon a profound and ongoing sense of relationship with the women they are studying. But is this enough? The fifth case also discusses the dilemma of how to tell subjects about the risks of participation in a study-here a longitudinal investigation of the impact of child abuse. The added twist is the inherent conflict between the legal requirement to report abuse (thus protecting children) and the ethical obligation of a researcher to protect the confidentiality of all information provided by study subjects (both parents and children). The final case is about a 1996 federal replatory exception to the informed consent requirement in "emergency research" for life-threatening illness or injury. The United States government determined that in this situation, the need for progress in research outweighs the imperative of informed consent, although some form of "community consultation and public disclosure" is required. While the issue of community is problematic for subjects in emergency rooms, equally difficult is the potential for conflict between the interests of the research community and their funding agencies, on the one hand, and on the other the potential subjects whose voices will not now be heard. These cases portray complex dilemmas because research relationships are complex, contradictory, and always changing. All of them address issues not commonly encountered in traditional discussions of research ethics-issues such as power, control, and trust. But most importantly, each of these issues is real, current, and potentially troubling for researchers engaged in human subjects research in any field. The range of disciplines engaged in research with human subjects is broad, and this book is informed by the perspectives of internal medicine, pediatrics, psychiatry, surgery, preventive medicine, dentistry, nursing, epidemiology and public health, sociology, anthropology, ethics and philosophy, history, law,

4

INTRODUCTION

health policy, and health services research. Each of these disciplines, and numerous others, has its own history and tradition of reflection on the moral dilemmas presented by research with human subjects, and makes unique contributions to the continuing debate about them. Three paths converged in shaping the concerns of this book. First are the continuing- stories of error and abuse in American research with human subjects, despite a nearly thirty-year experience in applying detailed federal regulations to shape and govern the design and conduct of that research (Advisory Committee on Human Radiation Experiments 1795; Faden and Beauchamp 1786; Rothman 1771). Next come the arguments, played out on the international stage, that those American regulations-gounded in the principles of autonomy, beneficence, and justice-should not apply in all cultures. For example, cultures that define or value those principles differently, or that do not acknowledge them or acknowledge principle-based moral reasoning at all, must be accorded respect (Christakis 1972; Newton 1770). Finally comes the counterargument that the regulations are not enough, that by virtue of their roots in acontextual moral principles that originated in the European Enlightenment, they fail to answer, and may even be incapable of asking, some intractable ethical questions (Angell 1788; Barry 1788). The regulations address some of these questions-for example, how to design and carry out a good informed consent process-but do not answer them completely or well. Other questions, such as whether and how research direction should be influenced by funders or by subjects, simply fall outside the regulations' purview altogether. It thus seems to us that, although the term is overused, we might be in the middle of a paradigm shift. Ethics generally, and bioethics in particular, have recently been debating whether the invocation of principles is adequate for moral analysis of the complex and diverse questions that arise today (Arras and Steinbock 1795; Beauchamp and Childress 1774). A persuasive case can be made that principlism, by itself, is a moral perspective that is insufficiently flexible. A variety of new moral perspectives, including narrative ethics (Churchill 1997; Clouser and Hawkins 1776; Hunter 1791, 1775), feminist ethics (Kittay and Meyers 1787; Mahowald 1993; McCarrick and Darragh 1776; Tong 1993; Wolf 1976), revivals of virtue ethics (MacIntyre 1784) and casuistry (Jonsen and Toulmin 1788), and a renewed attention to the meaning of culture, have begun to add richness and nuance to the portraits of moral problems, moving analysis away from a mathematized computation of moral "answers" toward a focus on mutuality. These debates and critiques are flourishing in a broad range of academic disciplines engaged in research with human subjects. We have attempted to characterize the current discussion as shifting from a regulatory, principlist p a r a d i p (based on rules derived from principle-based reasoning about moral problems in human

RELATIONSHIPS IN RESEARCH 5

subjects research) to a relationships paradigm. This new paradigm emphasizes relationships, interactions, power, responsibility, and contextual and historical considerations in examining moral issues. New paradigms, however, do not just appear. They evolve out of older ones, and those who attempt to implement and justify them must struggle with what to keep from earlier systems and how to defend their rejection of the rest. This is never easy, but it is particularly challenging when it takes place across academic disciplines, each with its own history, tradition, and language. For this reason, we have grounded this book in case studies from a variety of fields that present a range of ethical issues. Underlying all of them, however, is this recognition: traditional ways of overseeing the ethics of research are insufficient to address issues faced by researchers who are concerned about their responsibilities to and relationships with those whom they study, their funding organizations, their professional peers, and the audiences for their research.

A Brief History: Images and Ethics of Human Subjects Research

Problem Cases The ethics of human subjects research has enjoyed a resurgence of public attention in recent years. Most recently, this attention has been fueled by the President's Advisory Committee on Human Radiation Experiments, which began its work in 1994 and issued its Final Report in 1995. The Advisory Committee, which examined radiation research that took place during the Cold War era, concluded that lapses in protection of the rights, interests, and safety of human subjects are distressingly common even today, despite the elaborate federal regulatory process for the review and approval of human subjects research (Advisory Committee on Human Radiation Experiments 1995). Research with human subjects has a long tradition in many fields besides biomedicine, including social science, public health, epidemiology, and even the humanities. Concerns about the ethics of research with human subjects could be said to predate the existence of modern medical and scientific research, but certainly go back as far as Claude Bernard and other physician-researchers of the nineteenth cent~ry.~ Public interest and attention has generally been associated with problem cases and the governmental actions they engender. For example, the passage in 1906 of the Pure Food and Drugs Act, and in 1938 of the Food, Drug, and Cosmetic Act, came in response to concerns about quackery and the patent medicine industry. Similarly, the Drug Amendments of 1962, which laid the groundwork for close governmental control over the testing and approval of new drugs, were

6

INTRODUCTION

passed as a result of the Thalidomide t r a g e d ~Not . ~ surprisingly, most problem cases have concerned medical and scientific research that has caused harm to subjects. However, in a number of cases, violations of subjects' rights-for example, their privacy, or their right to choose to participate in research without coercion or deception-have also attracted public concern. Although the twenty-year period following World War I1 appears to be a time during which research ethics was "discovered," it is better understood as a period during which the concerns occasionally voiced throughout the first half of the century, by a wide range of professionals and scholars, and even by the courts, finally took form. "At the end of World War 11, there were virtually no noteworthy formal pidelines or codes governing research with human subjects. But . . . over twenty guidelines and codes, designed to protect subjects, [were] passed between 1948 and 1968 by major organizations" (Faden and Beauchamp 1986,186, footnote omitted). The increased attention to control of human subjects research was largely spurred by the Nuremberg trial of Nazi physicians, whose deadly "research" epitomized the horrors of the Nazi regime and resulted in the influential Nuremberg Code of research ethics (see Appendix A). New concerns about research ethics, however, were surely also influenced by profound and ongoing changes in American society, including attention to civil rights and individual freedoms, the development of formal professional organizations, and the growth of research, industry, and the university (Faden and Beauchamp 1986; Rothman 1991). A handful of highly publicized instances of obviously unethical research helped command attention to the issues. The most notorious cases included three medical research projects-the U.S. Public Health Service's Tuskegee syphilis study, the Jewish Chronic Disease Hospital case, and the Willowbrook study-and two psychology research projects- the Zimbardo prison experiment and Stanley Milgram's obedience study. All posed risks of serious physical or psychological harm to subjects, including failure to treat serious disease in subjects, exposing subjects to the risk of disease, placing subjects in distressing situations, and putting subjects at risk of shame or embarrassment. All also involved, at best, flawed or incomplete consent processes, or at worst, deception of subjects. In the Tuskegee syphilis study, poorly educated black men from the rural South were enrolled in a U.S. Public Health Service study to examine the progression of untreated syphilis. The subjects were not told that they were participants in research, and were deliberately deceived about the nature of the interventions they received. Worst of all, available treatments-including highly effective antibiotics-were kept from them. When this study came to public attention, it had been going on for some forty years, and revelations about it caused an uproar (Brandt 1978; Faden and Beauchamp 1986,165- 67; Jones 1992). The Jewish Chronic Disease Hospital case and the Willowbrook study both made use of decisionally incapable institutionalized patients for studies unrelated -

-

RELATIONSHIPS IN RESEARCH 7

to the conditions resulting in their institutionalization and incapacity. The Jewish Chronic Disease Hospital case involved elderly subjects, some of whom were incompetent, who were given injections of live cancer cells without their knowledge or consent. The researchers claimed to be convinced that the injections could not cause cancer, and that examining how long it took for a debilitated subject's immune system to reject the foreign cells would be important information in the war against cancer. However, the researchers had bypassed the hospital's research review committee, had failed to consult the subjects' physicians, and had determined that informing the subjects would only alarm them (Faden and Beauchamp 1986,161- 62). At Willowbrook, an institution for mentally disabled children, researchers sought to develop a hepatitis vaccine by studying children whom they had deliberately infected with the virus. Although this research was questioned on a variety of ethical grounds, it continued for several decades (Faden and Beauchamp 1986,163- 64). Stanley Milgram'.. obedience research used deception to examine the willingness of subjects to obey the commands of an authority figure. Subjects were told that they were participating in an experiment to determine whether aversive conditioning improved learning and memory. Each was instructed to administer electric shocks of increasing severity to a false subject, a confederate working in concert with the experimenter who gave wrong answers in a word association test. The false subject did not in fact receive any shocks, but screamed as if in pain, begged to be released, and otherwise simulated great injury. Subjects who objected to continuing to give the shocks were repeatedly told to continue for the advancement of science. Extreme distress usually accompanied obedience, but complete debriefing and reconciliation with the unharmed false subject followed the completion of each subject's participation. Milgram's research, in which sixty percent of subjects obeyed the authority figure a11 the way to the most severe "shock," arguably produced important results, but critics maintained that the price was far too high (Faden and Beauchamp 1986,174-77). The Zimbardo prison experiment used no deception, but was criticized, as was Milgram's research, for underestimating the likelihood and severity of psychological harm to subjects. Zimbardo set up a mock prison using volunteer college student subjects to assume the roles of prisoners and guards. The problem was how we11 it worked; student "pards" brutalized student "prisoners," and these results raised questions about how well researchers must anticipate possible outcomes and describe them to potential subjects (Faden and Beauchamp 1986,178-79). Of the well-known cases from this period, only the Wichita Jury Study and Laud Humphrcys's "tearoom trade" research, which were conducted without subjects' consent, were condemned solely because they violated the rights and interests of subjects without posing risks of harm to them. In the Wichita study, researchers secretly recorded and examined the deliberations of six juries. The jurors

8 INTRODUCTION

never knew that they were research subjects or that their deliberations were recorded, and this breach of the privacy and confidentiality of the jury process was bitterly controversial in both law and academia (Faden and Beauchamp 1986, 173-75). Humphreys observed the behavior of homosexuals in public restrooms without their consent, and in some instances surreptitiously determined their identities and later interviewed them under false pretenses. Again, it was argued that his results were important-they cast doubt on many stereotypes about homosexuality-but that the invasions of privacy and confidentiality of the unwitting subjects simply could not be outweighed by any benefits from the research (Faden and Beauchamp 1986,177-78). These problem cases focused attention on the development and refinement of rules, policies, and guidelines, resulting in some significant advances in public consciousness about and professional regulation of human subjects research. The effectiveness of such regulation depends, however, on its being understood as the vehicle for the application of principles to particular circumstances. And in the latter part of the twentieth century, this principlist paradigm of moral reasoning has been challenged.

Moral Principles All of the codes, guidelines, and even discussion of problem cases that emerged from the latter half of the twentieth century found their underpinnings in the moral philosophy of the European Enlightenment. The enunciation of principles and rules, and the use of logic and even of mathematical calculation to determine moral behavior, profoundly influenced the development of research ethics, just as it influenced Western intellectual history in general (Beauchamp and Childress 1994; MacIntyre 1984). The current system of federal regulation of research with human subjects is exemplified by the principlist paradigm in the Belmont Report, a document produced in 1978 by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research to ground and justify the federal regulations. The Belmont Report sets forth the three moral principles of respect for persons, beneficence, and justice as the basis for the protection of human subjects, applies those principles to human subjects research, and deduces from them a set of requirements for research with human subjects (National Commission 1978). Respect for persons, the first principle, includes both respect for the choices of autonomous persons and protection of the rights, needs, and interests of persons who lack the capacity to decide for themselves or have constraints upon their freedom of choice. This principle requires that the autonomy of subjects and potential subjects be supported in the informed consent process, and that subjects and

RELATIONSHIPS IN RESEARCH

9

potential subjects who may lack autonomy, such as children, persons with mental disabilities, or the institutionalized, be protected frorLL exploitation. Thus, for example, autonomous persons need more than information. They may need information conveyed at a level appropriate to their education or in their native language. In addition, persons whose decisionmaking capacity is questionable or limited may be precluded from taking part in some potentially risky research, must have legally authorized decisionmakers to act on their behalf, and may be asked to provide assent to participation after information has been provided at a level appropriate to their understanding. Finally, persons who are incarcerated or dependent upon the state may need additional protection against the risk of undue influence in the consent process. Beneficence, a familiar and well-understood principle, has tended to dominate discussion of research ethics. Like autonomy, it has two aspects: beneficence, or the duty to do good, and nonmaleficence, or the duty to refrain from causing or increasing harm. Thus, beneficence broadly understood includes both the familiar "do no harm" and the obligation to maximize the possibility of benefits and minimize the risks of harm. Beneficence in the research context governs both the design and conduct of research, including not only immediate risks and benefits to subjects, but consideration of whether the research is "worth doing" in light of long-term expected results and their societal implications. Justice is the least well understood and most neglected of the ethical principles presented in the Belmont Report. Justice is a principle designed to eliminate biases against groups of people; it is the principle that draws our focus from individuals to groups. However, justice can be interpreted in different ways (Mastroianni et al. 1994). The Belmont Report considers justice as a fair distribution of burdens and benefits among cultural, social, sexual, racial, and ethnic groups. As with each of the principles, what this means must be determined in the context of each specific research project, and this is of course no simple task. The Belmont Report's consideration of justice focuses principally on subject selection issues. However, in much of the literature and discussion related to changing research paradigms, justice has begun to appear as a stronger, more active principle-one that is related to past, present, and future distributions of power. This evolution reflects the view that research is part of a comprehensive system that unfairly distributes power and its perquisites, and that it is incumbent on researchers to compensate for both the past sins of research and the state of the world. Thus, many researchers-including feminist researchers, new paradigm researchers, and participatory action researchers-include the empowerment of research subjects as an ethical imperative (Stein 1997). This addition has resulted in an active dialogue on the need to recast such traditional research constructs as objectivity, bias, reliability, and validity It has led researchers to question their own

10

INTRODUCTION

positions of privilege and their relationships with powerful others, such as funders, policymakers, and their academic community, causing them to rethink what research they are willing to do and to pay much closer attention to how they ought to do it.

Domestic and International Research: Do the Regulations Apply?

Policy: The Regulations Current United States regulations governing research with human subjects, known as the federal "Common Rule," evolved over the past twenty-five years. The process began with a 1971governmental Guide (Public Health Service 1971), which was transformed into regulations in 1974.~ Substantially revised regulations followed in 1981 (Department of Health and Human Services @I), along with a second Commission,5 which produced even more impressive and influential writings (Advisory Committee on Human Radiation Experiments 1995; Faden and Beauchamp 1986). The consolidation of all federal regulations governing human subjects research into the Common Rule (see Appendix B) followed in 1991. Briefly, the Common Rule, which applies to all human subjects research conducted under the auspices of institutions receiving federal funding, defines research as "a systematic investigation . . . designed to develop or contribute to generalizable knowledge." It describes the research that is subject to the regulations and requires the establishment of Institutional Review Boards (IRB) to review and approve proposed research. In addition, the Common Rule establishes the conditions of that approval, which include extensive informed consent requirements, a requirement that risks to subjects be minimized and that risks be reasonable in relation to anticipated benefits, protection of privacy and confidentiality where possible, equitable selection of subjects, and appropriate protections for special populations of subjects. From the beginning of this expanded regulatory attention to research ethics, concern and controversy existed over the appropriateness of including "behavioral" research under the aegis of federal oversight. Social scientists in particular were dismayed, even indignant, to find that their work would be reviewed alongside clinical trials and other research posing physical risks to subjects, and would be subjected to the same standards of informed consent, which they viewed as inconveniences at best, and at worst as impediments that could render some research impossible to c o n d ~ c t . ~ Although the Belrnont Report featured respect for persons as the primary principle underpinning ethical research, beneficence continued to seem more significant to researchers. Much of the controversy surrounding the regulation of behavioral research stemmed from this view that protecting subjects from physical

RELATIONSHIPS IN RESEARCH 11

harm is more important than protecting them from wrongs, such as deception, failure to provide adequate information, breaches of privacy and confidentiality, or undue influence-that is, that violating the principle of beneficence is worse than violating the principle of respect for persons. Some of the research ethics literature during this twenty-five-year period has attempted to draw discipline-based distinctions between types of research and link those research categories to different priorities and concerns regarding ethics in human subjects research. It has been argued, for example, that biomedical research, because it is more likely to involve physical interventions, is more likely to cause or risk harm to subjects, and that social science research, because it is more likely to be limited to information-gathering, is more likely to cause or risk wrongs to subjects-that is, violations of their rights and interests, including breaches of privacy and confidentiality and failure to protect their autonomy (Beauchamp et al. 1982; Wax and Cassell 1981). Indeed, behavioral and social science researchers recognized early in the development of the regulations that some of their research designs, including ethnography, participant observation, and other longitudinal and interactive strategies, not only did not fit cleanly into the scheme of federal research regulation but also presented dilemmas of moral relationship that fell outside the regulatory paradigm's reach. There is a considerable literature of discussion and critique of research within the social sciences that has many rich insights to offer to consideration of the relationships paradigm (Fine 1993;Homan 1992).' In our view, however, distinguishing between biomedical and social science research represents no more than the crudest cut at the issues of concern to most researchers, and misrepresents the importance of evaluating the potential harms, wrongs, benefits, and burdens of all proposed research. Although some research does indeed fit these disciplinary stereotypes, much does not. Moreover, the prevalence of this disciplinary stereotyping has tended to obscure the existence of crossdisciplinary concerns that depend more on the particulars of the research than on how it is labeled. A few illustrations clarify this point. How should biomedical researchers think about the long-term relationships they develop with chronically ill patient-subjects, except by looking to the longterm relationships developed between nonbiomedical researchers and the populations with whom they do fieldwork? By contrast, is a research project that involves data collection by anonymous questionnaire very different from a project that collects anonymous tissue samples if the focus of both is, say, genetic risk or HIV infection?And how should questions of consent and confidentiality in crosscultural research be addressed, if not by the sharing of experiences and perspectives among researchers of a variety of disciplines who have faced such issues, both internationally and at home? Focusing on the relationships involved in all human subjects research serves not only as a means of addressing the characteristicsof re-

12

INTRODUCTION

search that give rise to particular ethical concerns, but also as a means of beginning to structure meaningful analysis and successful solutions.

CIOMS Guidelines Recent challenges to the traditional Western philosophical perspective on ethics have been articulated in the research context in two ways. Attacks on principlism are intended to dethrone the traditional triumvirate of bioethics principlesautonomy, beneficence, and justice-that underlies the regulation of research. "Western moral imperialism," with its emphasis on individualism, is denounced as holding researchers in developing countries to irrelevant and impossible-tomeet standards. This position both highlights the communitarian ethics of nonWestern traditions and seeks to promote the conduct of human subjects research in developing countries outside the standards represented by the principle-based research regulatory paradigm. The principlist underpinnings of Western moral theory have had such petvasive influence that even international research guidelines have conformed to that model. This includes two recent and influential sets of guidelines for epidemiological research (1991, see Appendix C) and biomedical research (1993, see Appendix D) produced by the Council for International Organizations of Medical Sciences (CIOMS). These two sets of CIOMS guidelines adopt the approach and organization of the Belmont Report and of the Common Rule, but they also address key points and issues in light of local conditions. Both these sets of guidelines are the result of many iterations and revisions, as their authors have attempted to respond to various critiques. Throughout these efforts, the guidelines have continued to reflect the basic premises of principlist ethics (Bankowski and Levine 1992; Ringheim 1995). Both sets of guidelines set up individual informed consent as a priority. This has been challenged by those who argue that the Western focus on individualism is being unfairly applied to cultures that are more communal or family-focused in their decisionmaking practices (Christakis 1992; Newton 1990). It has been applauded by others, however, who argue that when emphasis on individual consent is reduced, it permits the exploitation of women and encourages undue reliance on community "leaders" who are not necessarily endowed with authority by community members (Ijsselmuiden and Faden 1992). The guidelines for biomedical research specifically address research involving subjects in underdeveloped countries, the equitable distribution of burdens and benefits of research, and the obligations of sponsoring and host countries in international research collaborations. Guidelines addressing the first two issues focus on preventing the exploitation of subjects and communities of subjects; the p i d e line on the third issue requires research to first meet the standards of the spon-

RELATIONSHIPS IN RESEARCH

13

soring country and then the standards of the host country. The guidelines for epidemiological research consider the acceptability of community agreement where "collective decisionmaking is customary," and also address concerns including- "selective disclosure" of information to prospective subjects, undue influence and undue inducements to participation, respect for social mores, cultural sensitivity, and conflict of interest. The requirement that research meet the standards of the sponsoring country first and then the host country has drawn both praise and criticism-praise because it prevents the exploitation of host countries, criticism because it prioritizes the standards of the sponsoring (usually developed Western) country. This prioritization can lead to excessive attention to regulatory details like signature requirements for informed consent (often a barrier to research in countries with low literacy or with political circumstances that make signing one's name seem very risky), as well as to the tendency to take less seriously the review by the host country's IRB ("But our IRB has already approved this!"). The CIOMS attempt to take account of the needs and circumstances of the developing world thus can appear to be an effort to apply a superior (Western) model to inferior settings.

A Paradigm Shift: Are the Regulations Enough? The current wave of public interest in research ethics parallels a related but distinct set of ongoing concerns shared by researchers, scholars, and students worldwide: that the conceptual and regulatory framework for attention to ethics in research with human subjects is insufficiently rich and nuanced to be helpful in addressing certain key pervasive ethical issues (Kelman 1982; Phillips 1994). The principlist paradigm of moral theory that undergirds the oversight of human subjects research derives from the Enlightenment's attempt to develop and sustain a universalizable, acontextual, nonreligious morality. However, the late twentieth century has produced arguments from many quarters that the Enlightenment effort was misguided. Instead of hoping for a truly "rational" morality abstracted from emotion or circumstance, critics of principlism stress the inevitability and necessity of context and viewpoint in moral analysis, the impossibility of abstracting moral argument from emotion and experience, and the importance of considering social and political circumstances, and power relationships in particular, to give meaning to moral theory. Not surprisingly, concerns very similar to these have troubled researchers from a variety of disciplines whose work with subjects has dimension and durationwhereas those whose contact with subjects is a slice-of-time encounter have more often been content with rulebook-style moral guidance. While feminist philosophers have been criticizing the Kantian ethical perspective, ethnographers have

14

INTRODUCTION

been worrying about their depictions of their subject communities, and participatory researchers have been collaborating with communities to define research questions. At the same time, many epidemiologists have happily continued to compile surveys and crunch numbers, and physiologists pay healthy volunteers twenty dollars an hour to compare the effects of sports drinks and tap water, applying the regulations without thinking any more about it. The false opposition between "moral imperialism" and "cultural relativism" ought perhaps to be replaced with the realization that in some circumstances the regulations, and their underlying principles, come "close enough," and sometimes they do not. What we seek to discover is how to know what is "close enough," and how to proceed if it is not.

Toward a New Analysis This book draws on the perspectives of a multidisciplinary group of researchers and scholars from the humanities, social sciences, medicine, and health sciences, and asks readers to think about human subjects research in terms of the relationships it implicates. The keynote essay sets up the problem, ostensibly from the principlist paradigm. Six cases, each followed by two commentaries, afford the opportunity for detailed examination of the issues and the relationships among issues. From this examination we hope that readers can emerge with some shared perceptions, central themes, and continuing questions. Most importantly, we hope to move readers toward the recognition that a relationships paradigm is morally necessary to an adequate identification and consideration of the ethics of research with human subjects. The cases we chose for the conference and the volume are real, current, and complicated. Each case in this volume is controversial, in the sense that each has caused researchers and others to reexamine critically the relationships between the anticipation of benefit, the risk of harm, and the protection of the rights and interests of involved parties. All of the cases and commentaries in this volume examine or reexamine questions such as the following: What makes research worth doing? Should some research not be done? What constitutes an appropriate balance of risks and benefits? Whose interests count, and how much? What makes consent informed? How important is confidentiality, in the context of and with regard to the particulars of the research in question?At present, considering relationships can and should enable researchers and those who evaluate and oversee research to apply the principlist paradigm embodied in federal regulations thoughtfully and flexibly, with attention to their spirit and purpose. For the future, we anticipate that the relationships paradigm will be seen as having independent moral force. Its considerations should be incorporated into the regulatory process, and will ultimately alter the dominant paradigm underlying the regulations.

RELATIONSHIPS IN RESEARCH 1 5

A tentative outline ofwhat we are calling a paradigm shift, from the regulatory, principle-based paradigm to a paradigm based on relationships, helped to focus the presentations and discussions at the conference and the contents of this book. Although these paradigms are, in some sense, mythological creatures, they may also serve as guides for readers: The Principlist Paradigm:

-

Balancing principles-autonomy, beneficence, justice, informed consent, confidentiality Ethical universalism (not moral relativism)-truth (not stories) Atomistic focus-small frame, centered on individuals

The Relationships Paradigm (in process):

-

Layering of relevant relationships-individuals and groups Context-based-what are the relevant contexts? culture, gender, racelethnicity, community, place, others Cross-cutting issues, wider frames of reference Narrative focus Continuity-issues arise before and continue after projects Change-in relationships over time

In addition, we invite readers to apply a set of questions to their consideration of the cases: Define, describe, elaborate on the nature of the researcher-subject relationship. Do these cases have different takes on it? Is there an inherent "conflict of interest" between researchers and subjects? Is there a priority ranking of relevant relationships besides researcher-subject? Consider especially the difference between the researcher's relationships with individuals and the researcher's relationships with entities or groups. How much does/should the nature of the status quo and current practice affect the moral status of research? (That is, what does it mean to say that research is "necessary"?Who defines necessity and how does it count?) What are the problems of speaking for others in research?Are they different in consent, in research design, in drawing conclusions, and in dissemination of results? What problems do researchers identify with the regulatory, principlist paradigm? Are problems discrete or global?Can they be solved through better use of the principlist paradigm, or would they be better addressed by a new viewpoint? How can a relationships paradigm provide practical and theoretical guidance?

RELATIONSHIPS IN RESEARCH

17

Notes I. "From 'Regs' to Relationships: Reexamining Research Ethics," Chapel Hill, N.C., November 17-19,1995. z. See the extensive compendium of historical materials in Katz (1972). An eighteenthcentury English court decision, Slater u. BakerandStapleton, is one of the earliest reported disputes about experimental treatment. See discussion in Faden and Beauchamp (1986, 116 -18). Jenner's research on smallpox vaccination and Beaumont's investigations of digestion using Alexis St. Martin and his famous stomach wound represented major advances in research in that century and correspondingly raised many issues (Rothman 1991, 20 -22). Bernard's classic treatise was published in 1865 and first issued in English translation in 1927; it, along with Pasteur's dramatic rabies research, definitively situated ethical issues in the consciousness of researchers (Rothman 1991,22-24). But not only researchers were concerned about experimentation on human subjects. Gustave Flaubert's popular 1857 novel Madame Bouay contains a riveting description of an unsuccessful experimental surgical procedure to correct a clubfoot-complete with discussion of the decidedly mixed motivations of the surgeon, the desperation of the patient-subject, and the undue inducement represented by "free treatment" for the poor (Annas 1993,133-34). 3. The history of these laws is detailed in Edgar and Rothman (1990) and Faden and Beauchamp (1986). Thalidomide, a sedative used extensively by pregnant women outside the United States, but still considered experimental here at the time, resulted in the births of many children with serious limb deformities (King and Henderson 1991,1015-16). 4. At the same time, legislation established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, which authored, among other influential documents, the Belmont Report. 5. The President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research was authorized in 1978, after the expiration of the National Commission, and convened in 1980. 6. This controversy's long history is well illustrated by the sometimes vitriolic objections of social scientists, beginning in the 1970s (see, for example, Higman 1978) and continuing even today, to being included in research regulation. Many of the revisions in the 1981 regulations were designed to reduce the perceived burden on nonbiomedical researchers (Faden and Beauchamp 1986, 218-20). AS recently as 1994, the Tri-Council Working Group-representing Canada's Social Sciences and Humanities Research Council, Medical Research Council, and Natural Sciences and Engineering Research Council-which was given the task of drafting a new comprehensive "Tri-Council Code of Ethical Conduct for Research Involving Humans," was contending with the complaints of social scientists that there should be more than one code because of the differences between biomedical and social science research. The code, completed in 1997, and "Report on the Consultation Process," are available at http://www.hssfc.ca; click on General Interest. 7. See also Estroff's commentary in this volume.

22

KEYNOTE

care in their proper application, can really work all over the world. Indeed, it is central to her position that they must. Professor Macklin's perspective is challenging to proponents of a relationships paradigm, and it provoked vivid and passionate discussion in the genesis of this book. She presents the central challenge to the relationships paradigm. In essence, her argument is that the regulatory, principle-based paradigm is just fine, so long as we use it with appropriately nuanced attention to the concerns about power and viewpoint we have identified. In this view, the relationships model is not a new paradigm, but rather a set of considerations that add up to the context within which the current- principlist paradigm applies. Her concern is that what we call a new paradigm is merely descriptive ethics-a description of cultural practices that, however entrenched and venerated, are nonetheless wrong, because they fail to recognize and protect the moral imperatives of a genuine normative ethic. The archetypal example is perhaps that of the community where women do not give their individual consent to be research subjects, but the village elder consents for them. As is the case in cultures that practice female genital alteration, or where Islamic law restricts women's rights to drive, vote, or wear Western-style dress, many women in such a community would argue that individual autonomy is culturally meaningless, and that its imposition is moral imperialism. Professor Macklin responds that human rights transcend local customs. What is particularly striking about being in the middle of a paradigm shift is that for a period of time, both paradigms are true. Yet, if it is possible to organize all one's moral arguments and conclusions according to principle ethics, but if the chain of reasoning becomes longer and richer as we invoke more and more nuanced consideration of context, does there not come a point when the tail is wagging the dog? Is it not just as true that the consideration of power relationships, of women's circumstances, and of other contextual features has come to assume moral authority in its own right? If principles "work-in the sense that we can indeed invoke them to address moral problems in research-so what? Can we not say that relationships " w o r k this way too, and even sometimes work better than principles? To answer that last question affirmatively requires us to be able to say that we gain more than we lose in shifting paradigms. If we lose the capacity to engage in moral argument with researchers with whom we disagree, or if we can no longer think or speak about moral progress, or if we become lost in a sea of particulars when attempting to design and carry out ethical research, then a new paradigm does not serve us well. Professor Macklin has set forth the way things are now, in the middle of the paradigm shift, and challenged us to do better if we can. Her position is our starting point. The reader must judge whether the rest of the book shows that we can go further.

24

RUTH MACKLIN

of the information. Similarly, regulations require that ethical review committees (Institutional Review Boards, or IRBs, in the United States) evaluate risks and benefits and determine that "risks to subjects are reasonable in relation to anticipated benefits." ' How actually to weigh risks and benefits is a matter that no regulation could possibly specify. The description of the conference on which this volume is based begins by noting that "many ethical issues that arise in human subjects research cannot be answered readily by applying the federal regulations that govern this research in the United States." With that proposition I entirely agree. The conference description goes on to question "the extent to which a U.S. bioethics framework ought to be applied internationally." It is impossible to provide a fully detailed answer to this challenging question; I shall attempt the more modest task of laying the goundwork for discussing the limitations of regulations in the quest for a sound research ethics. In this paper I contend that the well-known and widely accepted ethical principles governing human subjects research are or ought to be universal in application. I argue for this proposition by distinguishing between descriptive ethics and normative ethics, between actual norms or customary practices and ethical ideals. . Additional distinctions need to be made between procedures stipulated by regulations, such as the need for written consent forms, and the fundamental ethical principles underlying this and other specific procedures. In the final section of the paper, I provide illustrations to show some ways in which research ethics are broader and deeper than any actual or possible regulations could encompass. Just as society's laws could not and should not embody all the ethical features of human relations and interactions, so too are regulations the bearers only of minimal ethical standards. - -

-

-

The Universality of Ethical Principles Governing Research The United States has elaborate and detailed federal regulations pertaining to biomedical and behavioral research using human subjects, whereas many countries have few or no laws or regulations in this area and others have the barest minimum. Does that mean researchers in those countries are not bound to adhere to ethical standards in the conduct of research? Does it mean that an imposition in those places of fundamental ethical principles recognized in North America and Western Europe is a form of ethical imperialism? My answer is that the ethical requirements for human subjects research rest on universal ethical principles, even if those principles are not recognized or adhered to in all parts of the world.2 To understand what this statement entails and, equally importantly, what it does not entail, several caveats are in order.

IS ETHICS UNIVERSAL? 25

The first caveat is not to commit the genetic fallacy. Just because certain ethical principles were first articulated in Western philosophy does not mean that such principles apply only in Western or Northern countries. Just because these ethical principles first found expression in governmental regulations applied to research in the United States does not mean that they are not equally relevant elsewhere. The origins of ethical principles are one thing; their domain of applicability is another. The second caveat is not to confuse the universality of principles with the quite different notion of an "absolutist ethics." Universality refers to the scope of applicability of ethical principles, whereas absolutism implies an exceptionless set of moral rules or prescriptions. The principles of bioethics are very general, and require interpretation in light of relevant empirical facts and contexts before they can be applied. Indeed, the principles are so general that they have been criticized as being "empty." In contrast, "moral absolutes" are ethical judgments with a specific content stating that something is "alwaysn obligatory or "never" permissible, such as Immanuel Kant's absolute prohibition against lying. A third caveat is that principles are necessary but not sufficient for a rich ethical analysis of human subjects research, among other human endeavors. A final caveat is the need to distinguish between the procedural requirements of U.S. regulations and the ethical principles that underlie them. There may be a variety of ways of implementing general principles, and these ways may take different forms in other countries or cultures. Broadly stated principles are designed to give pidance, but as principles they are, of necessity, general in their formulation. This is one difference between principles and rules of conduct. Principles require interpretation and elaboration before they can be applied in practice, while rules of conduct (for example, the Ten Commandments), specify what specific sorts of actions are ethically permitted, obligatory, or prohibited. Moreover, ethical principles embody ideals that may not be fully realized in practice. The distinction between normative ethics-what ought to be-and descriptive ethics-how people, in fact, behave-is essential for any critical reflection about research ethics. The well-known principle of respect for persons (National Commission 1978) focuses on autonomy, but that is not the only value embodied in the principle. Respect for persons includes respect for their spatial and informational privacy, as well as the notion that human dignity must not be ignored or compromised in the researcher-subjectencounter. Of course, the most fundamental application of the respect-for-persons principle to the research context is the requirement to obtain voluntary, informed consent from people before they may be enrolled as research subjects. As stated in the Belmont Report (p. 4), "Respect for persons incorporates [the ethical conviction] . . . that individuals should be treated as autonomous agents." Respect for persons can thus be understood as an ethical prin-

26

RUTH MACKLIN

ciple designed to protect the rights of human subjects of biomedical and behavioral research, in particular, the right to self-determination. The reason why informed consent is an ethical requirement even when the proposed research carries a very low or virtually nonexistent risk of harm, as is true of some social or behavioral investigations, is that people can be wronged even when they are not harmed. To carry out perfectly benign studies on human beings without their knowledge or consent thus wrongs them because their right to self-determination is violated. In the absence of ganting voluntary, informed consent, research subjects are being treated as a mere means to the ends of others, as objects or instruments rather than as persons worthy of respect. The three ethical principles elucidated in the Belmont Report are sometimes held to be peculiarly "Eurocentric." Differences among cultures-especially with regard to the primacy of the individual-have led some commentators to argue that informed consent is a concept understandable and applicable in the West but is irrelevant to social and cultural norms in Africa and Asia. Questions have been raised about the appropriateness of applying ethical standards and procedures from Western countries to non-Western societies. These standards and procedures are sometimes viewed as narrowly "American" in scope and content. For example, at a conference sponsored by the World Health Organization in December 1980, the WHO'S "Proposed International Ethical Guidelines for Human Experimentation" were first presented. Those guidelines were described by a conference participant from the United States as "essentially based on American standards of ethical review as well as on the international codes" (Miller 1981,~)-the Nuremberg Code, the Declaration of Helsinki, and the Tokyo Amendment. Critics from developing countries at the W H O conference objected to "ethical imperialism." "How far, they wondered, can Western countries impose a certain concept of human rights? In countries where the common law heritage of individuality, freedom of choice, and human rights do not exist, the . . . guidelines may seem entirely inappropriate" (p. 10). This debate was revisited in 1992 when a new set of International Guidelinesfor Biomedical Research Involving Human Subjects was under review by an international committee convened by the Council of International Organizations of Medical Sciences (CIOMS), which issued the 1982 guidelines and also the revised version. A report of the conference (Miller 1992, 8) notes that "No participants were willing to discard any of [the North American] values for research within their countries, and many acknowledged the need to examine their moral traditions in order to define the most appropriate value framework for ethics review." It is critical to distinguish between the specific procedures embodied in U.S. regulations and the fundamental principle that the research enterprise demands that we respect as individuals the persons who are enlisted as subjects in an activity that is not designed to serve their ends or purposes. Yet it is just this focus on

IS ETHICS UNIVERSAL? 27

the individual that is criticized as peculiarly American or, at least, Western. Lisa Newton, a philosopher from the United States, affirms that "it is 'ethical imperialism' at its worst to assume that the informed consent requirement, which does indeed serve one (only one) moral principle in the Western setting, is in itselfsuch a universal ethical standard." Newton contends that there is today a growing doubt surrounding the value of individualism and individual rights, so "the investigator might better stick to the research and accept the local assessment as to adequate protection of individual rights" (1990.11). Exactly the opposite position is stated by another American bioethicist, Ruth Faden, and coauthor Carel Ijsselmuiden, a physician from South Africa, who write: "Appeals to cultural sensitivity . . . are no substitute for careful moral analysis. We see no convincing arguments for a general policy of dispensing with, or substantially modifying, the researcher's obligation to obtain first-person consent in biomedical research conducted in Africa." Interestingly, Ijsselmuiden and Faden add that defenders of such a policy "have relied on limited and often dated anthropologic literature that does not reflect the rapid cultural changes brought about by colonialism and independence, warfare, and urbanization" (Ijsselmuiden and Faden 1992, 833). The danger - of Newton's reliance on local assessment is that in societies where there is no tradition whatsoever of individual rights, the local assessment may reject the very concept that individual research subjects have rights, and therefore they may be enrolled simply with the permission of the village chief. Even if it is the custom for a village chief to decide what everyone in that village is permitted or required to do in the ordinary activities of the village, it does not follow that the chief should be granted the same authority to submit his subjects to research maneuvers at the hands of biomedical or social scientists. The tenets of modern medical science were developed in the West, as well. Does it follow that they are only applicable in the West and that collaborators from developing countries should be free to modify research protocols by including interventions by native healers? The controversy over the universality of ethical standards governing research invokes a long-standing philosophical debate concerning whether ethics are relative to time and place. One side argues that there is no evident source of a universal morality and that ethical rightness and wrongness are products of the cultural and historical milieu from which they emanate. Opponents claim that even if a universal set of ethical norms has not yet been articulated or agreed upon, ethical relativism is a pernicious doctrine that must be rejected. The first group replies that the search for universal ethical precepts is a quest for the Holy Grail. The second group responds with the telling charge: If ethics were relative to time, place, and culture, then what the Nazis did was "right" for them, and there is no basis for moral criticism by anyone outside the Nazi society. Both sides appear to capture a kernel of truth. -

28

RUTH MACKLIN

The research enterprise today is global in scope. Can a coherent argument be made that because different countries or cultures have varying customs and traditions, different ethical standards based on those customs and traditions can be justified?To argue that different ethical standards are permissible because research in any culture is acceptable, so long as it follows the customs and norms of that culture, is either to sanction violations of human rights in the countries where those violations occur, or else to deny that there are any fundamental human rights. Either view embodies a form of ethical relativism that comes down in the end to the position that "whatever is, is right." Interestingly, just the opposite worry has been voiced by spokespersons from developing countries. In recent decades and even today, some research is carried out in developing countries that is not or cannot be conducted in North America and Western Europe. Critics of that practice, both from the West and from developing countries, argue that if a research project is deemed ethically questionable or unethical in a developed country it ought not to be conducted in a developing country that lacks a process for research review. But even if there exists a mechanism for research review in a developing country, if the norms or beliefs of that country do not embody respect for individual rights, it could be perfectly acceptable to proceed with a research design that would not be approvable in a developed country. It is a contradiction to seek to have it both ways, arguing on the one hand that research that is unethical in the United States should not be conducted in Africa or Asia, and at the same time contending that ethical values vary from one place to another so it is ethical imperialism to impose Western values on non-Western cultures. The enterprise of international collaborative research requires ongoing attention to these problems. In one example, a research proposal from an African country designed to study fertility awareness and pregnancy avoidance stated that "village chiefs will also assure the women of the need to cooperate." The program officer from the international agency to which the proposal was submitted wrote back to the principal investigator: "This implies pressure from the village chiefs, and also implies that the women who participate in the study will be known to them." The program officer's letter added: "You state 'the consent that we anticipate is almost taken for granted in our own part of the country.' Again, as you well understand, consent should be informed and entirely voluntary." In his reply, the principal investigator expressed his knowledge of and willingness to comply with these ethical requirements. The PI'S letter said: In Nigeria, there is no ethical problem in dealing with human subjects relating to information gathering. The intensity of research activities in various parts of Nigeria also reveals that Nigerian women are more than willing to answer questions relating to their sexuality. . . . Emerging new experiences

IS ETHICS UNIVERSAL?

29

corrected the earlier insinuations that Nigerian rural women might be unwilling to discuss what might be regarded as private affairs. In the present circumstance, the village chiefs are routinely informed of any project to be executed in their villages as part of courtesy to them as the head of the village and for him to be in the know of things going on around him. He, in any case, does not know the individual women to be interviewed. In our case, we do not anticipate any ethical problem on this: indeed the consent of the village chief is taken for ganted because it is freely and willingly given. The responses to questions from women are to be held confidential. Those unwilling to be interviewed are normally not coerced to respond. Therefore the issue of pressure does not arise at a1l.j It is evident that the investigator from Nigeria was well aware of the ethical requirements of the sponsoring agency and sought to assure the sponsor that the ethical precepts of voluntariness of participation and preserving confidentiality would be adhered to. What happened, of course, in the actual conduct of the research is impossible to know. A more nuanced issue is that of the requirement found in the U.S. regulations that informed consent be realized in the form of a signed, written document. Especially in the realm of social science research, but also in the broader sphere of biomedical research, this requirement is often questioned in the cross-cultural context. The requirement for written, signed consent forms is sometimes referred to as "an ethical standard," which it is not. It is a procedure designed to ensure that research subjects are adequately informed and that they freely grant permission to serve the ends or goals of someone else's project. The key distinction is between the process of informing and obtaining voluntary consent from prospective subjects and the piece of paper they are asked to sign. The latter is a consent document; it is not informed consent. It is amazing but true that even today, many sophisticated professionals from Western countries continue to fail to distinguish between the process of obtaining people's permission to invade their bodies or ask intrusive questions, on the one hand, and the consent form that is supposed to document that process. They also frequently misconstrue the purpose of an informed consent document. Many physicians believe that the purpose of informed consent is to protect the researcher or clinician from legal liability and they cynically observe that "you can be sued anyway, so informed consent is not worth the paper it's written on." But this is radically to mistake the purpose. The informed consent document is designed to attest to the fact that a research subject or patient has been told what the professional plans to do and grants permission for that maneuver, whether therapeutic or investigational. In some instances, a requirement ofwritten consent would constitute a barrier to doing the research. One team of researchers from a country in South America

30 RUTH MACKLIN

said that people in that country are reluctant to sign things. They may believe they are signing away their rights; or they may view signing a consent form as a "waiver" of some sort. Especially in countries where there have been authoritarian or military governments, people are wary of signing things. According to social scientists in more than one country, written consent is virtually never obtained at least for social and behavioral studies.* The procedural requirement of written, signed consent forms is just that-a procedural requirement. It is not a substitute for the consent process, and there may be ethically sound reasons for waiving this procedural requirement. One obvious reason is that the presence of a signed consent form places the subjects at risk. Social scientists conducting abortion research in the Philippines said that one reason written consent is not obtained is the need to preserve confidentiality.5 In places where abortion is illegal, the existence of a signed consent form in abortion research reveals the fact that the woman had an abortion. The same point was made in Chile by a group carrying out epidemiological research on AIDS.6 If the very existence of signed consent forms places subjects at risk of psychological, social, or legal harm, then it is not only ethically permissible, but ethically desirable to forgo signed consent forms. This is as true in the United States as it is in countries less accustomed to obtaining written consent for participation in research. Another obvious reason for waiving the requirement for written consent is that some studies involve illiterate subjects. The fact that people are illiterate may not be grounds for disqualifying them, and it is certainly not gounds for abandoning an oral explanation and gaining their permission to serve as subjects. It is, however, sufficient reason to abandon the need for a written document describing what they have been told and requiring that they make an " X on a signature line. A more substantive ethical concern about informed consent is the custom in some countries for husbands to make decisions about every activity their wives engage in, including giving permission for research in which their wives participate. A group with whom I met in Nigeria7 said that they could not carry out their research at all without first obtaining the husband's consent to involve the wife. This poses an ethical dilemma, as the only person who can properly consent to serve as a research subject is the subject herself. It therefore creates the dilemma ofwhether to do the research at all, if it violates a basic precept of research ethics, or on the other hand to forgo doing the research and thereby lose the opportunity to learn information that is critically important for improving the condition of women. What was not explored in the discussion was precisely what would happen if the researchers sought to do the research involving women without first obtaining the consent of their husbands. Would the women refuse to be subjects? If they did agree to participate, would their husbands discover that their wives were participants in the research? Was this simply a case of researchers adhering to a social practice because of the existence of the social norm? Might they seek to

IS ETHICS UNIVERSAL?

31

circumvent that norm and thereby succeed in doing- the research without violating the ethical requirements of informed consent? These questions were not addressed at the meeting I attended in Nigeria. They have been addressed, however, by the Scientific and Ethical Review Group (SERG) of the Special Programme of Research, Development, and Research Training in Human Reproduction (HRP) at the World Health Organization. This committee developed a document that was approved and disseminated by the W H O secretariat entitled "Guidelines on Reproductive Health Research and Partners' Agreement." These guidelines state that "a requirement of partner agreement or authorization for an individual to participate in research violates the autonomy of research subjects and their right to confidentiality. Therefore, as a matter of ethical principle, a requirement of partner agreement or authorization should not be permitted in studies supported by the Special Programme of Research, Development and Research Training in Human Reproduction" (p. 3). However, the guidelines also note that "because of existing cultural, religious, political or legal constraints, it is sometimes impossible to achieve the ethical ideal and exceptions to this general principle may have to be accepted (p. 2). The chief exception is stated as follows, under the heading "Social/Cultural Factors": "In rare circumstances, it may be necessary for researchers to conform to local custom and request partner agreement. An example would be the impossibility of recruiting any research subjects for a study in a particular country without partner agreement and the subsequent impossibility of gaining approval in that country for a new contraceptive drug or device. If failure to conduct the research would result in an inability of people in that country to receive the benefits of the drug or device, this consequence might be judged as sufficiently negative for the common good of the public to outweigh the usual prohibition against partner agreement for the individual subject" (p. 2). It is important to note several points about the rationale for this exception. The rationale certainly does not allow an individual to enroll a spouse in research without the spouse's own informed consent to participate. Moreover, the rationale does not endorse the practice of partner agreement or acknowledge the ethical acceptability of that practice. O n the contrary, these guidelines explicitly say that partner agreement "violates the autonomy of research subjects and their right to confidentiality." The relevant ethical principle does not permit the practice of partner agreement, and these guidelines uphold the principle. The guidelines do not assert that partner agreement should be permitted simply because in some cultures that is the custom or the norm. Rather, the pidelines justify the exception on grounds that a denial of the eventual benefits of the research to the entire society would be so great as to outweigh the usual prohibition against partner agreement for the individual subject. This is, without question, a utilitarian justification. Its use in this context demonstrates the difference between the univer-

32

RUTH MACKLIN

sality of ethical principles and the very different notion of "absolute" or "exceptionless" principles. The principle of respect for persons is universally applicable, which means it is not applicable solely in those countries or cultures whose customs and laws adhere to it. But is it an "absolute" principle, in the sense that it is exceptionless?Probably not. Special circumstances can justify overriding individual autonomy just as they justify abrogating certain individual rights in other situations. Yet this is a very different ethical justification from one that flows from ethical relativism. Informed consent is not the only "Western" requirement that poses problems for cross-cultural research. A question about the requirement to preserve confidentiality was raised at a meeting in Banglade~h,~ where cultural differences between Asian and North American countries are rather vivid. One participant stated that confidentiality is a Western concept, and therefore may not be applicable in the research setting in Bangladesh. She said that it is customary to share medical information with family members, and that insisting on confidentiality for the purpose of research would not be accepted in Bangladesh. Before I could reply, another participant, who had introduced herself as being from a women's activist group, objected to the comment about confidentiality. She appeared to be objecting on the basis of individual rights although she did not use the language of rights. It is significant that this participant was from a women's activist group, as these groups throughout the world are seeking to change longstanding customs and traditions that perpetuate the subordination and oppression of women. To seek to change a society's customs on the grounds that traditional practices subjugate women is to strive for moral progress.' A cynical view might hold that these women activists themselves have been influenced by the feminist movement in the West, and that therefore their view is not an authentic expression of their own culture. That cynical reply is flawed on two counts. First, it presupposes that cultures are or should remain static. Second, it confuses a causal explanation and an ethical justification. Even if it is true that the woman activist's comment can be explained by her acquaintance with and adherence to Western feminist concepts, that does not invalidate the ethical gounds for her endorsement of a requirement of confidentiality in research carried out in Bangladesh. Another example involving confidentiality, this one from the Philippines, illustrates the need to educate social science researchers about ethical requirements of research that may depart from the cultural norms of the society in which they live. I participated in a workshop devoted to ethics and social science research in which a researcher asked a question about her role in an abortion study.I0 She was making a follow-up visit after the patient was discharged from the hospital. This subject was a widow living in her home with the mother of her deceased husband.

IS ETHICS UNIVERSAL? 33

The woman had an abortion after becoming pregnant in an affair. The researcher asked whether she had an obligation to disclose the abortion to the mother-inlaw, since there is a cultural practice of everyone in the family knowing everyone else's business. However that cultural practice might be defended outside the research setting, the social scientist would not otherwise have had access to this information but for the research she was conducting. To disclose to the mother of the subject's deceased husband facts about the woman's sexual behavior and subsequent illegal abortion carried a potential for harm to the subject's interests. Nondisclosure on the part of the researcher is dictated not only by an obligation to preserve confidentiality, but also by the obligation to minimize harms to subjects. This is not a "peculiarly Western" notion or one that can be properly characterized as ethical imperialism dictated by U.S. regulations governing research. 1contend that when a society's norms and customs diverge from fundamental ethical principles, researchers are obligated to adhere to ethical requirements of human subjects research rather than to local or cultural customs. There are, however, undeniable instances of bureaucratic foolishness in international collaborative research. A researcher from the United States was doing a study in Bangladesh, for which he was required by his home institution's research ethics committee to use their standard consent form." O n the boilerplate portion of the form, there were instructions to subjects to call (collect) the research office of the university in the United States in case of any questions or problems about the research. The home institution's insistence on using its standard consent form is an example of a procedure that is misguided and utterly irrelevant to ensuring the ethics of research carried out in Bangladesh. Why insist on the relevance and applicability of ethical principles? Critics of the approach to bioethics that makes principles central to an ethical analysis tend to caricature the way principles are or ought to be employed. These critics misconstrue the use of principles as following some sort of deductive model, according to which right or wrong actions are "deduced" from very general ethical principles, such as respect for persons, beneficence, or a principle of distributive justice. But that picture seriously distorts the role of general ethical principles. Rather, principles play a justificatory role. They are used to support moral judgments of the form "this action is ethically wrong" or "this policy is ethically sound," or "this research methodology is ethically questionable." How else but by appealing to a general ~ r i n c i ~can l e a articular moral judgment be justified? Reasonable people may disagree over whether respect for autonomy should take precedence over promoting the welfare of an entire community, but that is not to reject the principle ofautonomy as irrelevant. If a communitarian approach to ethics places the value of maintaining tradition or promoting solidarity over the rights of individuals, the only way that approach can be justified is by setting priorities

34

RUTH MACKLIN

among competing ethical principles. Invoking one or another ethical principle may not succeed in resolving profound dilemmas, but it can make clear the source of ethical dilemmas and disagreements. Although generally speaking social and behavioral research poses less risk of harm to subjects than does biomedical research, there is much more to be said about this (Berman 1982; Cassell 1982; MacIntyre 1982; Nelkin 1982; Wanvick 1982). The potential for causing harm to subjects is something a conscientious social science researcher must consider in the design and conduct of the research. It is also a matter, as the following illustrations show, that could not realistically be addressed by specific research regulations. I begin with an example that illustrates the sorts of harms, as well as wrongs, that have occurred in certain instances of social science research. A social scientist working in a population institute at a university in a developing country in Asia described a proposed study on the incidence and social and behavioral consequences of abortion in her country, where abortion is a criminal offense.12The study was to be conducted in hospitals and in the community in three cities and one rural area, and consisted of interviews with women and service providers. The women to be interviewed were those admitted to hospitals following incomplete spontaneous abortion or complications of induced abortion. The service providers included midwives who perform abortions as well as medically trained personnel in hospitals, most of whom treat patients suffering the consequences of incomplete or induced abortions. The study design included the following features: the women coming into the hospital were to be initially interviewed by a record clerk. Then a second intetview was to be conducted by a physician just prior to discharge. A third interview would take place at a home follow-up visit. The researcher mentioned that women who come to the hospital for this purpose are frequently accompanied by their husbands, perhaps because of their poor physical condition that necessitates the hospital admission. This study design poses at least the following ethical problems: the capacity ofwomen to grant properly informed, voluntary consent to sensitive research at the time of hospital admission for incomplete or septic abortion; whether a record clerk, even if trained to conduct an interview, is an appropriate person to be questioning patients about highly sensitive and probably illegal matters; problems of confidentiality, both at the time of the hospital admission, when the woman might be accompanied by her husband or another family member unaware that the woman underwent an induced abortion, and also in the home follow-up visit, where others might be present when the interviewer arrives; and negative attitudes toward abortion on the part of physicians who would conduct the interviews with women at the time of discharge from the hospital. This is a type of study being carried out in a number of different countries in which abortion is a criminal offense and where unsafe abortion is a critical pub-

IS ETHICS UNIVERSAL? 35

lic health problem. Such studies are important and likely to yield significant information. I asked the researcher in a general way about ethical concerns posed by the research, and she initially replied that they did not have to worry about that since there was no ethics committee that would have to review the research. I responded that it was not the procedures for review that concerned me, but rather the substance of the project's methodology. Asking her permission to address some of these problems, I then described my concerns about the patients' decisional capacity, confidentiality, legal risks for providers as well as patients, and punitive attitudes toward abortion that could affect the way physicians treat women in the interviewing process. The researcher was thoroughly open in acknowledging that she had not thought of most of these issues and added that her collaborators on the research had also never addressed such concerns. This is not a type of research that is inherently unethical and thus should not be carried out at all. Instead, it is the type of research in which adequate safepards need to be established, both protecting the fundamental rights of research subjects and also seeking to prevent harm to their interests. The example illustrates the fact that social science research may well carry risks to subjects, and that the researcher's obligation to minimize such risks begins in the formulation of the study design. Another example from the same country relates to obtaining subjects' consent for a follow-up visit at home. When social science research on abortion began in this country, researchers did not seek informed consent from subjects interviewed in the hospital for a follow-up visit at home. They did the follow-up by gaining access to the hospital records and getting information about subjects' addresses from the records. However, when they actually tried to do the follow-up at home, subjects were indignant. They asked, "Who gave you permission to look at my hospital records?" and stated their belief that the records should have been kept confidential. In this case, the subjects not only were wronged, but felt as if they had been wronged. Feeling as if one has been wronged is no longer simply a wrong, but rather qualifies also as a harm to one's interests. Interestingly, however, social scientists in a different country reported that when they came to the homes of subjects who had also been identified through hospital records without their consent, the subjects thought nothing of the fact that researchers showed up at their door unannounced and unexpected. These subjects had no subjective experience of having been wronged, yet it is still legitimate to ask whether they were wronged nonetheless. Furthermore, in this latter study, the researchers had obtained information from the patients' hospital records that the patients were at risk for developing cancer, a fact that the patients themselves did not know. In contrast to the preceding situation, however, the subjects in this example did not seem to mind that people with no legitimate reason to look at their records had been given access to them. In addition, the

36 RUTH MACKLIN

subiects stood to benefit from the researchers' arrival at their homes since the researchers not only informed the subjects that they were at risk for developing cancer but the researchers also facilitated the subjects' return to the hospital for follow-up medical examinations for early detection and treatment of cancer.13 This poses the intriguing question of whether obtaining confidential medical information about patients and making a subsequent unannounced visit to the patients' homes can be ethically justified when the potential health benefits to the subjects themselves are substantial. A social scientist from a Latin American country recently told me that the "problem of ethics" is introduced from the outside into her country.'* When she was studying to be a social science researcher, ethics was taught in terms of preserving anonymity or confidentiality of subjects; the rule that subjects should not be paid (although now in social science research, subjects are being paid); and the dictate that social scientists should not make value judgments. What is new is informed consent, the requirement to introduce the notion of voluntariness into interviews, the idea that subjects can stop the interview at any time, and that subjects do not have to answer all questions. She reiterated the view that people in her country do not want to sign anything. This researcher said that the Social Science Task Force in Human Reproduction at the World Health Organization "requires the signed piece of paper" in research that it sponsors. However, she acknowledged, this is simply not adhered to in practice. It is not surprising that questions about the nature and scope of informed consent are raised by social scientists from developing countries. A debate has persisted for years in the United States over the need for consent and the procedures to be adopted, the ethics of withholding information from potential subjects or directly deceiving them, "shading the truth" in the informed consent process, and what is accomplished in "debriefing sessions" (Baumrind 1978; Bok 1995; Capron 1982; Dworkin 1982; Elms 1982; Macklin 1982; Marshall 1992; Murray 1980). There seems to be less attention to these questions in the literature today than there was fifteen years ago, but that need not mean that the issues have gone away. My own IRB grappled for months with a piece of deception research, which it disapproved until the researcher substantially altered the design of the protocol. The researcher, angry at the unwillingness of the IRB to approve his study, sought information from an IRB chair at another, quite prestigious institution. The chair of that IRB assured him that his own committee would approve the protocol and had approved similar ones in the past. The researcher sought an opinion from the Office for Protection from Research Risks (OPRR) of the National Institutes of Health, and was told that federal regulations did not explicitly prohibit the research design of his protocol. The researcher's anger at the IRB was heightened by his contention that the methodology would be seriously compromised if deception of subjects were not allowed and his belief that the benefits to be gained

IS ETHICS UNIVERSAL? 37

from the study were so great that they could justify deceiving the subjects. For its psi-t, the IRB used a high ethical standard both in rejecting deception and in assessing the risk-benefit ratio, a standard that went beyond what the regulations minimally require. It is not only the informed consent requirements, but also the use of riskbenefit assessments that is questioned by some social scientists. Some have argued that the language of risks and benefits is better suited to an ethical evaluation of biomedical research than to social and behavioral research. As the foregoing examples show, attention to and assessment of risks and benefits is just as important in social science research as in biomedical research, despite the fact that potential harms are rarely physical or life-threatening. People can suffer dignitary harms at the hands of researchers. Although dignitary harm may not be as palpable as kidney or liver damage suffered in the course of a drug trial, it is nonetheless a species of harm to human beings. The challenge to social scientists is to develop alternative research or consent strategies (Baumrind 1978; Capron 1982; Kelman 1968) that can enable their research to go forward without compromising fundamental ethical principles. Another questionable contention is that the risks and benefits in biomedical research can be measured and quantified, and therefore the task for researchers and IRB members to ensure that risks are reasonable in relation to anticipated benefits is a straightforward scientific enterprise. However, as anyone who has served on an IRB knows, the language of weighing risks and benefits in biomedical research is largely metaphorical. What kind of scientificcalculation can be used to determine whether the risks of toxicity to a small number of subjects in a phase 1drug trial are outweighed by the potential benefits to future patients, if the drug actually becomes approved for marketing? What sort of weighing is involved in the approval of a placebo-controlled clinical trial when the drug under study is only a slight modification from existing drugs to treat the same condition? The Belmont Report itself notes that "only on rare occasions will quantitative techniques be available for the scrutiny of research protocols." The report acknowledges that "the idea of systematic, nonarbitrary analysis of risks and benefits" is an ideal, and not realistically possible to achieve in practice. Still, the report maintains, this approach "should be emulated insofar as possible."

Research Ethics beyond Regulations For different reasons and in a variety of ways, regulations could not suffice by themselves to ensure that research is conducted in an ethically appropriate and desirable way. Concerns expressed by researchers eager to do the right thing go beyond what regulations do and can require. For example, in one Latin American

38 RUTH MACKLIN

country, researchers posed two concerns related to obtaining informed consent from subjects who do not understand the fundamentals of the research enterprise.l5 The first concern relates to the subjects' understanding: Can women in rural villages where social science research is being done understand the nature of the research process? Can women understand that they are being asked to participate in an activity that is not designed to benefit them directly? The second concern relates to the statement of benefits. The researcher noted that since women may derive psychological benefit from talking frankly to a researcher about topics they have never been able to discuss openly with anyone, is it acceptable to tell potential subjects that the research may benefit them by giving them that opportunity? The social scientists who voiced the concerns they experience as researchers were not questioning the need for informed consent, but rather endorsing it. Their concerns addressed the nuances of how best to accomplish that goal given the characteristics of the subjects and of the research itself. A different ethical concern that is not and could not be dealt with in regulations was brought up by a social science researcher in India.16She described a dilemma she faces as a researcher: how to make research more participatory, and how to level out the power relationships between researchers and subjects. She noted that it is more difficult for an academic researcher than for a service provider to establish legitimate relations with subjects. Research projects are grant oriented, and so the researcher does not have ongoing relationships with clients in the way service providers do. Many women who are subjects of her research are illiterate, and therefore could not read transcripts of the research if they were made available at the conclusion of the research. Her own research is on women's sexuality, and she had spoken with six groups of slum women in Bombay. A videotape was made of the focus group sessions she conducted. It was explained in advance to the women that the videotape would be shown worldwide. Although a rigorous process of informing was carried out in advance, the researcher was still left with a doubt in her mind: Did these women understand what research is? Did they understand the fact that the videotape would be shown around the world? The researcher has to explain the nature of research itself, as well as to tell subjects the content of the studies. Part of the ethical dilemma stems from the fact that research does not deliver services, so researchers cannot promise the women that there will be any improvement in their condition. The concerns mentioned by this researcher reveal that ethical sensitivity can and often does precede and go beyond formal regulatory requirements. Most developing countries have no regulations that deal with social science research, nor is there an outside review process. Nevertheless, ethically concerned and committed researchers do not need such regulations to instill ethical behavior in the conduct of their research. The reverse is also true, of course. Researchers who are

IS ETHICS UNIVERSAL? 39

unconcerned about ethics will not be magically transformed by the existence of formal regulations that focus more on procedures than on ethical substance. In the following cases, social scientists themselves wondered whether their research design and activities could result in producing more harm than benefits to the subjects. Working in a rural area of Mexico, a social scientist was involved in a project that contained both research and educational interventions. The researcher asked: "What risks are we taking when we introduce ideas that are utopian?" l 7 Her example was a workshop on the subject of women's sexual pleasure. Feminist ideas and attitudes were presented to the assembled group of rural, uneducated women. But these women cannot be isolated from their everyday life: their husbands, their mothers-in-law, their community. The researchers believe it is frustrating for women attending the workshop to be exposed to feminist ideals that they cannot act on within the traditional framework of their daily lives. Some research and educational interventions have unpredictable risks for the women at whom they are aimed, while others have predictable risks. The social science researcher asked: Is it ethical to subject women to risks that are indeed predictable, such as exposing them to feminist ideas that can undermine their self-esteem? A similar problem was mentioned by a group of social scientists in a different Latin American country. One researcher noted that women reveal to social scientists that they have undergone abortions, and they expect services that researchers are unable to provide, such as psychological support. Women expect at least a referral for such services, if not direct services provided by the researcher. This might be remedied, at least in part, by improving the quality of the informed consent process. Beyond that, researchers need to be cautious in planning and embarking on their research. Even in the social sciences, the ethical conduct of research requires a favorable benefit-risk ratio. Thus, if harmful consequences are likely to ensue from a particular piece of research without any compensating benefits, researchers must assess in advance whether they may ethically carry out the research. If researchers are able to provide for referral for women in need of psychological support, that would improve the benefit side of the equation. A variation on this theme relates to the proper role of researchers in areas of sensitive investigations. It often happens that subjects ask questions that the researchers are ill-equipped to respond to. For example, subjects (especially adolescents) may seek help for their sexual problems or ask technical questions of the researcher. For their part, researchers may lack knowledge of the technical information and lack the training or skills to provide the required emotional support. Here again, the question is whether it is the proper role of researchers to provide such support. A specific example of this general type of problem was posed at a workshop with social scientists in Argentina: What to do when very sensitive topics are raised by subjects where those topics are not part of the agenda of the interview?l 8 The

40 RUTH MACKLIN

case that confronted the researcher was one in which the research subject, an adolescent, wanted to discuss incest. For her part, the researcher was unprepared and felt ill-equipped to discuss this topic, and in any case it was not part of the research agenda. What was her obligation to the adolescent subject? A related, yet somewhat different question relates to the obligation of social scientists to report or take action on something discovered in the course of their research. This arose at a meeting I attended with social science researchers in Nigeria, and was also reported by social science researchers in other countries and settings. The example in the Nigerian case was that of two girls who had been raped near a scho01.'~The researchers discovered this in the course of a routine study, and when they brought the incident to the school principal, he told them not to say anything about the incident; they in turn wondered what action to take. Apparently, there was no mechanism available for legal action. But the researchers pondered whether there was something else they might do-for example, expose the incident to the parent-teacher association, encourage people to write letters exposing the incident, or try to identify other ways of preventing future occurrences. Although it is impossible to provide a general solution to this type of problem, social science researchers should address such issues as a general ethical concern about their obligations beyond the conduct of research itself toward research subjects and others. The problems described by the director of research in one institution in Latin America20are illustrative of an insistence on the part of international sponsors that all collaborating institutions adhere to the same detailed procedures relating to informed consent. A chiefcomplaint ofthe researchers is that the consent forms they receive from outside sponsors are too long, too complicated, and "too scary." Research personnel at the institution invest considerable effort to adapt these forms to a language the subjects can understand and to a length they are likely to read. The director of research expressed the thought that these lengthy and technical consent forms are insisted upon by lawyers seeking to protect the sponsoring institution or agency. Another complaint about the multi-institutional consent forms imposed from the outside is that the facts presented about risks are global data rather than those pertaining specifically to individual clinics. An example is the figure cited for risk of perforation from an intrauterine device (IUD). At this institution, there have been only two perforations in the entire history of the clinic, but worldwide data show a considerably higher incidence. This is an example of a procedural point: it is entirely appropriate to state the risks and benefits of research maneuvers in the statistical terms that subjects in this clinic actually face. A consent form could say that the incidence of harm is higher in other places, but the form can certainly include the statistical probabilities that pertain to the site where the research is carried out.

IS ETHICS UNIVERSAL? 41

A different problem with outside sponsors occurs at this same research institution. International collaborating or sponsoring agencies now insist that the consent forms for the research be signed before initial interviews with potential subjects, and thus before the rather elaborate process this institution uses to introduce women to the research. This timing is unacceptable to the researchers, since their procedures involve a lengthy process of explanation, over several visits, before a consent form is actually signed. The reason given for why the outside agencies insist that a consent form be signed at the outset is that the screening process leading up to enrollment includes procedures for which (potential) subjects should grant consent. Although these are not actually research maneuvers, they are procedures that women should consent to. There is a ready solution to this procedural problem. It is simply to use consent procedures at two different times: one for the initial screening examinations and one for the research itself. This is desirable not only because of the timing problem, but also because not all women who undergo the screening will actually be involved in the research itself. Moreover, screening procedures are not research maneuvers, so there may be confusion in the minds ofwomen about just what is research and what are standard biomedical examinations. To insist, as does the sponsoring agency, that all consent forms be signed at the outset is not only an unnecessarily rigid procedure but also one that is ethically flawed. The fundamental ethical principle of respect for persons is not well served by procedural requirements that thwart rather than ensure that subjects grant voluntary, informed consent to participate in research. Still another problem that researchers may confront has little to do with regulatory requirements but stems from the relationship between the sponsoring agency and the researchers. Faculty at one research institute in India conduct research sponsored by the government, which has led to some disputes.21Faculty researchers maintain that the public has a right to know what information the researchers have collected, whereas the government has sought to maintain secrecy. The director of the research institute said in one case that the institute was successful in seeing to it that the government secrecy clause was not implemented. Following a research project studying people who had been displaced when there are large government projects, the researchers claimed that people had a right to know the results of such a study. The government wanted the researchers to be apologists and spokespersons for the government, and was eager for the report to be favorable to the government's interests. But the director says, "We are researchers and have to be honest and objective in our work." After a six-month interval, the report was declassified. An especially egregious instance in which researchers were held hostage by the sponsoring agency involved the U.S. Public Health Service conducting epidemiological research on the risks of radon exposure to uranium miners. Substantial

42

RUTH MACKLIN

evidence was discovered during the course of this study that the radon hazard to miners was serious. In addition, the hazard could have been reduced by proper ventilation of the mines. However, in order to gain entry to the mines the researchers agreed that the Public Health Service would not "alarm the miners" by informing them of the dangerous conditions (Advisory Committee on Human Radiation Experiments 1995, 571). Duncan Holaday, a Public Health Service engineer who spent many years studying the miners, testified much later that "you had to get the survey done and you knew perfectly well you were not doing the correct thing . . . by not informing the workers" (Advisory Committee on Human Radiation Experiments 1995, 571). The Advisory Committee on Human Radiation Experiments noted in its Find Report that "current guidelines for the ethics of epidemiological research. . . would not counsel the original bargain with the mine owners, the minimal disclosure made to workers about the purpose of the research, or the failure to warn the workers as the hazard became clear" (p. 571).

Conclusion As I have sought to argue, we need to distinguish between the detailed requirements of regulations, most of which are procedural in nature, and the fundamental ethical principles that underlie the various declarations, guidelines, or regulations such as the Code of Federal Regulations in the United States, the Declaration of Helsinki, the Nuremberg Code, and others. Slavish adherence to procedures required in regulations leads to a foolish application of rules. At the same time, avoidance of an appropriate process for obtaining informed consent, on the gounds that people in a given society do not normally grant consent for anything as individuals, is to fail to respect the people in that society as persons, and therefore to violate a fundamental ethical principle. It fails to accord people in that society the dignity all people deserve by virtue of their being members of the human community. The fact that a particular culture does not recognize certain rights of its members, or human rights at all, does not thereby justify the failure of researchers to respect those rights. It should be clear that the poor fit between regulations and ethics in the sphere of human research is not entirely to be lamented. It is both necessary and inevitable. O n the one hand, regulations encompass numerous procedural items that have little to do with ethics in the conduct of research, such as how many members IRBs must have, the need for institutions to provide written assurances to a federal department or agency head that the institution is complying with the regulations, how records are to be kept and retained, and so forth. O n the other hand, the substantive ethical content of regulations is stated in such general form that it requires interpretation and analysis in order to be applied. More importantly,

IS ETHICS UNIVERSAL? 43

regulations do not address the actual conduct of research, but rather deal with the review of written protocols for research by IRBs and written documentation of research procedures, including consent forms. IRB review of the research proposal and the consent forms researchers prepare could not by itself possibly fulfill the aim of protecting the rights of human subjects. Jay Katz (1993b: 9 ) has argued that "respect for individual autonomy and for self-determination, which informed consent is intended to safeguard, will remain hollow aspirations until the nature and quality of the conversations between physician-investigators and patient-subjects about participation in research are radically transformed. Inviting such participation for the sake of science, society, and future patients is an awesome request that, in a democratic society committed to respect for human rights, requires the most punctilious attention to disclosure and consent." I would add that it is not only in democratic societies already committed to respect for human rights that we need to give attention to the process of disclosure and consent in research. If these rights are truly human rights, they transcend national boundaries, long-held traditions, and local customs. The right not to be used involuntarily as a means toward the ends of others, however noble may be those ends, should be recognized as a human right.

Notes I. Federal Policy for the Protection of Human Subjects, Notices and Rules, 56 Fed. Reg. 28002-28032 (June 18, 1991), p. 28015. 2. For a fuller argument and concurring views, see Macklin (1992), Angel1 (1988), and Ijsselmuiden and Faden (1992). 3. Correspondence provided to the author by the program officer. Names and institutions are omitted here to protect confidentiality. 4. Many of the examples in this paper are drawn from my own work in two international projects I have undertaken, supported by the Ford Foundation. The names of the researchers, their institutions, and, in some cases, the countries where the research took place are omitted to protect the confidentiality of the individuals and institutions. 5. Visit to the Philippines, December 1993, Ford Foundation Project. 6. Visit to Chile, October 1995, Ford Foundation Project. 7. Visit to Nigeria, October 1994, Ford Foundation Project. 8. Visit to Bangladesh, March 1994, Ford Foundation Project. 9. See generally Macklin (1977). 10. Visit to Philippines, Ford Foundation Project, December 1993. 11. Visit to Bangladesh, Ford Foundation Project, March 1994. 12. Ford Foundation Project visit, 1992. 13. Visit to South America, Ford Foundation Project, 1395. 14. Visit to South America, September-October 1995, Ford Foundation Project. 15. Visit to South America, September-October 1993, Ford Foundation Project.

44 RUTH MACKLIN

16. Visit to India, March 1994, Ford Foundation Project. 17. Visit to Mexico, Ford Foundation Project, February 1993. 18. Visit to South America, September-October 1995 19. Visit to Nigeria, Ford Foundation Project, October 1993. 20. 21.

Visit to South America, Ford Foundation Project, September-October 1995. Visit to India, Ford Foundation Project, March 1994.

48

CASE 1

the often very personal information that they elicit and observe. Anthropologists also often write well, resembling novelists or nonfiction authors more than their drier academic colleagues. However, writers of fiction and nonfiction do not usually concern themselves with their responsibilities to those they write about. They are "entitled" to present their version of events. What is different, then, about research?Why does Benjamin argue that researchers have such very strong responsibilities to their subjects? The community in Curaqao that he studied has a distinct history and context. Its members claim a particular vulnerability because they are Jews who trace their roots to the Holocaust and the Spanish Inquisition. They are also a small island community, concerned with their reputation (not unlike researchers). How can they not try to protect themselves, and how can they best protect themselves?In this case, they chose to establish a covenant, backed up by a contract. Should researchers accede to such a strategy, one that the community they propose to study deems most effective and that comes out of its traditions? The theme of trust, so powerful in all of the cases presented in this book, is described by Benjamin as generally one-sided: let the subjects trust the researchers. He, in contrast, chooses to trust the community he studies. The Cura~aoJewish community is elite, modern, and litcrate, with prior experience as research subjects. Thus it recognized its power and chose to negotiate terms. At one point, Benjamin suggested to Estroff, his advisor, that perhaps he should choose an "easier" community to study. But do we not need to study literate and powerful elites? Is it ethical for researchers to collect and reveal (to the powerful) information only about the powerless? O n the other hand, should we make deals with elites that we do not or will not make with less powerful groups? Benjamin incorporates his personal value system into his research-where social justice ranks at least as high as science itself, and where the interests of those studied are at least as important as the interests of those who study them. Estroff points out, however, that researchers have ambiguous and conflicting obligations, and both she and Fox present evidence of how Benjamin's agreement affects them professionally. These discussions make ever more clear why regulations are not in themselves sufficient to guide researchers through the ethical dilemmas and choices presented by their research. How one views concepts such as truth, trust, fairness, science, and justice is critical. Benjamin has chosen his ethical path based on his consciously formulated value system, as each researcher must. Through his difficulties, with the aid of his reflections and those of the commentators, we have an opportunity to explore our own values.

50 ALAN F. BENJAMIN

son, representing the board, continued until my third month in the field. Along the way the congregation discharged the original lawyer.2After the three of us met with a mediator, we were able to sign the seventh draft (Appendix E). The central clause of the contract is article three D: "The Review Committee shall not withhold its approval in matters regarding Alan Benjamin's interpretations andlor conclusions drawn from factual information, provided those facts are verified as correct . . . , and provided that these interpretations and/or conclusions will not, according to the Review Committee, adversely affect the reputation of any individual, organization or company which is directly or indirectly connected with The Congregation." The contract required that I receive prior written approval from the congregation of any material about them that I publish. Failure to receive approval would subject me to a fine, which was set at $2,800 per day from the date of publication until all copies were withdrawn from circ~lation.~ The Netherlands Antillean legal system could be called upon to prevent the dissemination of, or to withdraw from circulation, ethnographic representations I might produce about Congregation M i k d Israel-Eman~el.~ My fieldwork experience presents a challenge to anthropological-and, by implication, to scholarly-authority. In effect, board members said that having a "voice" in ethnographic productions about them was not enough; relying on the sensitivity of ethnographers was not enough; and the reciprocal relationshipwhat I call the "covenantn-normally established during the course of participantobservation was not enough for them. They insisted on the right to determine for themselves the acceptability of all material that I publish about them. My fieldwork experience indicates that the practice of research involves more than the neutral acquisition of knowledge. Scholarly research implicates power relations; it is not an innocent practice. Furthermore, if researchers fail to recognize and to address the concerns of those studied, then research does harm-as well as whatever good is achieved. Research is a practice that has value to society, but one should be skeptical of any claim that the value of potential research results supersedes the interests of those studied. Therefore, the central contention of this paper is for broader recognition among scholars that research occurs within contexts of social relations, and the central call of this paper is not to privilege the practice of research over the concerns of those studied. Below I briefly describe the research setting, focusing on the board and on the extended family system, two institutions central in the genesis of this contract. Then I discuss the issues-the board's and my own-reflected by and raised during our negotiations over this contract. I conclude with a discussion about the significance of this contract for my research, for anthropological research, and, primarily, for research practice in general. In doing so, I begin a critique of the term "scholarly integrity."

CONTRACT AND COVENANT IN CURACAO

51

The current population of the former Dutch colony of Cura~aois about 144,000 (Central Bureau of Statistics 1993). Of the two Jewish congregations currently on Cura~ao,United Netherlands Portuguese Congregation M i k d Israel-Emanuel, discussed here, dates to 1651. The 367 people (in 1991) affiliated with this predominantly Sephardi5 Jewish congregation constitute a small but significant and relatively elite6segment of the population, particularly active in local banking and in retail and wholesale businesses. Other than Dutch citizenship and the state control that it and Cura~aoanresidency entail-as is routine in any nation-state-congregation members are affected by two primary organizing institutions, the congregational administration (Direktiva, or board of directors), and the extended family unit (fdmiya). Rabbinical leadership can affect these latter institutions significantly, but is not itself a separate social institution. Social structural attributes of the Direktiva and the fdmiya, along with the congregation's past experiences with researchers, were significant factors in the board's request for a contract. The board of directors currently is a seven-member, all-male body with absolute control over congregational administrative functions (Benjamin 1996). The board tends to be a stable body that makes decisions in close concord with congregational opinion. There is an attempt to draw board members from a variety offdmiya to keep the board diverse and representative. Near relatives are not allowed to sit on the board at the same time. Minutes of board meetings are kept; however, the discussions are strictly secret. Formal involvement of the membership as a whole in decisionmaking is rare, but board members attempt to reach a consensus with each other that reflectsfdmiyd opinions. Sephardi Jewish life tends to revolve around thefdmiya (Karner 1969), though not all members of each fdmiya are Jewish. Holidays are celebrated in kinship groupings, as is most leisure time. Famiya members may share responsibilities in family businesses. Courtship includes spending considerable time with one's lover's extended kin. Everyday life is conducted within a thick web of interlocking, sometimes shifting, social relations. This social setting was a direct factor in the request for and the negotiations over our contract. Members of the board negotiated on behalf of their loved ones, on behalf of a host of people-living, dead, and yet unborn-whom they felt responsible to represent. For example, an older woman I interviewed during the period of negotiations asked whether people from the congregation would read my work before it was published. I replied that the board and I were working on a contract to arrange that. She said, "You see, they know what we want." Board members felt they carried a trust for those with whom their social relations were "multiplex" (Gluckman 1962,26 -27).

52

ALAN F. BENJAMIN

In recent decades there has been considerable interest in Congregation Mikvt Israel-Emanuel by outsiders, fueled by the fact that the Mikvt Israel-Emanuel synagogue (the Snoa) is the oldest building in continuous use in the Western hemisphere. There are visiting hours for tourists (some of whom attend services), a museum in the synagogue complex, and a steady stream of foreign journalists who visit to prepare articles about the congregation for tourist magazines and the Jewish press. In 1982, scholars came to lecture on the 250th anniversary of the Snoa. The congregation and its synagogue building are understood to have historic significance to Cura~aoanswho are not members, to Jews worldwide, and to anyone interested in European settlement of the Western hemisphere (Benjamin 1990). The congregation and synagogue building thus constitute a legacy over which current members have a kind of guardianship-what one member called "curatorship." Congregation Mikvt Israel-Emanuel and its synagogue building are a legacy "owned" personally by the living members of the congregation. The board takes a leading role in gatekeeping access to this and other legacies, acting on behalf of the members. I learned that twenty-five years earlier a previous board had undergone a heated argument with Rabbi Dr. Isaac Emmanuel over the contents of a history Emmanuel was writing under board sponsorship. In his 1970 publication (Emmanuel and Emmanuel 1970), Emmanuel included a short discussion titled "censorship" (pp. 8 -9), in which he wrote that he and several members of the board "literally fought over each word and comma" in the text (p. 9). Relations between Emmanuel and the board left bitter feelings on both sides. After his death and the sale of all the copies of his work, Emmanuel's daughter reportedly refused to allow his work to be reprinted. She is said to have commented that the congregation drove him to his death. At the same time, however, some members of the congregation were convinced that Emmanuel, who had access to the congregation's archives, took with him the earliest known document written in Papiamentu, the Creole language of Cura~ao.The document (Gomes Casseres 1990, 26; Henriquez 1988), a love letter written in 1775 by a congregation member, has never been recovered, and is felt to be a significant congregational loss. Current members of the board are relatives or friends of the members of the board who held discussions with Emmanuel and are informed about the discussions that took place at that time. Their understanding is that had it not been for the board's intervention, Emmanuel would have published material which may have had long-term "negative effects on individuals in that community"; they feel duty-bound to protect the good name of the congregation. Review and control over the Emmanuels' 1970 publication was predated by related instances. The American historian Jacob R. Marcus was required to obtain prior approval from the Mikvt Israel7 board before publishing any of the photostats he took of the congregation's archives in 1952 (Marcus 1953). Emmanuel him-

CONTRACT AND COVENANT IN CURACAO 53

self submitted an earlier work to review by a committee of three members ofMikvt Israel before its publication (Emmanuel 1957).

Issues and Motivations In considering the contract, the board and I drew on a variety of experiences and perspectives. We first had met two years previously, in the summer of 1989, when I spent almost six weeks in C u r a ~ a oconducting preliminary research. That encounter seemed to go well, and there was no request for a contract at that time. Based upon that research, I wrote a student paper about the Snoa (Benjamin 1990). Shortly before my return, a Sephardi acquaintance told me that the board wondered what I had done with the material I collected in 1989 and from a subsequent mailed questionnaire. The board explained that my 1991-92 research would be more "serious" than the earlier project; it would be more extensive and the resulting publications more likely to be read. Thus, the board sought to prevent some problematic consequences of my research that it anticipated. I accepted the notion of a contract between us, but wondered how it would work in practice. We each had an understanding of how we expected the other to act, relative to my research and writing, and of what the contract would do to shape or to limit our own and the other's actions. Not all of our understandings were accurate. In this section, starting with the board and finishing with me, I discuss some reasons we entered into discussions about a contract, and our concerns over the contract's clauses. The effect of my writing on the reputations of congregation members, the congregation as a whole, other Cura~aoanJews, and, to a lesser extent, Jews in general was a key concern to members of the board. The number of people with whom one is in regular contact on Cura~ao,and who can have a significant impact in one's life, is small (see Harrell-Bond 1976, who faced a similar concern when writing about social elites in Sierra Leone). Though Cura~aoanJews are in contact with people living in many other places, their social life might be characterized as intimate. "It's a small island," people say. Reputations-of individuals, kin groups, and ethnic groups-are crucial in every facet of one's life. They affect, for example, business affairs, prestige, potential marriage partners, and self-esteem. If damaged, they are difficult to change. Negotiating the contract helped me to understand this and to become engaged in community practices; the contract ensures that I respect these vulnerabilities in my writing. The ideological roots of this perspective were present from the formation of Congregation M i h C Israel. The first community bylaws (haskamot),in 1688, prohibited members from composing "in jest or otherwise . . . a verse, sonnet or pasquin injuring his fellowman" (Emmanuel and Emmanuel 1970, 545). Moreover,

54 ALAN F. BENIAMIN

when discussing the discipline of members who lead "an indecent life," one section of those haskumot explained a rationale for congregational authority: "If we do not do so, it will be said that we approve of improper conduct and on that account we may suffer prejudice, etc." (Emmanuel and Emmanuel 1970, 544-45). From the outset, Congregation Mikvk Israel, mindful of its members' and ancestors' oppression in Spain and Portugal, was concerned to squelch all possible justifications for persecuting or harming its members. The recent Holocaust of Jews in Europe-in which one member of Congregation M i k d Israel was killedserved to recall and to reinforce this perspective. I initiallywas accepted by the board because I amJewish and represented scholarship. As a Jew I was expected to be sympathetic to concerns about anti-Semitism. As a scholar my purpose was deemed to be fair-minded-that is, neither sensationalist nor bigoted-and to be directed toward increasing knowledge. In general, the board valued endeavors to increase knowledge. During the negotiating process, I was told repeatedly that Congregation Mikvk Israel-Emanuel had no intention of censoring my research or interfering with academic freedom. However, although accepted as a scholar, I had never published, and my knowledge of publishing practices was limited. Members of the board could not refer to published works of mine in order to evaluate me,8 and I was not confident that I could anticipate my needs in the realm of scholarly publishing. The board's request for a contract was not related to the nature of my research proposal or to me personally, but arose from che congregation's past experience with historians, as described above, coupled with their feeling of responsibility to protect the legacy and reputations of those most important to them: the congregation and people associated in a variety of ways with congregation members. The community is an elite, literate group, aware and protective of their position, and familiar with legal matters. Virtually all members of the community speak English. Some members read some English-language scholarly material, and understand that publications about them in English will be read by people in contact with them, affecting their community. The willingness of the members of the board to discuss a contract with me-rather than to oppose my researchindicates that they feel a link with Jews and a respect for text, including a desire to see their community remembered in text.' Knowing that I would be observing many congregational members in large gatherings, for example, during religious worship and at parties, I wanted to ensure that people understood the purpose of my presence. Since it would be impossible to ask every person present on such occasions to sign an informed consent form (see Mann 1976,who faced the same problem while conducting research in a bar), I askcd the board to publish a lcttcr from me in the congrcgational newsletter. The board thought this would constitute an endorsement of my research by them to members of the c~ngregation.'~ Their reluctance to provide such an

CONTRACT AND COVENANT IN CURACAO

55

endorsement without some control over what I would write about the congregation was an additional reason they asked for a contract. After the contract was signed, my letter was published in the newsletter with a statement from the board that it had endorsed my research (see Appendix F). As elites, the congregation had much to risk from my research, and within the context of Curacaoan social relations probably assessed the potential for harm as a result of my research more knowledgeably than I could. Though members of the board seemed to imbue me with more power than I felt, I could not ignore the forceful request of this elite group for a contract. The board did not have the power to prevent me from conducting research among members of the congregation who wished to cooperate with me; however, their opposition would have limited my effectiveness. As a mere graduate student, with limited funds, I doubted the power apparently ascribed to me. After all, members of the congregation had access to governmental authorities and to financial resources unavailable to me in Curacao. However, I came from a powerful nation, was a representative of a respected university, and-despite my junior academic status-was deemed to be due respect as a scholar. As such, my writings might be read and deemed to be accurate by people who would come into contact with members of the congregation, thus affecting the lives of congregation members. I immediately agreed to the idea of a contract because I believed that people being studied should have some control over representations of their lives. Complete absence of control on their part would constitute exploitation of them on my part-it would be using their lives to advance my research and career without considering their wishes. Justice would seem to require that they have the opportunity to attempt to protect themselves from harm that I might not foresee resulting from my publications about them. Too, I hoped that a review process would help correct inaccuracies in my writing- though board members demurred from accepting this responsibility My early reaction was that negotiating our contract seemed a step toward "gaining rapport." I understood negotiating the contract-and my signing the contract, whenever that would come to pass-as the board's way of testing me before endorsing my integration into their community. I expected that the contract would be a nascent "covenant," helping to establish rapport and shaping our future relationship and the processes of ethnographic research and representation beyond the contract's legal terms. In our developing "covenant," I committed myself to sensitive and respectful conduct; the board committed itself to cooperative conduct. In a process at times adversarial, we learned something of each other's expectations and anxieties over my impending research. Early on I gave my research proposal to the board. Few read it. We came from different cultural worlds. I was there to learn about their culture; they were not engaged in research into my culture-academic anthropology in the United States.

56 ALAN F. BENJAMIN

Most people in Curaqao who discussed my work seemed to expect me to engage in a straightforward process in which I would collect information that I would then report and interpret. The practice ofwriting about selected events, not all of which I could predict, in order to illustrate theoretical points-which may hold no interest for members of the community-was difficult to convey. I thought that members of the board would not and should not be expected to understand the research and writing processes with which I expected to be engaged. They were not familiar with contemporary anthropological scholarship and its processes. Brettell (1993,104) writes, "While we may be intrigued by a turn of phrase . . . or attracted to a mode of analysis that ignores the details in order to evaluate the structural relationship . . . our readers [that is, those studied] may not be." The primary reference points to which the board could compare my project were histories and novels, not anthropological monographs. I also was concerned that future congregational political conflicts, as well as anthropological literary devices and theoretical assumptions, might impede the acceptance of my work by congregational reviewers." To address these issues, the board agreed to an arbitration procedure as a way to resolve conflicts short of !going to court, and to a two-week deadline for each forty pages reviewed to ensure timely responses to my material. A further concern of mine was that members of the congregation would respond in a wide variety of ways to my material and that it would be impossible to please everyone on every issue. Thus, I sought to limit the number of reviewers. We agreed on a committee of five to increase the likelihood that a quorum of three always would be present. Now geographic distance has been added to cultural distance, another issue I anticipated during the contract negotiations. Spontaneous and informal discussion between us is difficult, making it harder to work together on reviews of my material. Phone calls are expensive, and mail delivery is uncertain. Members of the board and I are less well known to each other, as our lives continue in a less connected fashion. The constellation of concerns presented in this section serves to illustrate the many connections, the ambiguity, the conflicting interests, and the layered character of research relations. Although idiosyncratic, these issues and motivations hint at the complex social interactions present in research projects generally.

Discussion The contract I signed constructs a relationship in which power over published material is more easily contested than in other research-subject relationships, because the relationship is more equal. The authority of the ethnographer derived

CONTRACT AND COVENANT IN CURACAO

57

from "being there" (Geertz 1988) is no longer sufficient to publish material that those studied do not wish to see in print. The ethnographer's voice is not privileged over the voices of those being studied. The members of Congregation M i k d Israel-Emanuel and I share an ongoing moral relationship that includes mutual responsibilities-in other words, a covenant. During our negotiations this covenant was tentative and fragile. After the contract was signed our covenant grew stronger. Our relationship continued to strengthen as we observed each other, joked with each other, and discussed other topics during the succeeding year, while I was still conducting fieldwork in Cura~ao.Distance and time have both weakened and solidified our covenant. There is a loss of intimacy between us, coupled with an inkling of experience in what it is like to implement our contract. As board members read my material I expected that we would misunderstand each other and differ in opinion. I hoped that, by having worked together in good faith in the process of resolving our disagreements, we would continue to build trust. Yet, as we entered into our encounter, the chance that we might establish a morally reciprocal relationship, a covenant, was determined by the members of the board to be insufficient to their needs. They forcefully requested a more formal means of addressing their concerns, recognizing that my published representations would exist outside the sphere of our everyday social relations, among people not party to our developing covenant. Although the congregation is not a "typical" research community, their actions may express the attitudes of many other people. As Skomal (1993, 26) writes in response to an open letter by a "subject" of anthropological research, "people everywhere are increasingly anxious to control the information that is being released about themselves." During field research the contract contributed to-though was not always effective in-easing the concerns of those with whom I wished to speak. Now that my research in the field is complete, the review process entailed in the contract adds time, energy, expense, and worry to the task of writing; it is an inconvenience for me-as I assume it is for the reviewers. Since this paper was the first material to be reviewed seriously, it is somewhat premature to discuss the contract's impact on my publications. Nonetheless, the implications of this contract in the arena of social science research are far-reaching. Chief among those is to remind us that those we study have interests of their own, that those interests will not always coincide with conventional research practices, and that failing to address such conflicts may well constitute subordination of those we study. Below, in order to contextualize the change in perspective that I propose, I briefly review the role of ethics in the discipline of anthropology. Although the argument of this paper is directed toward scholarly research in general, the experience that informs it arises out of the practice of anthropological research. Representation itself, and control over the ~ublicationof ethnographic representations, are anthropological issues particularly relevant to this discussion.

58 ALAN F. BENJAMIN

Many ethnographers voluntarily have sent their material to the people studied for comments, and at the onset of research projects there have been attempts by those studied to exert various types and degrees of control over what would be published about them (for example, Appell1978, 208; Barnes 1979, esp pp. 14144). However, Claire Farrer's (1976) explicit agreement, sealed by a ritual-in which the Mescalero Apaches she studied were to review material about them prior to publication, and in which their consent would be required to publishis the only similar arrangement between ethnographer and people studied that I have found in the realm of academic ethnographic research not sponsored by parties interested in applying or controlling the results.12 The agreement between Farrer and the Mescalero Apache was verbal. A written contract would have been too reminiscent of the many broken treaties between European Americans and Native Americans. Although they always have existed in traditional anthropological and social science research, ethical issues about how to implement research have not been in the forefront of research practice until recent decades (Fluehr-Lobban 1991a: 1535). It is especially important that the effects of scholarly publishing on those studied (Appell1978,199-200) was rarely examined. In the 196os, predated by a small group in the 1940s (the Society for Applied Anthropology), American anthropologists formally recognized a responsibility to protect the dignity of people they study. This principle of attempting to prevent adverse effects as a result of research activities, and of refusing to undertake research that has that potential-in short, of attempting not to cause harm to those studied-has remained at the core of anthropological research ethics (American Anthropological Association 1990). Over the past three decades discussion of research ethics in the field of anthropology has increased significantly.Discussions cover ethical issues in many aspects of all types of anthropological research, including applied research (for example, Appell 1978; Brettell 1993; Cassell and Jacobs 1987; Fluehr-Lobban 1991b; Rynkiewich and Spradley 1976). This trend seems to enjoy disciplinary institutional support. For example, during the 1993-94 academic year the theme of the Anthropology Newsletter was the ethics of fieldwork. Some of the issues discussed by anthropologists relate to the practice and the effects of scholarly publishing They include, for example, whether to maintain subjects' anonymity despite their wish to be named (Jacobs 1987,24-27); a variety of issues relating to plagiarism and authorship credit; and whether to disclose illegal activities of those studied (Appell1978, 209). However, although there has been concern about the effects of scholarly publishing on those studied, researchers and sponsoring institutions, rather than those studied, have been understood as the primary arbiters of ethical correctness-that is, as having final say over what is published (for example, Appell1978,zo5-7)-even in studies characterized as collaborative (Brettell1993,1-24; Clifford 1988).An occasional exception is found

CONTRACT AND COVENANT IN CURACAO

59

in which an ethnographer's authority to determine what will be published is restricted by those studied. For example, Cassell (1987, 62- 65) reports that an ethnographer was told explicitly not to publish information about a Native American group's religious beliefs and practices because they were deemed sacred and private. In addition, some ethnographers (for example, Lawless 1992) are significantly successful in involving those studied to collaborate on what is published about them (Brettell1993.1-24). However, this approach requires a considerable time commitment on the part of those studied, and is most likely to be successful with one or a small number of informants (McBeth 1993). The arguments of Scheper-Hughes (1982, V-xi) are an example of the privilege over publication decisions that has been asserted by anthropological researchers. Scheper-Hughes reports that her published anthropological research offended the Irish people she studied. Their "commonsense world was "shattered" by exposure to different interpretations, and things only spoken about in their community were written down for any reader to know. She argues, however, that if a work "resonates," then, perhaps, it is all right to anger and hurt those it describes: "Any ethnography ultimately stands or falls on the basis ofwhether or not it resonates: it should ring true, strike a familiar (even if occasionally painful) chord. It should not leave the 'native' reader cold and confused. Angry and hurt, perhaps, but not confused or perplexed" (p. viii, emphasis in original). Scheper-Hughes argues for a kind of scientific objectivity that serves to maintain anthropological privilege. She partially justifies her work by claiming that it satisfies some higher good, for example, that people will gain new insights and that unacknowledged problems will be discussed. This argument, however, is antithetical to the stance taken toward my research by the board of Mikvk IsraelEmanuel. The members of the board insisted that no outsider-not even a social scientist-can claim to act on their behalf without their consent and careful monitoring. Members of the board, I think, would argue that they, too, possess moral authority over representations of their lives-this is not held only by the "expert" who claims to know what is best. A 1986 publication by Marcus and Fischer applies the issue of representation to anthropology. Arguing that there has been a loss of "generalizing frameworks" (p. 7) in such varied disciplines as economics, literary criticism, law, philosophy, physics, and mathematics, Marcus and Fischer draw our attention to a "crisis of representation" arising "from uncertainty about adequate means of describing social reality" (p. 8). They extend Geertz's (1973) approach to ethnography as an ". mterpretive" project, in which representations are partial, and need to be as layered and as complex as possible in order to translate cultural understandings to mostly American, English-speaking readers. Marcus and Fischer argue that "experiments" in the representation of social life should consider both the individual people who experience and shape culture, and the effects of large-scale politi-

60 ALAN

F. BENJAMIN

cal and economic conditions on social groups. Aspects of the problematics of representation thus are in the forefront of current anthropological concerns (see Conquergood 1991). However, the focus of Marcus and Fischer (1986) was on literary style. They wrote that many current ethnographic representations are "experimental" in two ways. One, they are an attempt to find new ways to represent the voice of the other adequately. Two, they are an attempt to use the study of other cultures to critique our own culture. Neither experiment addresses the concerns of board members. Board members do not insist on seeing their own voice in ethnographic representations. They do want the power to reject material they find unacceptable. Members of the board are saying, in effect, we do not appreciate being used for your cultural critique if it ignores our sensibilities. Neither literary experiment nor social critique is a higher good to which board members would be willing to be sacrificed. Thus, although connected with a variety of "threads" in anthropological ethics, and with issues of representation that are prominent in a variety of disciplines, this paper calls for an extension of ethical considerations in anthropology. During the remainder of the paper I discuss implications of the contract I signed for fields of scientific research other than social science; for here, too, this contract serves as a vehicle to introduce significant questions about current research practice. Two code phrases often seem to crystallize the opposition to my call that researchers must no longer be privileged over others when pursuing knowledge. Concerns center on the perceived threat of my work either to what tends to be called "academic freedom," or to "the integrity of scientific research" and reporting. Academic freedom is an important and hard-won achievement that refers to freedom from the state. I do not recommend, the board did not request, and the contract discussed in this paper does not invoke state infringement on academic freedom. Too often adherents advocate a radical understanding of "scholarly integrity" that fails to consider the ways in which all research is "compromised." I 3 The notion seems to be that scholarly thought is entirely independent and that any infringement on scholarly independence reduces "scholarly integrity." However, in contemporary scientific practice we find that funding directs research topics and methodologies, as do paradigms (Kuhn 1970), fads in theoretical issues, and political pressures (for example, Becker 1967, concerning sociologists).Governments and communities-however defined-historically have put a variety of limits on research, as have businesses and colonial administrations (Barnes 1979). Reviewers and colleagues shape research. Journal editors and publishers influence what topics are published.

CONTRACT AND COVENANT IN CURACAO 61

As Sue Estroff points out in this volume, scholars who question the integrity of my research because of board influence make apparent the one-sidedness of the trust expected by researchers in the relationship between researchers and those they study. Researchers routinely expect those studied to trust that the goals of research are worthwhile, that the knowledge gained will be used for positive ends, and that researchers will treat those studied with respect. O n the other hand, those studied are perceived routinely to be biased or short-sighted-that is, concerned primarily with their own narrow interests rather than with the research questions. Thus, the people being studied often are perceived as a threat to the integrity of research, but the subtle, hidden threats of researchers and research toward those studied is "solved conveniently by the signing of informed consent formsa procedure limited in scope and replete with its own problems. Researchers too often expect the trust of those they study as their due, yet seem to find it difficult to extend the same trust to those studied. Other notions of scholarly integrity are changing, too. Realizing the social interdependence of scientific research, some now contest the right of scientists to gather and dispose of the knowledge or resources of others. There is growing recognition that the characterization of scientific research and scholarship as objective, value-free, and rational is not universal but situated in particular Western cultural understandings. Supporters of the practice of science are no longer comfortable with the claim that science is a transcendent, absolutely good human activity, with a "right" to override the concerns of those studied. Other values, for example, social justice, may rank as high as science. Objective, value-free, rational research is an impossible ideal and an inaccurate description of scientific practice. l'his ideal seems to assume that knowledge can be, should be, and is divorced from social relations, rather than produced within them. The ideologies of science and the activities of scientists are embedded in, not divorced from, social worlds; and social worlds are never context free. In social relations, there are differences in power; people are trained, disciplined, loved, and hated; people perform social roles (Goffman 1959). Similarly, research questions, methodologies, and findings are produced by people shaped to problematize the unknown in certain ways and likely to notice some things but not others. In short, scientific practice is shaped and directed by the social matrix within which it exists. Moreover, the position that scholarly research is an absolute and disinterested social good serves to perpetuate the interests of researchers themselves. It aids in obtaining prestige, job security, and funding-and is not a disinterested claim (see for example Barnes 1979, concerning social scientists). However, in the ideology of scientific research there is little rccognition of the personal or social context that shapes and "biases" research methods and conclusions. The influence of social context can be difficult to accept when discussing

62

ALAN F. BENJAMIN

contemporary research, so I draw on Gould (1981), who discusses research bias in the nineteenth-century work of Samuel George Morton, holder of two medical degrees. Upon Morton's death, in 1851, the New York Tribunewrote, "Probably no scientific man in America enjoyed a higher reputation among scholars throughout the world, than Dr. Morton" (Gould 1981, 51, citing Stanton). Morton studied the cranial capacity (brain size) of different "races" in an attempt to substantiate a biological basis for cultural and behavioral differences and to prove "Caucasian" superiority over "American Indians" and "Negroes." Morton, renowned for his objectivity, published the raw data from his studies, enabling Gould (1981, 50 - 69) to review those data. Gould found four types of errors, probably unintended, all of which served to confirm Morton's prejudices. For example, all miscalculations and omissions were in Morton's preferred direction, and the mustard seed used to measure skull size consistently was packed more tightly in Caucasian skulls than in Negro skulls, thus resulting in measurements of Caucasian skulls that increased their reported sizes.'* In consequence, Morton-the renowned, objective scientist-produced evidence that supported his prejudices. He claimed to have found that, due to innate differences, Africans were less intelligent than Europeans. In reviewing Morton's raw data, however, Gould found that the "racial" differences in brain size either were not significant, or correlated with difference in body size. In our own times, biomedical research often classifies African Americans as a race, that is, as a genetically homogeneous group-a salient point in purportedly biological research. However, a considerable number of African Americans have European genetic ancestry. African Americans are not genetically homogeneous. Furthermore, the distribution of genetic material within people from Africa and from Europe varies significantly, both from one region to another and within regions-further evidence of the lack of genetic homogeneity among African Americans. Thus, when biomedical or other research classifies those studied by race, and "explains" difference according to these supposedly biological categories--which fail to acknowledge genetic diversity-researchers perpetuate the racist generalizations so prevalent in American society. The contract I signed implies a vision of scientific authority more limited and contextualized than usually is invoked. In this vision, researchers are reflexive about and critical of the very endeavor of research. We regularly think about the possible injustices that research may initiate or perpetuate, and the proper response to these injustices. For to disregard this issue is to maintain unequal power relations between researchers and those studied. Educating researchers and students to make themselves aware of the concerns of those they study is an important way to focus attention on this issue, but systematic regulations are not. Regulations, too, reflect and are produced within unjust systems. Moreover, as Osborne (1993, I, 6 ) points out, research occurs at a sin-

CONTRACT AND COVENANT IN CURACAO

63

gular time and space, involving specific people and histories. He advocates a "contextual ethics," in which researchers and those studied, by virtue of their shared engagement at a particular moment, share responsibilities "to each other's lives and cultures." Similarly, Graves and Shields (1991) point out that we do not have the power to keep all the promises that we make to those we study, notwithstanding the American Anthropological Association's Principles of Professionul Responsibility (1990). Graves and Shields suggest, instead, that ethical responsibilities, like other social relations, in fact are multivocal. In practice, ethical decisions devolve to the voices of researchers, sponsors, and the people being studied. Researchers can ask those studied what they want in exchange for the favor they grant in agreeing to participate in research. A contract or agreement providing those studied the right to review publications is not necessarily what I recommend. The key is to ask those being studied what they want. One size does not fit all. Those studied may wish, for example, for a contract such as mine, for medical services, for a report on the research findings, for an educational program on the research topic, for a cash payment, or for any number of things unimaginable to me. Of course, not every wish can be gratified. Negotiation may resolve the issue; in the alternative, researchers may decide to study a different group, or to alter the research topic or methodology. Not all research is focused on a clearly defined community, and there will not always be a community representative with whom to negotiate. In such cases the need to consider the concerns of those studied remains, but the means to this end cannot be prescribed. All research is a social activity, embedded in socially contextualized ideologies and practiced for socially recognized goals. Research brings people in contact with each other in situations fraught with differential power relations and cultural and personal complexities. As a social relationship, research implies the existence of mutual obligations between researchers and those studied. In effect, a covenant is always established. The covenants implicit in research should reflect increased reciprocity in the relationship between researchers and those they study. As the philosopher MacIntyre (1993, 5) writes: "The outcome of successful fieldwork is achieved understanding and achieved understanding is defective as understanding, when it is not morally informed understanding"

Notes I. Special thanks go to Sue Estroff, whose assistance was invaluable at every stage of this project. Julia Crane, James Peacock, and Jack Sasson also contributed to the development of this paper. Carole Cain and Peter Hervik read an early draft closely and critically. Dorothy Holland read several drafts closely, and Margaret Wiener read a late draft closely. The Memorial Foundation for Jewish Culture contributed funding for the research which

64 ALAN F. BENJAMIN

led to this paper. The Archaeology and Anthropology Institute of the Netherlands Antilles, both individual members-especially Director Edwin Ayubi, Rose Mary Allen, and Jay Haviser-and the institution as a whole, graciously assisted me during the research. Participants at the conference, "From 'Regs' to Relationships: Reexamining Research Ethics," sponsored by a number of departments at the University of North Carolina at Chapel Hill in November 1995, hclpcd focus my discussion, and conference organizers supported my unusual requests openheartedly. Finally, without the cooperation and the steadfast insistence on accuracy and integrity by the congregational members and the members of the board of directors of United Netherlands Portuguese Congregation Mikvt IsraelEmanuel, my research would not have been possible and this paper would never have been envisioned. All failings, of course, are my responsibility. 2. I wanted the lawyer discharged because he had been slow to produce revised drafts and because several times in the revised drafts returned he either did not include or inaccurately wrote clauses that the board and I had agreed upon and had both asked him to produce. I told this to the board negotiator, who said that the board simultaneously and independently had decided to discharge the lawyer. I am not certain of the reasons for the board's decision. 3. Although the fine appears exorbitant, especially for a graduate student or an anthropologist, I knew I would not do anything to cause the fine to be levied. Thus, I did not raise the amount of the fine as an issue even once during contract negotiations, hoping to place one potential source of contention to the side. Second, the high dollar amount of the fine reveals a major concern of the board, that 1 may choose to leave academia and try to get rich by writing a scandal-mongering piece aimed at a wide audience. This is not an unreasonable concern (see, for example, Sheehan 1993, 87-88), but they were not interested in censoring scholarly publications. 4. John Conley, an anthropologist and lawyer, advises (personal communication) that the U.S. Constitution does not prohibit the kind of contract discussed in this article. The Bill of Rights protects freedom of speech from government infringement. 5. There are a variety of ways to define Sephardi Jews. Walter Zenner (personal communication) sets out three ways. From most to least restrictive the definitions are: (I) Jews from the Iberian peninsula; (2) Jews from the Middle East and many of ;he lands bordering the Mediterranean Sea; and (3) any Jew who is not an Ashkenazi Jew. A further distinguishing characteristic is that many Sephardi Jews have lived for long periods in Moslem rather than Christian countries, and practice a religious tradition that may reflect that history (see Elazar 1989, 13-40) The Sephardi Jews on C u r a ~ a odescend from Portuguese Jews. The term Ashkenazi has referred to peoples of different regions, but since the eleventh century C.E. (Common Era, rather than the Christian "A.D.") it has referred literally to Jews from Germany. More generally it now refers to a geographically diverse subgroup of Jews who share traditions and who mainly are from north, central, or eastern Europe (Elazar 1989,15-17). 6. I use this term relatively loosely, to indicate people who, in general, earn or own more than many others on the same island and thus live a comfortable lifestyle, have access to or are themselves people with power in a variety of realms, and may occupy prominent, honorable, social positions. As a group, the relatively elite Sephardim of Curapo,

CONTRACT AND COVENANT IN CURAVO

65

however, face competing elite groups on Curaqao, include a majority who are not exceptionally wealthy, and include some who arc economically impoverished. In addition, their total numbers are so small that their electoral power is insignificant and they are unlikely to be present in many Curaqaoan social activities. 7. M i h i Israel was the original congregational name. In 1864 the congregation split. The merged congregation, dating from 1964 and reuniting the two, is named United Netherlands Portuguese Congregation Mikvt Israel-Emanuel. 8. At least three members of the congregation, however, had read an unpublished Fourth Semester Paper that I wrote in 1990 (Benjamin 1990). 9. Some claim that respect for text, and a tendency toward a contractual approach to social relations, are characteristic of many Jews. Anecdotal evidence for the former can be found in the example below. However, one should be cautious about endorsing these characterizationsas typical of Jews. Ron Strauss (personal communication) encountered a concern over textual representation among orthodox Ashkenazi rabbis in Israel that resolved itself differently. In 1985, he wished to study birth defects and their prevention among Orthodox Jewish women in Israel. He planned to incorporate Orthodox perspectives on the topic. Although living a traditional Orthodox life, the people with whom he wished to speak were aware of how they appeared to outsiders and did not wish to be misrepresented or denigrated. While arranging introductions through the Office of the Ashkenazi Chief Rabbi of Israel, Strauss was requested to submit his manuscript to them prior to its publication. Strauss offered, instead, to send his manuscript for comments, without accepting an obligation to incorporate the comments. No official response was given; nevertheless the introductions were made. Prior to publication he sent the completed manuscript to be reviewed anyway, but the Chief Rabbi's Office had no comments and requested no changes. 10. I also had made the commitment to write a letter in the congregational newsletter prior to my fieldwork, in the "Human Subjects Proposal" to my Institutional Review Board (IRB). The board's reaction to my illustrates the inherent inability of IRBs - request and researchers to anticipate the concerns of those studied. 11. Political shading of how those studied read ethnographies about themselves has been reported, in particular, when media misrepresentations of ethnographies invoke and thus incite social antagonisms already present (for example, Brettell1993). In such instances the anthropologist has become a symbol around which opposing interests crystallize-despite nuances present in the original ethnography. 12. Contractual arrangements in which research results are reviewed by interested parties or by those studied prior to publication are not unheard of in other realms. This may happen with research performed under contract to a government agency, a private company, or, for example, with research performed for an authorized biography. Such research may or may not be proprietary. Similarly, institutions testing the effectiveness of a product or a program may desire to control access to, and the content of, their tests. These and other types of applied research do not present the same types of ethical issues raised in this paper. 13. That my research might be "compromised" was a criticism put forth by several researchers at the conference at which a shorter version of this paper was discussed.

66 ALAN F. BENJAMIN

14. Mustard seed can be packed tightly or loosely in a skull. If packed more tightly, when transferred to a graded beaker it would result in a greater amount of seed than if packed loosely. Mustard seed measurements of cranial capacity, thus, are inconsistent. Recognizing this, Morton switched to lead shot, yet his published results kept the measurements that favored Caucasians.

68 RENEE C. FOX

Benjamin's contract with Congregation Mikvt! Israel-Emanuel went through seven drafts, over a period of three months; involved the congregation's president and its seven-member, all-male board of directors, who retained a Netherlands Antillean lawyer's expert counsel to help draw up the first three drafts of the contract; established a review committee of five representatives of the congregation (of whom their American rabbi is one) to read, comment on, and approve in written form "any published or publicly distributed material" forthcoming from Benjamin's research (including the "final, bound version" of his Ph.D. dissertation); subjects Benjamin to a fine of $2,800 a day if he publishes any material about the congregation they have not approved in writing-a fine that extends from the day of publication until the day the last copy of the publication has been withdrawn; and establishes a mediation mechanism whereby any disputes that may arise between Benjamin and the review committee "as to whether approval rightly has been refused or not" can undergo binding arbitration. An early draft of the paper that Benjamin prepared for the conference upon which this volume is based had the following admonition stamped on the upper right-hand corner of its first page: "At the request of Congregation MikvC IsraelEmanuel do not distribute this further: this is FOR YOUR EYES ONLY." The contract requires that such a presentation, along with term papers, proposals, and drafts of Benjamin's dissertation, carry this instruction. As he explained in a July 27, 1995, note to the organizers of the conference, "I understand the purpose of that statement to be to deter readers of unpublished and unreviewed material from citing or 'gossiping' about (the Congregation's term) what I write." Although in principle, Benjamin's conference paper did not fall in the category ofwritten communications that he is contractually obliged to submit to the congregation's review committee, in fact, he felt a responsibility to do so. The committee not only read this preliminary draft of the paper, but signaled to Benjamin that they would like him to modify the opening sentence of its concluding paragraph, which read: "The contract I signed is a great inconvenience for me-as I assume it is for reviewers." He was asked by the committee to remove the word "inconvenience" from the text. In response, Benjamin reworked the paragraph. He retained the noun "inconvenience," but deleted the adjective "great" which he had previously used to characterize it. Instead, he specified that "now that my research in the field is complete, following the review process . . . in the contract adds time, energy, expense, and worry to the task ofwriting" Benjamin also prefaced the paragraph with a new sentence affirming that "during [the] field research the contract [had] contributed to . . . easing the concerns of those with whom I wished to speak-a statement that he qualified with a parenthetical phrase indicating that the contract had not been "entirely effective" in this regard. It appears that these changes were acceptable to the review commit-

CONTRACT AND COVENANT IN ETHNOGRAPHIC RESEARCH 69

tee. In any case, the revised draft of the paper that Benjamin presented at the conference was no longer imprinted with the cautionary "FOR YOUR EYES ONLY" statement on its title page. I am appreciative of Benjamin's sensitivity to what he characterizes, both in his paper and in personal conversation I had with him, as the "unequal power relations" that tend to exist "between researchers and those studied," and his respectful insistence on the moral importance of not imposing a "kind of subjugation [on] those whom we study" by "subordinating [their] voices" to the intricately entwined professional and personal interests, needs, and goals of questing social scientists. With each successive draft of his paper, he has more fully and fervently enunciated his commitment to a reciprocal model of research that is responsive to the cultural context and social relationships within which it is conducted, and does not "privilege" the investigator over the persons who consent to be studied. The contractual agreement that Benjamin signed with Congregation Mikvt Israel-Emanuel surpasses the conception of mutual responsibility and exchange that he advocates, by inverting the "professional dominance," of which he disapproves, of the researcher over the persons studied. In my opinion, it goes so far in the direction of subjecting not only the researcher himself, but also the teachers, academic department, university, discipline, and profession with which he is affiliated, to the continuing micro-control of the Mikvt Israel-Emanuel community that it is questionable on both practical and ethical grounds. Most particularly, although the contract explicitly states that the congregation's "right of prior approval does not imply a right to academic or intellectual censorship of Alan Benjamin's writings," it empowers the congregation to curtail the circulation of whatever he writes about them (extending even to term papers and drafts). Indeed, it mandated the review committee to wield gatekeeping influence over whether or not his dissertation was accepted in fulfillment of the requirements for the Ph.D. in anthropology at the University of North Carolina at Chapel Hill. This opens up the possibility of serious restrictions on Benjamin's academic freedom, and encroachment on the academic authority of his teachers and university by a nonacademic religious institution that, in the words of the contract, is "domiciled" and has "its place of business" in another society. Furthermore, even though any issues that arose over Benjamin's dissertation have been satisfactorily resolved through exchanges between himself, his dissertation committee, and the congregation's review committee, it is very likely that he will subsequently be faced with additional difficulties and impediments when he writes articles andlor a book based on his study of the congregation that are destined for publication. Journal editors and book publishers will not easily accede to the conditions that the congregation has set down in the contract, or to the long-armed, adversarial, and fas-

tidiously controlling way in which it reaches out to implement them. Nor should they be expected to do so. In the eyes of Mikvk Israel-Emanuel, the underlying rationale for the contract they forged is to "protect the good name" of the congregation, its individual members, and their extended kinship system, and to shield them from any presentation to the public outside their confines that, to quote the contract, "will have a clear adverse effect [on their] reputations." More is involved in the congregation's concern than the mere fact that it is a "small community" (367 members in 1991), "on a small island" (with a population of about 144,000 persons), where, as the contract states, "their reputations are especially vulnerable to public opinion," and "an adverse effect to a reputation may be irreparable." What the congregation is intent on guarding stretches beyond the everyday, here-and-now sociological meaning of reputation; it encompasses their venerable historical and religious heritage, which extends back to the time of their establishment in C u r a ~ a oin 1651, and to an even older Sephardi-Jewish history of the fifteenth and sixteenth centuries in Spain and in Holland, of which they are descendants. Moreover, as Benjamin points out, in negotiating the contract with him, the members of the congregation's board defined themselves not only as representatives of the living members of their community and the extended family units to which they belong, but also of their deceased ancestors, and of the generations of the "yet unborn" progeny who will follow them. In their view, then, the persons for whom they spoke embody over time and space their religious heritage and spiritual kindred. The agreement at which they arrived with Benjamin is not just a secular contract. In certain regards it has the weight of a solemn religious covenant, as the title ofAlan Benjamin's essay implies-one that binds him to it as a person who was accepted in their community because he is a Jewish male, as well as a representative of academic "scholarship." In addition, the authority that the contract effects over Benjamin is reinforced by the more worldly fact that he is dealing with, and doing research in, what he describes as a relatively elite, well-educated, affluent, and powerful group of persons whose individual and collective social status is higher than his own. The more I have thought about this intriguing and troublesome case of "Contract and Covenant in Curaqao," the more convinced I have become that I would not have consented to do fieldwork under these conditions, or advised a student writing a Ph.D. dissertation under my direction to do so, even though it would undoubtedly mean forfeiting the congregation's permission to carry out the research that Benjamin has accomplished. But as I wrote this, and anticipated how I would feel making such a statement in the public forum of a conference on research ethics, I did so with the "trepidation" that Benjamin invoked in an earlier version of his paper. "I am keenly aware of some absent 'observers,"' he declared. "The people I studied 'look over my shoulder' as I speak." I, too, have

CONTRACT AND COVENANT IN ETHNOGRAPHIC RESEARCH

71

the sense that the eyes of Congregation Mikvt Israel-Emanuel are fixed upon me-disapprovingly-as they bring their sanctioning power to bear on me, albeit vicariously.

Note I.

This phrase is borrowed from John Van Maanen (1988).

THE GAZE OF SCHOLARS AND SUBJECTS 73

appreciate the complexity inherent in a relationship-based approach to research ethics, the first task is to ethnographize the MikvC Israel-Emanuel case from my vantage point as now inadvertent participant-observer in this process. Fortunately, I have retained in my files much of the ethnographic data-the letters, faxes, and other written material-that chronicle the process through which the contract was conceived, negotiated, and signed. It is my intention to contextualize the Benjamin-Mikvt Israel-Emanuel contract, and then to raise questions about its implications for the many roles, relationships, and moral obligations that accompany ethnographic research with human subjects. Situating myself in this process is not an altogether comfortable exercise because I had and retain conflicting views of the nature and implications of the contract, and yet found myself in the position of giving advice to Benjamin. It was difficult to be the sage when I had not encountered previously such a demand from a group of research subjects. From the outset, I considered Benjamin's wishes, views, and judgments about the situation and how to proceed to be of paramount importance. While learning to do field research is a life-long process, the dissertation experience is a unique scholarly developmental landmark. My role was to guide him through this complicated enterprise, not to dictate what he did. Had I been aware that he had breached a scholarly or moral principle in the conduct of his work, I would have intervened. But in this circumstance, it was most important for him to learn to analyze and work his way through this dilemma. Benjamin also knew much more about the principals and the setting than did I. I accepted his desire to negotiate the contract and took as my task making certain that his interests and the interests of scholarly inquiry were adequately protected. Engaging in this critical retrospective has also brought to light the layers of relationships and sometimes ambiguous and even conflicting obligations involved in the research process-this time the relationship between student and advisor, between Benjamin and me. As I prepared this commentary, Benjamin became a subject of my scholarly gaze-in addition to being a student. Our relationship and the expectations, obligations, and boundaries temporarily shifted accordingly. I sought and obtained his consent to quote from his letters to me from the field, and agreed to omit passages or phrases that he found problematic because they were too private or because they might be misunderstood by members of the MikvC Israel-Emanuel congregation. When first introduced to the idea of a contract, I was pleased to see explicit negotiations between subjects and scholar taking lace as the research began. This is yet another era of much hand-wringing among anthropologists about the problematic~of "other knowing," about the moral politics of the production of knowledge, the appropriation of other ~eople'sstories, and contested claims about who represents whom and with what authority (Estroff 1995). Geertz (1988, 130) refers to a "pervasive nervousness" among us about whether our authority to authenti-

74 S U E E. ESTROFF

cally represent "the other" is adequately accounted for by just having "been there" with them. In her vexing and persuasive analysis, "U.S. Academics and Third World Women: Is Ethical Research Possible?" Patai (1991) answers no, repeatedly and elegantly. She warns feminist researchers in particular that they too "reproduce the very inequalities and hierarchies [feminist scholarship] seeks to reveal and transform" (149). In a recent collection aptly titled When Thq Read What We Write (Brettell 1993), Brettell and the other contributors analyze instances in which "they," the subjects of ethnography, read what "we," ethnographers, write, and "they" are none too pleased. The complaints range from what Macklin (1982) has called "inflicted insight" to feelings of betrayal of trust, disagreement about matters of representation and interpretation, and disputes about whose version of events and persons-the subjects' or the author's-should be privileged and published. The problematics of "other knowing" represent significant challenges to research ethics practices and principles formulated prior to the emergence of assertive self-advocacy on the part of biomedical research subjects. Individuals with chronic illnesses and disabilities, and increasing numbers of literate research participants around the globe, can and do read what we write about them. Recently I received via e-mail a scathing critique of my publication in a psychiatric journal, from a member of the psychiatric consumer/survivor/ex-patient movement. I had provided the members of this community an electronic version of the article, which was then posted on their Internet bulletin board for discussion and debate. In this critique, the author accused me of furthering my career at the expense of his sensibilities by getting research funding and publishing articles on the topic ofviolence among persons with psychiatric disorders. He asserted that in his view I was a "wolf in sheep's clothing, tearing at the vulnerable necks" of people like him who were the subjects of my investigations. In his view, my research further stigmatized persons with psychiatric disorders by linking them with violence, and failed to place sufficient emphasis on the violence in American society and the violence ~ e r ~ e t r a t eby d the treatment system on psychiatric patients. Ironically, the conceptual thrust and empirical findings of the article focused attention on contextual and interpersonal factors that were associated with violence by persons with psychiatric disorders. We argued that people with psychiatric disorders lived in interpersonally threatening environments, which had to be considered when assessing the risk posed by psychiatric disorder for violence toward others. My critic and I had similar views, but he did not read my text as I had intended. In this climate, I am challenged constantly by the unfolding complexities of conducting scholarly work among and about persons who claim their right to be active participants in the production of knowledge about them. I wish to respect this claim. Yet what is it that we seek when we consult with and respond to the concerns of research participants? Do we seek their approval for what we say and

THE GAZE OF SCHOLARS AND SUBJECTS 75

how we say it? Do we really want to be challenged about our interpretations, conclusions, and methods? Whose version of a portrayal of an individual-mine or theirs-should be published or privileged?When I distributed the angry critique to a graduate seminar on research ethics, the participants asked difficult and pertinent questions about my response to this man. What did I seek in corresponding with him? What was the goal of any further exchange? Did I seek to him of my goodness as a person? To convince him of my altruistic motives and scholarly integrity? Did I seek absolution? What are the goals of collaboration or of review of texts with informants? While the moral dilemmas p e r a t e d by ethnographic work may have changed little over time, our concern with and attention to them has changed much, in response to these new claims from subjects and colleagues alike. The anthropological community is in the midst of a lively debate about fieldwork ethics and is yet again revising the American Anthropological Association principIes of professional responsibility. Ethnographers are experimenting with a variety of means to include the voices and views of informants in their published work. In his last book, Elliot Liebow (1993) asked each of the homeless women about whom he wrote to review those portions of his manuscript that were about them. Their commentaries on his representations of them arc included in the text. McBcth (1993) describes an ongoing process of life history production with one informant /collaborator in which both their agendas, narratives, and interpretations change over time. Stoller (1988), a psychoanalyst, gives what he calls veto power to patients about whom he has written for publication up to the stage of galley proofs. Despite the enduring tensions in my own work and substantial sympathies with many of the claims of research subjects, I was and am deeply concerned about the reach of the approval mechanisms set out in the contract and the financial penalty associated with publication of any unapproved material. I was and am very troubled by the vagueness of the terminology associated with the congregation's main concern: the "adverse effect to a reputation" of any member of the community resulting from Benjamin's published work. The congregation retains sole rights to define "adverse effect to a reputation1'-even if Benjamin reports information accurately. I was and am concerned about the actual and potential pitfalls of the Directiva's review process, particularly the time factor. But much more than this, I have been concerned about the more subtle, perhaps out-of-awareness yet far more regrettable influences that the contract may have on Benjamin's scholarly freedom and analytic interpretive tasks. Indeed, I, like Benjamin, have felt the gaze of the contract and the congregation on me as I have critically reflected on the process, product, and sequelae of this agreement. With Benjamin's permission, I recount just a portion of the immediate impact that negotiating the contract had on him. In a letter to me dated December 6,1991,

76 SUE E. ESTROFF

he began, "The contract has finally been signed! You saw the 3rd draft. We signed the 7th. I'm not entirely happy with it because the Cong. will have the right to stop anything (including interpretations & conclusions) which harm reputations, but I hope the 'WHEREAS' part, & the part that says they don't have the right to censor, make it clear that the scope of their authority to prevent publication is limited. . . . The process got real sticky. It got me pretty depressed. Much of Nov. I was 'Mr. Lassitude' & 'Mr. Ennui.'" After giving details of unsettling encounters with the Directiva's representative, he wrote, "Anyway, I had already been thinking that it might not work out. I thought, maybe 1'11 just study the other Jewish community-I get along with them better anyway. Well, I felt I had done nothing more than stand up for what I feel I need. . . . I knew I had to show my face at services the Fri. night & Sat. morn. before the mediation, but I tell you, those were some of the hardest things I've done. (I tried to call you for an 'attitude check' but you were out. Probably just as well, the mediation went fine & I'd only have worried you.) So, for much of Nov. I felt like I was in a bunker-trying to keep the bombs from blowing me up. Right now, then, I'm at a turning point." It was clear to both of us that either the contract would be negotiated and signed or there would be no fieldwork. I took the stance of supporting Benjamin as he negotiated with the congregation's representatives, and then providing more technical advice about the details of the contract. At the same time, I found myself also representing the university via the Institutional Review Board, which saw me as the responsible party and a virtual co-investigator, and also representing the other members of the dissertation committee, the anthropology department, and at base, the larger academic and anthropological communities. These various roles and relationships carried with them differing, sometimes conflicting obligations. The roles, relationships, and obligations that Benjamin and I incurred in this process are enumerated in the accompanying table. Benjamin functions in two primary roles in this case: researcher-scholar and gaduate student. He has relationships with numerous parties with attendant moral obligations, depending on his role and on commitments made. It is important to note that much of what is called for in the contract can be subsumed and assumed to be invoked by the principles of autonomy, beneficence, and nonmaleficence. Arguably, the contract simply adds specificity to what must be protected from harm and imposes penalties. In a significant shift, the contract locates the authority for defining harm solely with the congregation, not with the scholarly community. When objections like those of Fox (this volume) are raised about the potential reach of this authority, we must ask why we expect our informants to assume the best about us when we are quite willing - to assume the worst about them. That is, we assume that the congregation will unfairly restrict Benjamin's academic freedom, but that scholars such as Benjamin and I can be trusted to first know and then protect the sensibilities of the congregation. If not the congregation, who -

-

Roles, Relationships, And Obligations ROLE/RELATIONSHIP

OBLIGATIONS

Researcher-scholar

BENJAMIN

Emanuel Informed consent

I

Memorial Foundation

1 Fulfill research plans

Graduate student EstroE Other dissertation committee members UNC-IRB UNC-Department of Anthropology AAAIAnthropology Larger academic community ESTROFF

Contribute to knowledge and practice Learnlengage in critical analysis and scholarship Seek approval and advice Represent, protect interests and reputation Disclosure of activities

Teacher-advisor Benjamin

Advise, support, protect, and represent [advocate Train, monitor, evaluate Inform ldisclose

UNC-Department of

Benjamin

M i h 6 Israel-Emanuel

Larger academic community

I

78 SUE E. ESTROFF

indeed is the appropriate arbiter of competing claims here-Benjamin's to the unfettered production of knowledge and the congregation's to protect their reputations and representations? One of the important questions raised by this case is from which of the many audiences and communities that receive our work should we seek and require approval? In reviewing Benjamin's two primary roles, it is apparent that conflicts between researcher-scholar and gaduate student may arise and that inadequate attention has been given to these inherent tensions in the research ethics literature. A relationship-based research ethics must take as one of its developmental tasks attending to these as yet insufficiently specified and explored topics. Even less well examined are the roles and moral obligations of faculty advisors. In this case, the congregation also viewed me as a responsible party. This was made explicit in June 1993 when I received a letter from them intended to clarify their position about Benjamin's first publicly delivered analysis of the contract-to which they took some exception. In their letter, they historicize their concerns about being studied by an outsider, recounting the difficulties encountered when Dr. Isaac Emmanuel's manuscript underwent review by the Book Committee and the Directiva. In view of this experience, I would argue that unlike many groups and individuals that we study, they were unusually well-informed about what they were consenting to. They had already welcomed an outside scholar to their midst, attempted a collaborative process, and seen themselves portrayed in print in ways that were discrepant with their own views. The letter continues, "The present Board wishes you to know that the point of academic freedom was not-and is not-in question but rather the good name of our congregation and of each and every member of our community which was-and is-to be respected." Herein lies perhaps the most profound dilemma posed by the contract. The congregation's respect for scholarly freedom does not, in their view, supersede their claims to protect-and their desires to monitor and arguably determine-how they are represented by another to others. From our point of view, their claims might, but do not of necessity, subvert scholarly freedom. In my response, I assured the Directiva that Benjamin and I had their best interests fully in mind, that he was following diligently the letter and the spirit of the contract, and that as Jews also we understood their position. In retrospect, I see that I was invoking our shared ethnicity as yet another reason for them to trust us, but offered no other assurances. Nor did I voice any objections to the contract per se. This raises a host of questions, inventoried in the table, regarding what my obligations are to the congregation, to the contract, and to Benjamin. Should I implement and enforce the contract with Benjamin even if I am not a signatory and if I disagree with its provisions? Does the contract apply to me and my scholarly work regarding this case? What are my obligations to Benjamin? Must I consider

THE GAZE OF SCHOLARS AND SUBJECTS 79

how my scholarly views of this case might have an impact on his rapport and standing with the congregation? Further, what are my obligations if or when enforcement of the contract by the congregation impedes Benjamin's scholarly work or damages his professional career, as Fox suggests is inevitable? My view is that while I am no't bound by the contract, I am bound by my commitments to Benjamin, to his successful completion of a graduate degree, and to his professional future. I am obligated to inform him if I think that he has altered or diminished the authenticity and quality of his work because of the contract. At the same time, I am also obligated to the university, the anthropology department, and the larger anthropological community to work toward a productive and morally robust resolution of any conflicts that may arise. The Benjamin-MikvP. Israel-Emanuel contract is relatively uncommon at present, but it will not be so in the future. The challenges posed by this particular form of assertion of shared authority by the subjects of research demand considered debate and resolution that go beyond many current ethical guidelines and codes. To whom shall we turn for these relationship-based principles and practices? How do we engage in actual collaboration without placing the quasi-sacred value of academic freedom in jeopardy? A staggering number of questions arise from taking a relationship-based view of the human subjects and moral issues raised by this case. Does the contract in effect, if not in intention, limit or interfere with Benjamin's academic and scholarly freedom? How might the very legitimate concerns of this group of people have been better addressed? Indeed, who decides what constitutes a legitimate concern? Should Benjamin have refused to sign and then conducted his work elsewhere? If he had done so, should he have returned his funding to the Memorial Foundation? How far do we go in negotiating representational authority and privilege with the subjects of our research? What moral tenets guide my relationship with Benjamin as I take on the role of scholar and critical analyst-in addition to being his advisor? Benjamin's paper in draft form cautioned that it was "FOR YOUR EYES ONLY," at the request of the congregation. We might reverse this request for limited visibility and ask, "How far should the gaze of the congregation reach among us?" By posing the question in this way, perhaps I simply reassert academic privilege. How then do we engage in scholarship that acknowledges and respects this gazewhile at the same time acknowledging and respecting the scholarly obligations to look, to see, and to reveal?

Note I. I was helped enormously in formulating the views expressed in this commentary by conversations with Grayce Sills, the students in my moral reasoning seminar in the spring of 1995, and of course Alan Benjamin. I am also grateful to Quinn Rossander for permis-

80 SUE E. ESTROFF

sion to quote from his e-mail critique. It is impossible to underestimate the long-term influence that Nancy King and Larry Churchill have had on my grasp of moral reasoning, whether or not I have learned to practice it as well as either of them. Renee Fox and Ruth Macklin provoked and stimulated my thinking during the conference at which this case was presented, though neither will be surprised that my views were not changed substantially by that extraordinarily enjoyable exchange.

ers, was then invited to participate in the implementation phase. With all the best intentions, the researchers have no idea how the collaboration might continue through the dissemination stage and how it might affect the use to which the findings will be put. However, this project has broken new gound, and many of the questions that the authors raise cannot be answered in full. We are drawn again, through this case, to the importance of trust in research. This study was designed to overcome African Americans' distrust of the biomedical research establishment. The descriptions of the interactions provided by Blanchard and Strauss indicate that the building of trust on the part of the African American community, or at least its selected representatives, in the research was a major component of their work. To that end, they carefullydefined the procedures that they followed but allowed for modifications based on input from the Community Advisory Board (CAB). The university-based researchers were also intermediaries between the federal government and the community, and needed to trust both their funders and their co-investigators in different sites. Wailoo dissects the process of creating the CAB, the members of which are supposed to represent a "community" at risk for HIV infection, pointing out that, in the end, the CAB'S role is to "grant tacit approval" to the research. Historically, the simple presence of community members in advisory or decisionmaking groups has not ensured effective community representation. Thus, more is called for than just the existence of such a committee. The member-selection strategy, the processes that are followed in and outside of meetings, the relationship of the members to the larger community or communities that they are to represent, and their relationships with the funding organizations are all sensitive issues for which few useful models exist. The previous case was concerned with what some see as too much power in the community. This case reflects the tension that is inevitable when the researchers are seen as those with power by communities where they do their research. The of the researchers in this study was to redistribute and share power in order to facilitate other research, which, as Strauss points out, may or may not be a priority to the African Americans who are considered to be a community. But who really holds power in this case: the governmental funders, the researchers, their universities, the CAB, or the community? The situation is tainted, as is all research that is centered on African Americans, by the legacy of the U.S. Public Health Service's Tuskegee syphilis study. Traditional HIV research has been fraught with ethical dilemmas because of power imbalances, because of the stigma associated with the infection, because of the high costs associated with treatment, and because of the desperation felt both by at-risk populations and by researchers looking for successful preventive or curative strategies. This case represents one attempt to face those dilemmas by building a partnership from which to proceed, thus breaking new ground.

86 LYNN BLANCHARD

There are also some very real concerns associated with participation in an HIV vaccine trial. Primarily, these are in the area of social risk. Social risk is the term used to describe the potential nonmedical, adverse outcomes associated with vaccine trials (Belshe et al. 1994). Significant social risks result from the likelihood that those who receive the vaccine may be misdiagnosed in the future as being infected by HIV because of antibodies generated by immunogenic vaccines. The fear of being identified as HIV positive is rational and reasonable given the discrimination and social consequences that so often result. There may be significant economic repercussions from a positive test; individuals often face the loss of employment and insurance. This misdiagnosis also could result in exclusion from national military and service programs or other employment opportunities (MacQueen et al. 1994). Furthermore, persons who actually become HIV-infected after participating in a trial may only be able to determine their infection with sophisticated DNA sequencing tests. Even trial participants who receive a placebo rather than a vaccine could be placed at social risk. Knowledge that individuals participated in a vaccine trial may result in their being identified as belonging to a group at higher risk for HIV infection. Behaviors associated with acquiring HIV are often those that stigmatize individuals and alienate them from social and institutional support systems. There is little incentive for participating in a vaccine trial that might increase the likelihood of disclosure or stigmatization. Recent media attention to HIV vaccine trials may also have a negative effect on how individuals and communities perceive future clinical trial efforts. The impact of news reports may be similar to the effects of publicity concerning HIVIAIDS on blood donations in the mid-1980s. At that time, donations dropped drastically, even though donating blood was never considered a risky behavior. Widespread misconception and fear of contracting HIV infection through donating blood affected donations as significantly as if donating blood had been suspect. Wire service articles have reported that a limited number of participants in some initial vaccine trials have tested positive for HIV with such headlines as "HIV Infections Cast Pall Over Expanding Vaccine Experiments" (Brown and Weiser 1994). As with many similar reports, information is limited and presents many questions. However, while questions raised are addressed by the scientific community, public perceptions and misconceptions will build. Among African Americans, a group disproportionately affected by the HIV crisis, there are additional known barriers to participation in research such as vaccine efficacy trials. The strong mistrust and suspicion of research and public health programs is grounded in history, particularly the Tuskegee experiment, a U.S. Public Health Service study of the natural history of syphilis in the African American male. In that research, begun in 1932, infected individuals continued in the study without benefit of treatment until 1972, long after the discovery of peni-

COMMUNITY ASSESSMENT AND PERCEPTIONS

87

cillin. This well-documented experience with unethical research provides a rational basis for community mistrust regarding future trials. In addition, a common misconception exists that subjects of the Tuskegee experiment may have been intentionally infected with syphilis by the government, intensifying the pervasive distrust still felt toward public health programs. All this has left a legacy for a significant portion of African Americans, who believe that AIDS is a form of genocide. In a 1993 survey of 1,000 church leaders conducted by the Southern Christian Leadership Conference, 35 percent felt that AIDS was a form of genecide, and another 30 percent felt that it might be (Guinan 1993). Thomas and Quinn (1991) report numerous instances in which this perception has been publicly recognized andlor discussed, including television shows, news articles, and widely distributed literature. The nature of HIVIAIDS preventive programs further feeds the belief of some African Americans that public health AIDS initiatives are detrimental. Programs promoting condom use, pregnancy prevention, and needle cleaning or exchange are all subject to suspicion and may be seen as ways to reduce the African American population.

Description of Project As a consequence of this background, there is general recognition that designing HIV vaccine efficacy trials must involve collaborative efforts between communities and research institutions, and that research strategies should be responsive to community concerns. Developing true community partnerships necessitates comprehensive and formative community-based research efforts. Ideally, information gained from our study will be used to help inform the design and implementation of future HIV vaccine efficacy trials in a positive way. When first beginning to grapple with the ethical questions raised in this project, the overall research team agreed on the following statement of principles to p i d e their planning: This project seeks to learn about community and individual perceptions of medical research and drug trials related to HIVIAIDS. The decision to participate in a clinical trial is an individual one, but it happens within a community context which must be considered when planning research. Collaboration with the community is central, and projects should be designed in partnership to be sensitive to community concerns, build on existing strengths, and increase capacity to address future issues. This project is not directed toward enrolling persons in HIV research or using results to circumvent community concerns. Rather it seeks to provide a forum for

88 LYNN BLANCHARD

communities to identify their concerns and ideas and then to develop ways communities, researchers, and government agencies can work together. An essential first step in developing community partnerships is to define "community," a term that has been given numerous operational definitions. Community may be used to describe a geographic, functional, or structural entity. This project was developed using the term described by Hatch and his associates (1993): "Community can be defined also by shared interests, common fate, social and political history, and cultural affinity. Blacks share a bond to the history, ethos, and institutions that form the heart and soul of the identity known as the black community. . . . Researchers seeking community models must consider both the geographic boundaries of the black community and the sense of cultural identity that also unites it." As described earlier, Project LinCS was a multisite, collaborative study. Thus the most significant effort was directed at establishing core research questions and protocols across sites. The U N C site defined the African American community in nearby Durham as the community of interest; the University of Pennsylvania worked with injecting drug users in Philadelphia; and the University of California at San Francisco worked with young gay men. As one would imagine, along with some similarities, there were substantive differences in considering these target "communitiesn-and in developing appropriate data collection instruments. However, through a process of development and review from research teams and Community Advisory Boards at each site as well as the Centers for Disease Control project group, the following methods were agreed upon. First, the target communities were identified according to three dimensions: behavioral risk for HIV infection, geography, and a shared worldview concerning biomedical research and HIVIAIDS. Each respondent participated in open-ended interviews covering three content areas: community environment, vaccine issues, and sources of information and information networks. These three interviews were designed to consider both community level and individual level factors that are likely to have substantive influence on supporting or opposing HIV vaccine trial participation. The community environment interview was designed to elicit information on how individuals in the target communities define community, conceptualize the parts or segments of their communities, view leaders and roles of influence in their communities, see the strengths and needs of their communities (particularly as related to HIVIAIDS), and feel about biomedical research, past, present, and future. The vaccine issues interview was developed to find out how individuals in targeted community segments view vaccines in general and HIV vaccines in particular, how they consider specific concerns and benefits around testing of HIV vaccines, how they respond to information about HIV vaccines and testing, and how

COMMUNITY ASSESSMENT AND PERCEPTIONS

89

they would seek information about participating in a vaccine trial. The interview about information sources and networks was directed at identifying the media outlets that individuals turn to for information, particularly information relating to HIVIAIDS, whom they consider to be trustworthy sources of information, and the primary social context in which people share information and make decisions. The interview was also aimed at gaining insight into individuals' "functional communitiesm-the people with whom an individual directly interacts. During a later stage of the study, we employed focus groups as a confirmatory method, basing their design on what was learned from the open-ended interviews. Each site also developed site-specific activities addressing their unique target community, as well as an interview for leaders encompassing aspects of the open-ended interviews. Other activities undertaken or planned at the U N C site included an extensive review of the literature, a media watch, a community diagnosis, and a survey of awareness and perceptions of the Tuskegee study. The process of developing the methods for this study was challenging, positive, and enlightening. The development of and relationship with the Community Advisory Board (CAB) was critical to the progress and success of the project. To illustrate this, I return to the Durham experience, describing the rationale for choosing Durham as a study site.

Location of Project Durham, North Carolina, was chosen as a study site for several reasons. These include the high prevalence of HIVIAIDS, a strong history of community commitment to addressing social, poiitical, and medical issues, the presence of community organizations and agencies interested in working together to address the HIVIAIDS crisis, and existing community-university partnerships. Durham County, located in north-central North Carolina, has a total population of 181,835, according to the 1990 census. The percentage of African Americans there (37.2 percent) is substantially higher than in North Carolina as a whole (22 percent). While Durham is sixth in total population among North Carolina's IOO counties, it ranks third in the total number ofAIDS cases reported (395). Durham County has the highest prevalence and incidence rates of AIDS in the state, triple that of the state rate (North Carolina Department of Environment 1994). In the immediate and surrounding area, there are three major providers of care for persons with HIVIAIDS in Durham County. U N C Hospitals has an N I H AIDS Clinical Trials Unit, and Duke University Medical Center has an HIVI AIDS treatment program with special emphasis on women and children. Both institutions have large HIVIAIDS clinics that work closely with the Durham County Health Department, which is the primary ~ u b l i chealth agency ~roviding care for persons living with HIV or AIDS in the county and provides HIV

90 LYNN BLANCHARD

antibody testing. From January through April of 1994, over 1,000 persons were tested for HIV, Go percent of those tested were African Americans, evenly divided between females and males. The Health Department's Early Intervention Clinic (EIC) receives federal funding support for providing outpatient services to individuals with HIV and AIDS. Eighty-two percent of their clients are African Americans. Serving the population seen in the clinic presents several challenges to consider when developing community-based clinical research efforts. First, clinic staff report that there is a high no-show rate for appointments, and the reasons are not well understood. If these patients are enrolled in a research protocol, missed appointments would have a significant impact. However, it is important to note that the staff states that most individuals who do not show up at the time of their appointment do come in for care eventually, just not at the scheduled time. Second, many EIC clients have drug and alcohol problems. Staff indicate that a significant number are under the influence of alcohol or drugs when they come to the clinic for care. This raises serious questions in relation to informed consent and treatment. Last, almost one-fifth of EIC patients are homeless. Whether their living situation results from their HIV status is not clear, but these individuals are particularly disenfranchised. They would certainly present unique issues if they were participating in any longitudinal research. Durham County was appropriate for the study not only because of the magnitude of the HIVIAIDS crisis there, but also because of its strong history of community involvement. Organizations such as the Durham Committee on the Affairs of Black People, The People's Alliance, and the Piedmont HIV Health Care Consortium are well established and experienced in addressing community concerns and issues. The largest African American-owned insurance company in the world is located in Durham. In a book detailing North Carolina politics, Luebke (rg91,1zz-z3) reports that Durham is home to one of the only effective and wellorganized political coalitions of African Americans and whites in North Carolina. Since the 193os, African Americans have played a substantial role in the political and economic life of Durham.

Community Advisory Board A challenge to the research team was to establish a CAB to provide insight into how this project could explore the myriad of issues that HIV preventive vaccine trials would present. The team recognized that the CAB should include members with diverse experiences and opinions, but who could also work together effectively and represent larger community constituencies. Initially, we had discussions with the CAB for the Health Department's EIC about serving in this role. However, EIC CAB members felt that they could not adequately represent the Dur-

COMMUNITY ASSESSMENT AND PERCEPTIONS

91

ham African American community and suggested that a new CAB be formed. They generated a list of potential representatives, taking into consideration key individuals in HIVIAIDS work, political activities, and community development circles. Through several additional channels, the list of potential members was expanded. Research team members talked with agency contacts, personal friends and acquaintances, and individuals suggested by others. People were told about the project and asked who would be important to include on a CAB. We asked not only for the names of individuals, but for the types of representation we should include. A list ofabout twenty-five people resulted from this process. All were contacted with an introductory letter which was followed by a phone call. In general, response was positive; the only reason anyone gave for not participating was being too busy with other work. Those who refused were asked if they could suggest others. At initial meetings there were discussions of "who we are missing," and additional invitations issued. In the end, fourteen people agreed to serve, including an attorney, a fireman, a minister, several community activists, a director of a youth program, and a resident of public housing. There were also outreach workers for a Health Department program, a counselor for a substance abuse treatment center, and the director of a community resource center. There were equal numbers ofwomen and men. In addition, there were individuals who fall into the defined risk categories for HIV as well as at least one individual who was HIV positive. Monthly meetings were held at the local historically black university in Durham, lunch was provided, and members were given a small stipend (though many chose not to take it). As the CAB was being constituted, and once it began meeting, everyone recognized that individuals agreeing to serve were taking on a level of risk. For example, at the first meeting, one individual asked a researcher to go to the campus security office and explain to the personnel there that he did not have AIDS. H e had stopped for directions and felt that when he mentioned what the meeting concerned there was a tangible difference in how they responded to him, based on an assumption that he was HIV positive. At other times, the CAB discussed what it would mean to members personally if information about or from the study were misused or misrepresented. Despite or perhaps because of this, the CAB was very much a "working board," and there are numerous examples of the importance they place in their role. At the first meeting, as the CAB and the investigators were sitting down to the table, the door opened and a young community activist stepped into the room. She announced, "I know I was not invited, but I was sure you wouldn't want to start without me." She has continued to be an active and vocal member of the CAB. The investigators asked for permission to tape-record the CAB discussions, because points raised were valuable and the investigators did not want to miss them.

92

LYNN BLANCHARD

The CAB itself desired tapes so that a member who missed a meeting might review the deliberations. As the content of the discussions became more informed, the investigators wondered if they might serve other research purposes. However, this had not been anticipated, and the CAB members had not been given the opportunity to provide formal informed consent. In discussing whether or not their meetings should become a source of data, the CAB unanimously stated their desire that meeting transcripts be used as primary data. In addition, they suggested signing a consent form-to enable the use of the data and to indicate that they were willing to formally consent to participate in activities into which they were enlisting other community members. They helped the investigators to draft an informed consent form, which each CAB member signed. In addition to giving consent to use CAB meetings as an official source of study data, a number of' members asked to be study participants. Several stated that they would be willing to help with recruitment, but only after they themselves took part in an interview. However, during a lively discussion of recruitment, in which most members decided that they would each recruit three subjects, one member quietly but strongly stated that he had not agreed to participate in recruitment when he agreed to serve on the CAB and was not comfortable in that role. The ensuing discussion clarified that recruitment of subjects, like other CAB activities, was voluntary, and that individual members should freely express any concerns about participation. The CAB felt strongly that before recruiting individuals for the interviews, there needed to be an informational brochure and press release. These were developed over several meetings with substantive and thoughtful discussion. CAB members decided to have their names listed on the brochure. The research team was approached by a ~linicianl~raduate student interested in observing one or more CAB meetings. The CAB discussed their need to be open and frank in exchanges, as well as how permitting an observer would affect their feelings of cohesiveness, and they denied permission. They agreed, however, that there would be circumstances in which they would be willing to have visitors. The research team was interested in doing a small survey of people's awareness and knowledge of the U.S. Public Health Service's Tuskegee syphilis study. After reviewing the proposed questionnaire and methods, the CAB called a halt to it. Although they supported the idea, they felt that doing the survey could contaminate future findings for the study. They wanted no chance that community members would have their thoughts initially drawn to the memory of Tuskegee, just when this project was beginning to enter the field. It was clear that they wanted to ensure that no confusion developed between the projects and that no overlay of negativity would be imposed on the current research. This was a very astute scientific recognition-and a notice to the researchers of how word of the study can be expected to travel.

COMMUNITY ASSESSMENT AND PERCEPTIONS 93

CAB meetings were extended from ninety minutes to two hours, and often members remained for the "meeting after the meeting." One meeting was scheduled on a day in which the area experienced record-breaking low temperatures, and schools were closed due to ice and snow. Over half of the CAB was in attendance. The research team and CAB representatives attended a national meeting on HIV vaccine research. At a poster session describing the project, a researcher from Africa expressed a great deal of interest in the relationship between the Durham CAB and the researchers. When he asked what would happen if the researchers wanted to ask a question the CAB did not want them to, the principal investigator deferred to a CAB member. The CAB member stated that he could not imagine the researchers wanting to ask something that should not be asked, particularly as community input had been a part of the study from its inception. He explained his belief that the role of the CAB was to work with the researchers to determine how the needed questions should be asked. Despite this reassuring response from the CAB member in the last example, the situation brings up important issues to consider when working in partnership with communities, including whether they would have "veto power" over what is being asked. This extends to whether they have control over findings and how they are disseminated. In Durham, the CAB was particularly concerned that the study not depict the community from a deficit perspective, but that a balanced view be given. This was taken into consideration in the design of questions being asked, particularly in the community environment interview. Thus, respondents were asked about their perceptions of the strengths of the African American community, in general and in response to HIVIAIDS. However, it is interesting to consider how the issue might be addressed differently if concern had been expressed not by the community board but by an industry involved in the research. The model of community-based research in partnership with communities raises many ethical and social issues. The researchers and the CAB considered all of the issues that have been raised, along with others such as how to develop and implement research and disseminate findings as true collaborators; how information gained from the study can be used to benefit communities; how to involve the biomedical establishment in community-based research issues; and whether effective HIV vaccines that might be developed will be available and accessible for underserved populations.

Note I. Project LinCS was sponsored by the Division of HIVIAIDS Prevention, National Center for HIV, STD, and T B Prevention of the Centers for Disease Control and Prevention and with funding from the Division ofAIDS, National Institute for Allergies and Infectious Diseases, National Institute of Health.

ADVISORY BOARD-INVESTIGATOR RELATIONSHIPS

95

legacy of racism, reinforced by the recall of the Tuskegee syphilis study, have given rise to very reasonable misgivings about the conduct of research (Brandt 1978; Jones 1992; Phillips 1993; Thomas and Quinn 1791). Studies such as Project LinCS (Linking Communities and Scientists) must enter the field aware of the distrust that they are likely to encounter (Cohen 1991; El-Sadr and Capps 1972; Mays and Jackson 1991) and anticipating the difficulties of doing HIV vaccine-related research (Des Jarlais et al. 1991; Koff and Hoth 1988; Mays and Jackson 1971; Porter et al. 1989; Porter et al. 1993; Tacket and Edelman 1990). It is critical that researchers be aware of ethnicity-specific preventive and HIVIAIDS issues (Airhihenbuwa et al. 1992; De La Cancela 1989; Foster et al. 1993; Friedman et al. 1987; Hutchinson 1992; Seltzer and Smith 1988), including those that relate to !gender (Dancy 1991; Flaskerud and Flaskerud 1987; Kline et al. 1992; Peterson and Marin 1988; Quinn 1993). In the Project LinCS study, the investigators shared a genuine commitment to "doing research differently" so that the trust engendered in developing and implementing the project would not be betrayed. Thus, the research model the project employed is a community-based research design in partnership with community organizations and community members and engaging a CAB. This model of research departs from much ethnographic or social science survey research in that it explicitly incorporates the subjects of research into the design and conduct of the inquiry, raising a variety of issues this commentary will explore.

Ethical and Social Issues We need to be open to hear any kind of concerns andfears, not only about the terminology that? being used, but also about how to plan to keep communication open . . . as they take part in a clinical trial, they will go out and talk to their friends and neighbors . . . and all of that, and they need to be comfortable using the jargon, or being able to translate thejargon into whatever means they can use to share with fdmily members, andfriends and whoever else they want to talk to about it. -A

CAB member

From the outset, the investigators in this project knew that if true collaboration was to exist between communities and the university, relationships had to be developed before actual projects could be undertaken. The partnership-building process does not fit easily into private foundation or governmental grant funding cycles. The short turnaround in the production of many funding proposals is built on the expectation that communities are readily available to serve as the source of research subjects, particularly if a substantial financial incentive is offered. In the conduct of community-based research, partnerships had to precede the grants. In the North Carolina study the grant application was developed only after close consultation with sponsoring community agencies that have had long-

96 RONALD P. STRAUSS

term previous service and research projects with the investigators. Community organizational support was a prior condition to seeking funding. The immediate dilemma faced by both the investigators and community organizations was in how to define "the community." Many people suggested that in the field site, Durham, North Carolina, there is not just one single African American community; rather, as is generally the case, there are many different African American communities. African American leaders and community residents found themselves standing together in front of city maps, arguing about where to draw the lines delineating community segments. The search for what the community is, and where it is situated, was a prelude to the more difficult consideration of exactly who speaks for the community or communities. One of the first steps to conducting this research was the formation of a CAB. Deciding who should sit on the CAB was a careful and complex process. Should it be leaders, "grassroots" citizens, or both? Whose voices needed to be heard on the CAB? Who makes the decisions about the composition of the CAB? What power does the CAB have? What is the CAB'S role in the research process? In this case, the agencies and investigators took the first steps and then the initial CAB group selected additional members itself. For the investigators, the motivations and agendas of CAB members required close scrutiny, as they were seen as holding substantial power over the research. I don't think the investigator team will ever quite forget the first CAB meeting where, as Blanchard describes, a carefully balanced, hand-picked group of African American advisors was interrupted by the entrance of an uninvited community activist. Needless to say, we saw this as an opportunity to hear another voice and invited her in; she remained a vital and challenging member of the CAB. The lesson taken home was that the control that investigators have traditionally exerted over the conduct of their research is substantially altered in this model of inquiry. This shift in control also implies a fundamental change in ownership of the data and findings. In this study, the CAB was the principal vehicle for community ownership of the research. However in the beginning, the roles of the CAB were unclear to both the conveners and the advisors, and they continue to evolve. To the conveners, the CAB was there to assist in the formation of the research project and to provide a visible public identity for the study. They would, it was hoped, facilitate access to the community and enhance community acceptance of the research team. The CAB members appeared to have various understandings of their own roles that included: monitoring the research process for signs it could damage the community; developing assurances that the community would not be used and then disregarded when the data were collected;

ADVISORY BOARD-INVESTIGATOR RELATIONSHIPS

97

exerting political power and province over a segment of the community; contributing to reforming how federal research is conducted in communities of color; and helping stigmatized communities cope with the HIVIAIDS epidemic. Some of the CAB members also considered themselves to be co-investigators who might help to collect data and interpret findings. Some also envisioned themselves as arbitrators of resources to help decide where and how research funds might be spent. All CAB members considered their public credibility and image to be at risk with this project and were very invested in ensuring that only positive outcomes occurred. The CAB'S deep involvement raised basic questions about who holds power and ethical responsibility over the research; is it the subjects, the community advisors, the investigators, the community partnership agencies, the university, or the grant funding agency? The distribution of power over research under this model remains a dilemma. If the CAB did not want something to occur in the research, their wishes were respected. For instance, as Blanchard describes, when the investigators wanted to do an initial small-scale community survey to learn about the degree of community recall regarding the Tuskegee experiments, the CAB voiced absolute opposition. In this situation, the investigators saw the CAB'S stance as astute and concurred. However, if under other circumstances the investigators wanted some research component to proceed, would thc CAB acquiesce?The answer has depended on the rationale and justification offered for a course of action. For example, when the investigators initially considered what questions to ask about intimate sexual and drug behaviors, the CAB expressed a high degree of concern about generating controversy. The investigators prepared a justification for including these sensitive questions, and after considerable discussion, the CAB approved their choices. True tests of the sharing and equality of power have not occurred in this project. However, it is worth pondering whether, when one party has complete control over financial resources, there can be true equality. Furthermore, in this project there was a contract with the funding agencies, and it was apparent that they saw the university as responsible for the delivery of a research product. This product was to be defined in discussions and meetings between the three research sites, the funding agency, and its outside consultant anthropologist. It was difficult for the CAB to participate in those discussions since CAB members had work and family obligations. Their medium for input into the multisite agreements was through advice to the investigators from the university. Funding agency staff sitevisited the project periodically and did meet with some members of the CAB and community agencies on those occasions. Their reception, while cordial, appeared

98

RONALD

P. STRAUSS

to be affected by CAB perceptions of the agency ~ersonnelas expert representatives of the government, who were expected to be fully knowledgeable about all aspects of HIV vaccines and research. The nature of the CAB member role raises further questions. Are CAB members actually collaborators in the research? Should they serve as informants to the investigators? Should they be co-investigators on published papers? Should they hold power over what will be included, or not included, in future publications? Can they publish results in community media, just as investigators publish in scholarly media? Should they serve as recruiters for subject enrollment? Should they serve as ~ u b l i crelations representatives from the project in the community? Perhaps the most vexing issue in Project LinCS was whether CAB members themselves could serve as subjects in the research. In elaborating this research project it became apparent to the investigators that the CAB itself constituted a representative, in-depth, ongoing discussion group on the research topic-in other words, an excellent source of primary data. As Lynn Blanchard describes, CAB members saw themselves as subjects of the research in ways unanticipated by the investigators. This is hardly the course that investigators generally envision for the informed consent process. The CAB'S informed consent was not initially approved by the Institutional Review Board (IRB), and it was retrospective in nature-the CAB wanted the investigators to use months of tapes as project data, even though they had been recorded before that use was contemplated by the investigators and before formal informed consent to that use was sought. The CAB indicated that they had assumed that their words and ideas were already part of the data to be used in the project. The investigators expressed a worry that changing roles months into the research was a risky violation of existing agreements with the project and the IRB; the CAB itself did not appear concerned about this change. Several other specific ethical issues arose in the conduct of this project. One issue related to the use of financial incentives or payment for research subjects. In communities where there are high rates of poverty, financial incentives for subjects may be highly compelling, even appearing to some as bribes. The issue of compensation is particularly poignant for subjects in studies who are known substance abusers and who are likely to use the project compensation to purchase drugs. The investigators and CAB had to consider whether this project was going to be seen as feeding addiction in the community-hardly a stance one would desire of an HIV prevention investigation. The investigators felt that it was important that research compensation be moderate in amount and be framed as paying for subjects' time and costs associated with participating. Partner agency representatives frequently cited another community research project for its misuse of subject compensation, set so high as to be almost irresistible. Another specific ethical issue that arose related to confidentiality regarding

ADVISORY BOARD-INVESTIGATOR RELATIONSHIPS

99

issues such as HIV status, sexual behavior, and substance abuse activities, when community members are employed as project interviewers. The CAB held that if possible, trained community members should be employed as interviewers and transcribers, thus retaining the project's economic resources in the local area. The project investigators worried that local community interviewers might easily or unwittingly violate the basic protection of confidentiality, just because in a small city they would have more routine and social interactions with their research subjects. The goal of protecting confidentiality proved compelling in this research and nonindigenous African American interviewers were used. The study of sexuality and substance use raises questions about response bias, honesty, and the validity of subjects' responses. In the North Carolina study, persons were sampled from three subgroups: persons at highest likelihood of contracting HIV infection, but not already infected; younger persons at unknown risk for HIV infection; and older established persons in monogamous relationships and presumed to be at lowest risk of HIV infection. The CAB and investigators extensively considered how to manage study participants who claim to be in the lowest risk category but who, in truth, are at high risk for infection given their actual behaviors. There is a validity concern that many persons will not respond honestly to questions about the behaviors that place them at risk for HIV infection, particularly those subjects in long-standing and supposedly monogamous relationships. In addition, studying sexual behavior raised reporting concerns very much like those discussed in Case 5 in this volume. The investigators decided not to collect data from minors, and to apply for certificates of confidentiality-both admittedly only partially satisfactory solutions for the ethical dilemmas raised (see Case 5).

The Rationale of Research The black intravenous drug user has been chosen as the guinea pigfor vaccines. . . . People aregoing to ask why we were chosen. -A

CAB member

This research project represents an investment of funds toward the development of vaccines targeted at currently noninfected individuals in populations at high risk for HIV infection. The concept of a preventive HIV vaccine has broadly appealing aspects. It would limit the spread of the future epidemic. Federal research funding agencies have supported such research and several vaccines are in early trials. However, an effective and community-sensitive methodology for testing HIV vaccines does not as yet exist. An effective research design must be based on understanding more about community perceptions of the dangers and social risks

100 RONALD P. STRAUSS

of undertaking a vaccine clinical trial. The data from this project will facilitate the respectful conduct of a community clinical trial of an HIV vaccine. Nonetheless, as the project's statement of principles, quoted in Blanchard's case presentation, makes clear, the goal of this research project was not to begin a vaccine trial, or even to prepare a community for a future vaccine trial, rather to contribute to a basic understanding of communities that are highly affected by HIV. The project sought to understand how communities are organized, how they envision and respond to biomedical research, and how information about HIV and research flows within a community. The project entailed collecting focused community ethnographic information that should result in deeper understandings of community norms, cultural beliefs, organizations, media, and history. In order to test a vaccine, vaccine trial participants must be at risk for infection. Those trial participants can only be found in communities where there is a reasonably high rate of HIV infection and where behavioral risk factors for transmission are common. Those communities may reasonably be portrayed as often suspicious of researchers and governmental beneficence. Research agencies wisely perceived that without understanding and working with these communities, it is unlikely that clinical trial enrollment would be facilitated. A great deal of community disruption might occur if trials were foisted on targeted vaccine trial communities. Long-term concerns might also exist, since researchers might need to engage in a series of trials that would depend on the continued and satisfied participation of communities. "The laboratory scientists don't know much about communities, they've never really dealt with them, and the community people don't know much about the laboratories," said a CAB member. "Along with teaching the government researchers about how to do vaccine testing we also need to teach communities about what goes on in the scientific world." Feedback from this project's CABS resulted in a new study in which HIV research scientists are being studied, using much of the community research protocol, to ascertain their perceptions about the ethics of community-based research. CAB members initiated and endorsed this research as a vehicle to increase community understanding of science and scientists.

The Uses of the Research The African American community is used to hearing a lot of bad news-that is, that they're at high risk for lots of things, they're at high risk for infdnt mortality, they're at high risk for this . . . so, there are a lot of negative associations with being an African American. Here: yet another thing, oh well we're asking African Americans now [to enroll in AIDS research], because the whole community is at high risk for HIVinfection. That: not true. There: certain people who may be at high risk and identzhing who those people are may be dzficult, but we

ADVISORY BOARD-INVESTIGATOR RELATIONSHIPS 101

dejinitely don't want to addyet another thing that people have to feel bad abou or don't want to be associated with. -A

CAB member

The most critical issues facing this community-based HIV project relate to the uses of the research. Will it be used to build future research in collaboration with community organizations, respecting the values of community members? O r will the information be used to circumvent community resolve, as researchers will have in-depth knowledge of how community media work and how they might effectively deliver a message to engage trial participants? Governmental funding agencies have begun to plan and conduct large-scale preventive HIV vaccine clinical trials. One wonders whether that is the goal of the communities studied here. Perhaps they think other preventive approaches are preferable to vaccines, or perhaps HIV infection does not constitute a principal concern when compared to poverty or violence. One might ask: ifAfrican Americans in the U.S. South were opposed to HIV vaccine trials, would research agencies and drug companies heed their perspective? Some members of the North Carolina site study population raised the question about whether a successful vaccine would be made available at no or low cost to all Americans at risk. Clearly, they anticipated that the historical inequities in access to U.S. health care might actually work to keep persons who are potential trial participants from receiving the eventual vaccines produced. No research team can promise how the information derived from a research project will be used in the future. However, research that is built on trusting partnerships implies that data will be used to benefit the constituents, not only the research enterprise. Findings will be published and their uses may be unanticipated by the groups or individuals who generated them. It is too early to know whether the CAB will seek to publish findings in community media and whether they will seek to exert selective control over scholarly publications. The investigators have a longterm commitment to community-based research in the study setting and have a real interest in ensuring that the integrity of the community be protected. They have told community members that as far as they know, this research is not the prelude to a specific vaccine trial in their city. The use of the research findings to circumvent community desires or norms would be a serious breach of trust. This research model implies that both investigators and research funding agencies will respect the wishes and values of the community in the conduct of future and further investigations. "Beforehand, it was we're the government and we're going to come in and do what we want and you're just gonna have to listen to us," noted a CAB member. "And it looks like they really want to change their tune, and if we're going to be doing this, we're going to have to work with you. Nobody's on top, nobody's on bottom, we're partners in this research."

RESEARCH PARTNERSHIPS AND PEOPLE "AT RISK"

103

are characterized as having large numbers of people "at risk for HIVIAIDS. This particular paper focuses on Durham, selected "not only because of the magnitude of the HIVIAIDS crisis there, but also because of the strong history of community involvement" in social, political, and medical issues. Researchers also selected Durham because of its "existing community-university partnerships." Thus, this particular study defined the community by its disease profile (an assessment made by researchers, media, and-presumably-by numerous individuals within the community). What linked the communities in the minds of researchers was their common experience with HIVIAIDS, or what researchers predicted will be the rising incidence of the disease in the community. The desire to test HIV vaccines is a broad national goal, as AIDS continues to be a prominent and much-feared public health problem. A vaccine trial will be influenced by a variety of research interests that are quite distant from specific neighborhoods. Blanchard sees this fact as a cause for concern, and as the driving force behind this "community assessment." Such a preparatory study may uncover many communication problems before they arise in a full-blown vaccine trial study. The public health perception of these communities as "at r i s k may not be sharedwithin the community. Thus, the study investigatorssought to document local perceptions about research, vaccines, and community health. Moreover, Blanchard recognized from the outset the complexity and importance of "building community research partnership" in order to carry out the anticipated HIV vaccine efficacy trial. The goals of this assessment were thus anticipatory, an exercise in both data-gathering and diplomacy. The assessment sought to gather information about community leadership, the webs of local influence, and prevailing community attitudes toward research. It attempted to examine local concerns about vaccines in general and HIV vaccines in particular, and to build a stable research partnership in anticipation of HIV vaccine research trials. The researchers wished to avoid exacerbating community distrust of research, and to avoid fomenting historical antagonisms toward research. (Here Blanchard evokes the Tuskegee syphilis study as just the kind of research enterprise they do not represent, and from which they seek distance. Blanchard also mentions briefly the widely held perception of AIDS as part of a conspiracy to "wipe out the black race.") ' The assessment sought to avoid such problems by documenting community concerns, and by addressing them openly with citizens before the HIV vaccine trials began. My first comments focus on the concept of "the community," and therefore the ideal of community partnership, implicit in this assessment. As Blanchard suggests, it is crucial that we define and understand the contours of a community, in order to build a relationship with "it." The researchers created a CAB at each site, envisioning this entity as an informal regulatory mechanism-a voice of the community. The members of the CAB are described as people of "diverse ex-

104 KEITH A. WAIL00

periences and opinions, but who could work together effectively as well as represent larger community constituencies." No other precise descriptions are given of these people and their relationship to a preexisting "community," and one wonders immediately about the ethics of researchers' assuming this role-"creating" a voice of the community in this manner. Indeed, the CAB'S first incarnation was problematic. As Blanchard recounts, an already existing CAB, formed for a related but distinct purpose, was originally approached to fill the role for this project. However, its members "felt that they could not adequately represent the Durham African-American community and that a new CAB should be formed." That CAB, clearly, was aware of its status as an appointed and anointed group, chosen by outside organizations to speak for others. The problem of creating a CAB is not unrelated to the challenge facing anthropologists like Marcel Griaule in the 195os, seeking reliable informants and collaborators to p i d e ethnographic studies of the Dogon people. As Griaule noted, he sought to "rely on other specialists and try to form with them a thinking group . . . a tactical unit of research in which each person, while holding to his own personal cpalities, knows he is an intelligent cog of a machine in which he is indispensable but without which he is nothing" (Clifford 1988, 72). As historian of anthropology James Clifford concluded in his discussion of Griaule, -. the anthropologists understood that "every informant . . . enunciates a different kind of truth, and the ethnographer must be constantly alive to its limitations, strengths, and weaknesses" (73). Griaule was well aware of the ways in which informants and teams would themselves legitimate the research enterprise, by embodying the very principles that underlie modern inquiry. Such insights are relevant to our interpretation of the modern-day CAB, current research, and the enterprises of community assessment and HIVIAIDS vaccine trials in the Durham neighborhood. First, let us consider the CAB more closely-the character and concerns of this anointed group, and their own ambiguous sense of their relationship to a "community." One member placed herself on the CAB without invitation, suggesting that a vigorous desire to speak for the community was itself encouraged by researchers as a characteristic of community representation. CAB members accepted the role as spokespersons, allowing their own ideas to be recorded as part of the matrix of community beliefs and perceptions. However, some felt strongly that recruiting people to participate in the assessment was not an appropriate role for them-seeking, it seems, some degree of public anonymity and distance from the research enterprise itself. Their attendance at a national meeting on HIV vaccine suggested that visibility in other professional realms was not a major concern. Their attendance can be seen as a cementing of a lay-professional partnership, a validation of their "expertise," and also a blurring of their roles as informants, community spokespersons, and members of the research team. Finally, they -

-

RESEARCH PARTNERSHIPSAND PEOPLE "AT RISK"

105

maintained a concern (presumably as residents of the neighborhood) that the depiction of this community based on the assessment should be favorable. They realized, it seems, that the assessment would result in a broad-strokes portrait of Durham as a place where HIV and AIDS resided among black people. They sought to temper this probable portrait by convincing researchers to ask respondents about the strengths of the African American community. The CAB was given advisory functions-its members suggested other members who might join, thus making the CAB itself a network of articular relationships intersecting with institutions and individuals who themselves could claim some lace in the community. But in many ways, the CAB was a community unto itself. Over time (by the process described above), the CAB was handed the power to speak for and protect the community. It was granted limited power to direct the study, but it was denied the power to veto some aspects of the study. It was certainly not the same as the community nor a representative of the community in any democratic sense, but it was a temporary enabling institution ("intelligent cogs," in Griaule's phrase) in the research enterprise. For the sake of provoking thought, I would like to suggest some other historical parallels-specifically between these research relationships and those forged between U.S. Public Health Service researchers, the Tuskegee Institute, and residents of Macon County, Alabama, in the Tuskegee syphilis experiment. This is the very study of untreated syphilis in black men in Alabama that is so abhorrent to Blanchard, Strauss, and the CAB. My intention is not to suggest that this preliminary assessment study is in any way the same as the Tuskegee study, nor is it my intention to suggest even a close analogy. Rather, the comparison (raised by the authors themselves) raises for me important questions about the role of African American institutions in the promotion of community-based research (whether these are ad-hoc created institutions such as the CAB, or longestablished institutions such as Booker T. Washington's Tuskegee Institute). Describing the early years of the Tuskegee study, historian James Jones noted the central role of Tuskegee in the early 1930s: "The Institute's cooperation would permit the study to go forward without arousing the fears and suspicions of private physicians. The participation of black physicians would also help secure the cooperation of subjects for the experiment, for the Tuskegee Institute commanded trust and respect among the black population of Macon County" (Jones 1993, 100). Decades later, when the experiment became a public scandal in the early 197os, the Institute "acknowledged that its medical facilities and personnel had been used in the study . . . but maintained that the study 'was acceptable under the clinical conditions prevailing 40 years ago.'" As Jones makes clear, this stance was somewhat disingenuous: "While technically true, the statement ignored the fact that the Institute had given its tacit approval to the experiment by permitting its facilities and medical personnel to be used repeatedly over the

106

KEITH A. WAIL00

years" (Jones 1993,208). The point is not that Tuskegee can be blamed for the callous practices of U.S. Public Health Service researchers through the decades, but that institutional cooperation was integral to the legitimacy of the study in its time. Yet tacit approval offered no long-term immunity from charges of unethical conduct. The CAB, like Tuskegee, is an institution (yet it is an ambiguous, weak, and "constructed" institution) that is externally designated as a voice of the community and granted the power to guide ethical research relationships. Such an institution warrants closer scrutiny than Blanchard and Strauss have given it, as a crucial enabling modality in the research enterprise. To evaluate the nature of the research relationships being forged by this institution, one needs to know much more about the structure of the CAB and how it works. This information is crucial to any evaluation of its ability to guide, speak for, and protect those "at risk for HIVIAIDS. Does the CAB itself constitute a reliable depiction of "the community"? Clearly, one of the virtues of the CAB was its ability to be selfcorrecting and flexible, in the sense that the members of the original CAB approached to fill this role perceived their own limitations as spokespersons and sought a new composition. But whatever its composition, it is clear that the CAB granted tacit approval to the HIV study in a way similar to the ways in which Tuskegee granted tacit approval to the syphilis experiments. Yet the tacit approval of such an institution does not, by itself, constitute ethical, principled, just, or community-sensitive research practice. This study, then, raises a host of issues about institutions and communitybased health research that will not be easily resolved. One significant consequence of the creation of the CAB and the pursuit of HIVIAIDS vaccine trials in this community, for example, is the creation of a group of subjects who (after enrollment in a vaccine study) would subsequently test "HIV positive." This development may have lasting effects on these individuals' search for employment, insurance, and health care, and thus will leave a lasting mark on the individual subject and the community. How does this mesh with the CAB desire that "no overlay of negativity" devolve from this study? The CAB exercised a considerable role in answering such questions and in shaping research practices, and yet the CAB also was made a willing arm of the research enterprise. The relationship between CAB and researchers should therefore be conceived as external to "the community" itself, and only marginally representative of community concerns. Such relationships, in fact, redefine the boundaries and character of "community" in the name of a particular research agenda. The CAB-researcher relationship is a unique institution. Such institutions should be regarded with extreme caution-they are ad hoc, take on ambiguous roles and changing functions, and respond to incentives and forms of credit in a research economy that are alien to the "community" itself. Such institutions

RESEARCH PARTNERSHIPSAND PEOPLE "AT RISK"

107

pose significant ethical challenges and dangers for individuals in the community, for long-term trust, and for the etiterprise of improving health by way of such community-based interventions.

Note I. See Jones (1993).According to one poll, some 35 percent of blacks believed AIDS was a form of genocide.

tion-specifically, accepting the funds as a gift and the decision to publicly acknowledge industry support in all publications. Is this enough? How are we to evaluate whether or not it is?And should we do the same when the funder is the NIH, or an advocacy organization, as we do when research support comes from industry? What ldnd of stake in research results is it permissible for a funder to have, and what is not? In order to address these questions, we need a framework in which to locate the players' interests and motivations. Loretta Kopelman, a philosopher, elegantly demonstrates that potential conflicts arise because science, business, and advocacy have different underlying values and goals. For researchers to counteract conflicts of interest with potential funders, it is particularly important for them to protect their integrity through openness, objectivity and honesty. Popkin's actions were, for the most part, successful in achieving those aims. Allan Brandt and Lara Freidenfelds are historians. They argue that an individual case and an individual choice must be placed in an even broader context before thoughtful decisions can be reached. Researchers exist in a community of scholars who might plausibly all be affected by the choices of individual researchers in Popkin's position. In essence, the second commentary asks that we balance the rights and interests of an individual researcher against those of an entirz research community. Is this fair? The industry that wants to fund research may be purchasing subtle advantages from support of "unbiased research. Are the benefits that accrue to a funder the responsibility or concern of an individual researcher?If so, is there a point at which they cease to be that researcher's responsibility?Who or what agency could adjudicate these issues? Popkin asserts that the NIH is too rigid in its review of research design, and that having no oversight other than the integrity of the researcher may be a good thing. This argument reflects issues raised in several other cases in this volume about who gets to define the rules of evidence and whether or not something is "good science." This is a critical component, since if it is not good science, it is unethical research. A final issue raised by this case is the nature of the "gift" of funding. Popkin's arguments rest upon the belief that accepting the Nest16 funding as a gift to the university provides an "unfettered environment in which to conduct his research. But is this accurate? Much has been written about the complex web of psychological obligations that arise from "gifts" of blood transfusions (Murray 1990; Titmus 1971) and organ donations (Fox and Swazey 1978), or even from gifts by pharmaceutical companies to physicians who might prescribe their products to patients (Chren et al. 1989). Is it ever possible to accept a gift with "no strings attached?

114 BARRY M. POPKIN

The author of this paper headed a group of economists and social scientists who were involved in research on infant feeding in lower-income countries during the late 1970s.~We were approached in late 1980 by a senior executive of Nestle England. He wanted to visit us to discuss our interest in developing a research project on infant-feeding behavior that would extend work already undertaken by our group, and would be funded by Nestle. What followed was a long period of dialope and discussion with friends and colleagues at other campuses, and then with faculty and administrators at the University of North Carolina at Chapel Hill (UNC-CH). We invited the Nest16 representative to UNC-CH, and several months after his visit conveyed to him both our interest and our concern regarding the ethics and feasibility of such an undertaking. We then took a series of steps that helped us to decide whether we would move forward with this option.

History of the Infant-Feeding Project at UNC-CH Our first action was to meet with the director of the Carolina Population Center, the multidisciplinary institute that housed our research. Our group, the director of the center, and his advisory council deliberated and made several key recommendations. The most important was the need to establish a scientific advisory panel of international health scholars to ensure adherence to the best scientific principles in the design, implementation, and analysis of our proposed survey research project. Second, we wanted to receive funding that did not require accountability to Nestle. In 1981, UNC-CH was not experienced in the receipt of major gifts from indu~tr~-~articularlyindustries that were facing the social scrutiny that had been turned on the infant formula industry. Moreover, there was great concern among some of the deans of the university that such a gift would hurt the university's image. After several meetings of the University Chancellor's Advisory Council of Deans and Vice-Chancellors, the chancellor decided to accept Nestlt's @ft. He worked with the investigators to draft a letter that delineated the terms of acceptance, the most critical being requirements that the gift be unconditional (with no financial or other type of accountability) and that the university would be free to fulfill the aims of the gift as it saw fit. Our group was pleased with these guidelines. As a team, we also discussed the importance of being open about funding sources, in order to deal directly with any criticism related to the receipt of this gift. We decided to acknowledge all sources of support in all future grant proposals and papers related to this p r ~ j e c t . ~ A series of literature reviews (Hamilton et al. 1984; Popkin et al. 1982, 1983, 1986; Akin et al. 1985) was used to design a field study that would allow us to ex-

TRUTH-IN-FUNDING 115

amine causal relationships in the areas of infant-feeding patterns to the best extent possible. The project was longitudinal, following a large representative cohort of women and their infants from pregnancy through two years postpartum. We were able to study in great detail the patterns, determinants, and consequences of infant feeding and health, as well as a series of related topics regarding health service use and maternal diet and nutritional status. We included a focus on the role of the food industry's formula promotion in the context of a larger set of factors that affected our key outcomes of infant feeding, infant growth, and morbidity and survival. Our study thus focused on issues of direct relevance to the infantfeeding controversy. The criteria for a study site included variability in infant-feeding patterns within a sample of lower-income women who did not breast-feed, and a mix of urban and rural women. We did not expect to develop a nationally representative sample, but we required a site that could provide enough variability over time and space for a meaningful attempt to unravel key causal relationships. The project was designed to be a study of a natural population. There would be no interventions. As it turned out, midway through the first several years of the project, there were a number of exogenous national macroeconomic shocks on our sample area that created significant shifts in real incomes, prices, and employment patterns. Our UNC-CH group selected as partners in this effort a group headed by Dr. Wilhelm Flieger, a priest who is an eminent demogapher, and a collaborating institution, the Nutrition Center of the Philippines. We developed an initial budget of $1.8 million for data collection, preliminary analysis, and an overview of the design of the survey. We obtained gifts totaling $8~0,000initially from Nestle and Wyeth (American Home Products). This was adequate to cover data collection and the scientific advisory panel, so we initiated work with this support. As it turned out, the inflationary period that the Philippines experienced midway through the project led to a depletion of our funds, but we were able to fund the remaining data coliection and processing costs through individual investigators' proposals to the National Institutes of Health (NIH). Most of our funding came from the NIH, but we also obtained smaller amounts for additional data collection and analysis from the Ford Foundation, the Agency for International Development (USAID), and the National Academy of Sciences. More recently, funds for re-analysis have come from the Thrasher Foundation, the World Bank, and the Asian Development Bank. As a result of our industry-supported research, from 1981 to 1991,over nineteen . - grants, including five large NIH individual grants and one program project, were developed and funded. The list of rcferccd journal publications from the Cebu Longitudinal Health and Nutrition S U N ~(CLHNS), Y as it is called, covers over

116 BARRY M. POPKIN

five pages. In addition, at NIH's request, the group applied for and received funds to create an indexed CD-ROM of the data set from the 1982- 86 survey. We have also received, from a variety of funding agencies, financing for several follow-up surveys of the 3,327 mothers and infants who were surveyed intensively over three years of pregnancy and infancy. A large number of N I H grants have funded other scholars to undertake research on this data set.

What Did We Study? The sets of maternal and infant nutrition and health studies carried out under the CLHNS umbrella are varied. The core studies were represented by five central projects: an analysis of prenatal behavior and medical behavior and their effects on pregnancy outcome; an analysis of the initiation and duration of infant feeding with a focus on breast-feeding; an analysis of the set of underlying and intermediate factors that affects infant health; an analysis of the patterns and determinants of maternal diet and body composition; and an analysis of the role of maternal nutrition and infant feeding in birth spacing. Other projects focused on aspects associatedwith infant health (for example, several projects focused on sanitation and the water supply); the nature of pregnancy outcome and its effects on infant feeding and infant health; natural family planning patterns and their determinants; the effects of the food industry on infant-feeding patterns; and other dimensions of household behavior during the pre- and postnatal period.* This full set of studies represents a thorough examination of most of the health and nutrition issues raised during the Nest16 controversy.

Why Not NIH? When the initial discussions with the food industry took place, we had not considered that this type of project was feasible. It was difficult to imagine the successful completion of such a complex set of surveys. In the early 1980s, the analytic and statistical techniques for handling the issues raised in this effort were primitive and could not accommodate full analysis of the data sets we proposed to develop. Moreover, the scope of the questions we addressed was quite broad and did not fit a traditional individual grant application or site review of a program project grant. For all these reasons, we did not try to present this project to N I H for peer review prior to our industry funding When we did present proposals from the CLHNS to N I H for funding, they were much more narrowly and methodologically focused.

TRUTH-IN-FUNDING

117

Ethical Implications of the Issues Several issues arose over the history of this project that relate to the themes of this volume. One was the general reaction of the scholarly community. Despite our initial concerns, our colleagues were surprisingly open and positive. We were never publicly chastised or accused of being pawns of industry. Second, we found that our support from industry was among the least restrictive we had ever received. We were never directly or indirectly faced with discussions about our methods or results. This, of course, is not all positive, because the peer review approach has important strengths. When we proposed the exceptionally complex research design that we undertook, few colleagues believed we could carry it off and, had we gone to NIH for our initial funding, we would probably not have received a positive review. We needed funds that would allow us the freedom to design the kind of study we felt was needed to answer the research questions we faced, and then to ensure that we could implement this design. In comparison, our experience with grants from what one might call "advocacy institutions" was most interesting. In one case, a close colleague and head of planning and evaluation at UNICEF requested that we undertake a review and case study of the effects of infant-feeding patterns on infant diarrhea. This person was not concerned about our multiple sources of funds, but he received a great deal of negative feedback from his breast-feeding advocacy colleagues inside UNICEF. We received the grant but had to suppress all mention of funds received from industry in publications and reports that resulted from UNICEF funding. A second case related to two proposals that members of our team submitted to Wellstart, a prominent and successful breast-feeding advocacy and counseling group that had received a very large USAID grant to promote breast-feeding Part of this grant provided support for studies on infant feeding and nutrition. Our group submitted three proposals, two ofwhich received strong scientific endorsements and one of which received no review. In the letter of rejection, Wellstart noted that they declined to fund our grant proposals because of our financial support from the infant feeding industry. We developed a carefully worded response to Wellstart that challenged their criteria and stated that federal funds could not place such stipulations on their support. This led to a rather prolonged debate within Wellstart. In particular, the parent organization's board of directors and the group handling the USAID grant differed on the issue we raised. In fact, although the charter of Wellstart specifically bans receipt of industry support, this stance contradicts guidelines linked with their receipt of federal funds. Further, in the course of their deliberations it became clear that dozens of university research groups-including those who argued most strongly against the infant food industry-had also received infant formula company support. U1-

118 BARRY M. POPKIN

timately, the Wellstart Board of Directors revised their guidelines to provide funds to groups who would not commingle industry and Wellstart funds, and would not acknowledge industry support for any data used in Wellstart-funded research. They also grudgingly provided us with two small grants.

The Current Situation Regarding Industry Sponsorship of Research When the food industry approached our group of researchers, the presence of industry on our public research campus was not very visible. A large number of drug trials and related activities were funded by industry, but there were no major joint efforts and there had been no campuswide dialope regarding the role of industry on campus in general, or related to research support in particular. There was certainly no encouragement of academic-industry partnerships, there were no facilities on campus that were linked to industry, and few faculty members in the arts and sciences were even aware that research on campus had any industry support. The climate has changed completely during the past fifteen years. There is currently a wide range of partnerships with industry. Buildings on campus have been constructed with industry funds to support joint university-industry research. Development officers actively seek gifts from industry and support in the nature of gants and contracts for our research enterprise. Academic entities such as the Frank Hawkins Kenan Center support entrepreneurial activity, and the university actively pursues patents related to business-university partnerships. The set of ethical issues has also shifted. Now the main subject of concern is that a researcher might have a financial or personal stake in research funded by industry, and that this might distort research findings. Our concern in the early 1980s about the need to avoid the appearance of bias still exists. If we were contemplating the NestlP gift today, we would still feel the need to insulate ourselves from the concerns of our colleagues that we would somehow be tainted by receiving food industry funding But today we would not face scrutiny regarding the gift, its receipt would have been assured, and the university would be far more active in helping us work out a relationship with the infant feeding industry.

Personal Conclusions Funding Source Biases This project represented the first time that I received industry funding and, moreover, the first time that I had any contact with the infant food industry or any other industry. My own activist background made me hesitant to be funded by in-

TRUTH-IN-FUNDING 119

dustry or even to associate with it.5 However, my experience with this project has shown me that the food industry is willing to support research activities in a manner that, in selected circumstances, can allow scholars to pursue a wide-ranging set of research activities. That experience and subsequent ones have shown me that the food industry is quite willing to remove all types of requirements of accountability that one normally needs to follow for most research funding The gifts are not completely unfettered. Clearly, industry obtains important public relations benefits and in some way is shaping an agenda that is favorable to its broader set of interests. And there are obviously other aspects of the research that benefit industry. In contrast, the advocacy groups mentioned above presented a very narrow and focused agenda for funding. While there were no overt signs of coercion, these groups strongly desired results that favored their own initiati~es.~ This, of course, is understandable given the goals of any advocacy organization. But as scientists, we often focus a much more rigorous gaze on research funded by industry than we do on research funded by other groups that may have an equally restrictive view of the results they wish to see. Conflicts of interest considered by journals seem to relate to financial rather than ideological interests. Without industry support, I doubt that we would have been able to develop the wide-ranging survey research project that became the CLHNS, with its innovative data collection methods. It would have been, I believe, quite difficult to engage in this type of research, which covers a large number of topics and encompasses several disciplines. My N I H funding must continually conform to the current paradigm in terms of acceptable research methods and models, while attempting to push forward our frontiers of knowledge. I am free to explore questions that conform to these requirements, but it is difficult to move into areas outside the current purview of my discipline.

Openness about Funding Sources Both Wellstart and I were surprised by the number of researchers who did not openly acknowledge funding sources. Modern medical and public health journal guidelines require such acknowledgments; however, in the era of large, complex, multiyear projects, it is easy to obtain research support and hide this behind acknowledgments that focus on the most narrow interpretation of support. This is particularly true for the types of public health and social science research in which I engage. I think that the dual standard applied to industry in comparison with other sources of funding is an issue worthy of discussion. I think that journals should learn about all sources of the funding that a laboratory or large project receives. Researchers may only mention the source that funds a particular analysis, but con-

120 BARRY M. POPKIN

cern about conflicts of interest requires that journals provide a much more careful examination of funding sources.

Postmortem I find it paradoxical that I, a former full-time activist, found myself much less constrained in receiving funding from an infant food industry that was so beleaguered and under such scrutiny than from other potential funding sources. More than that, I find it remarkable that I am an advocate of obtaining industry support for research that might directly investigate the industry funding the research. When does the investigator begin to censor himlherself? Our group explored issues that dealt rather directly with some industry practices that were claimed by advocates to be very harmful, and ended up with results that were highly damning of the effects of some of the infant food industry's actions. But as with most science, we also produced results that showed that the world is much more complex than was presented in the early debates on the infant formula controversy. Let me provide some examples: Breast-feeding is declining. This is clearly shown by our work and that of other researchers, but we also found that formula is not the food that replaces breast milk. Rather, for the poor, it is sweetened condensed milk and evaporated milk; only middle- and upper-income groups consume formula (Fernandez and Popkin 1988; Zohoori et al. 1993). Theformula companies are the cause of the decline in breast-fpediq. We and many other social science and public health researchers have examined the causes of breast-feeding behavior. The Cebu project provided some of the more definitive results, showing that a wide range of health professional, household, and community factors affect infant feeding intentions and decisions. In contrast, the role of the infant formula industry in this is limited. However, we also found that the formula industry has a significant adverse effect on the length of breast-feeding and the length of exclusive feeding. Our research could not explain fully the reason for these effects but the statistical controls used lead us to believe that it is a real industry promotion effect (Adair, Popkin, Guilkey 1993; Guilkey and Stewart 1995; Stewart et al. 1991). The decline in breast-feeding leads to a huge increase in infant mortalig. The Cebu Study Teani has shown that breast-feeding has significant effects in reducing infant growth retardation and morbidity (both diarrheal and febrile respiratory), but that other factors are equally or more important (Adair, Popkin, VanDerslice et al. 1993; Cebu Study Team 1991; Cebu Study Team 1992). These include, inter aha, the water supply, sanitary conditions inside and around the household, and orher environmental factors (VanDerslice and Briscoe 1993; VanDerslice and Briscoe 1995; VanDerslice et al. 1994).

TRUTH-IN-FUNDING

121

The benefits of breast-feeding are so great that any decline has major consequences for the infdnt and the mother. We have shown significant effects of cessation of breast-feeding on increased fecundity and reduced length of postpartum amenorrhea, but we have also shown that the lower-income women who breast-feed for long durations- from one to two years-can suffer significant depletion of their body fat (Adair and Popkin 1992; Popkin et al. 1993). In these and other ways, science and research lead to complex results that muddy the waters of public debate. Industry certainly gains from supporting such studies, even when it cannot control the research.

Notes I. Gail Henderson and Nancy King provided invaluable advice in the preparation of this case. In particular, Gail Henderson deserves thanks for leading me through all phases of this write-up. Frances Dancy and Lynn Igoe are thanked for assistance in preparation and editing. 2. John Akin and David K. Guilkey, both professors of economics and fellows of the Carolina Population Center at UNC-CH, shared all the effort, travails, and rewards of this project. Many other collaborators were involved in this research project but not in the events related to its initiation. 3. We subsequently worked out the exact wording of this acknowledgment with our collaborators in the country we selected for this work, the Philippines. The wording for the generic component of the acknowledgment is as follows:

This study is part of a collaborative research project by the Nutrition Center of the Philippines, directed by Dr. Florentino S. Solon; the Office of Population Studies (OPS), University of San Carlos, directed by Dr. Wilhelm Flieger; and a group from the Carolina Population Center, University of North Carolina at Chapel Hill (UNC-CH). Barry M. Popkin (UNC-CH) is coordinating the total project. Funding for the project design, data collection, and data base development was provided by the National Institutes of Health (NIH) (Grants R O I - H D I ~ ~ ROI-HD18880, ~~A, and ROI-HDz318z), the Nestle Coordinating Center for Nutrition Research, Wyeth International, the Ford Foundation, the U.S. National Academy of Science, and the U S . Agency for International Development (AID). Funds for analysis were provided by N I H Grant. . . . We thank. . . for their extensive assistance. 4. A full set of references for these studies is presented elsewhere. Interested readers can obtain free use of the original CLHNS data set by accessing the CLHNS's home page. The project continues with a number of follow-up studies of the longer-term consequences of the early environment on the child and the mother. The home page can be found at http:/lwww.cpc.unc.edu/projects/cebu/cebu~home.html. 5 . I have since received some support for my U.S. research from the Kellogg Company and other food companies. None has approached the size and scope of the original gifts,

122

BARRY M. POPKIN

but all companies have been willing to agree to broad terms of reference and allow us to receive gifts that do not require close financial and research accountability. 6. In each case, the organizations reviewed very carefully the results of our research and strong suggestions were made regarding changes in the content and conclusions. In our experience, however, none of these suggestions was inappropriate and related to a political agenda.

124 LORETTA M. KOPELMAN

bias I mean unwarranted influences or one-sided deliberations that interfere with careful reasoning (Durbin 1988; Kopelman 1994; Paul and Rudinow 1988). Good scientists take elaborate measures to protect results from bias because, whether conscious or unconscious, it threatens disinterested study. Another way to protecr science from conflicts of interest rakes the form of social policies and practices. For example, many journals, state employers, and universities require investigators fully to reveal conflicts of interest. Conflicts of interest are conditions where a potential for bias exists, such as ties to industry. There is reason for concern. For example, one investigation comparing biotechnical academics found that those having ties to industry were four times as likely to delay reporting or publishing their studies (see De Ville 1995 and Blumenthal et al. 1986). This difference was attributed to attempts to protect the industries' proprietary interests. Part of the problem is that science, business, and advocacy groups have different standards. To understand these differences and investigators' duties, we have to look at the values held in these cultures.

The Scientific Framework Popkin found that Nestlt was "open" and "unfettered" and "not restrictive" in supporting his studies on breast milk substitution and breast-feeding behavior. Such attitudes, which he finds typical of industry support he has received, suggest some important goals and values governing science-honesty, objectivity, and openness (Cournand 1977; Kopelman 1995).' In contrast, he found that some advocacy groups placed troubling restrictions on him that seemed to threaten these primary values. UNICEF and Wellstart were advocacy groups that supported Popkin's research but only if their names did not appear with an industry criticized for promoting substitutes for breast milk in the world. This lack of openness was troubling to Popkin because he wanted to reveal all sources of funding for his projects. Thus, measured in terms of the two ways to protect science from conflicts of interest mentioned above-the investigators integrity and social policy-Popkin concluded that the industry funding agencies fared better than the advocacy groups. By refusing to have their names associated with Nestlt, advocacy groups seemed to Popkin to threaten the important practice of revealing all funding sources as a way for investigators to protect research from bias (Durbin 1988; Kopelman 1994).

The Framework Assumptions of Advocacy Unlike scientists seeking knowledge, advocates sometimes appropriately adopt special or biased perspectives. Effective advocates for children's health focus on

BIAS AND CONFLICTS OF INTEREST I N SCIENCE

125

children, despite the existence of other good causes. Advocacy groups are not committed to strict impartiality in how they present themselves or their cases. Scientists who adopt some of the advocates' goals may alter how they select, record, interpret, or publish data. Socially established boundaries regarding advocacy exist. Advocates should not lie, cheat, or ask scientists to rig their studies to help their campaigns. Arguably, Wellstart and UNICEF did not overstep any socially established boundaries in insisting that their names not be associated with Nestle. They did, however, challenge a principle Popkin and others in the scientific community regard as important: full disclosure of funding sources. The advocates' and scientists' frameworks and goals were different. If they could not resolve their disputes to their mutual satisfaction, then perhaps Popkin should have refused either their funding or that from Nestle, so he could list all funding sources.

The Framework Assumptions of Business Businesses primarily seek profits. They are unlike advocates, who aim primarily at helping or avoiding harm, or academics, who seek truth and knowledge. The framework goals and assumptions of business activities differ from those of both advocacy groups and academic dis~iplines.~ Thus, when industry funds research, prudent people are on guard for proprietary interests. What was the advantage for Nestle to fund Popkin's studies? This beleaguered company might have been trying to improve its image through apparently disinterested research funding. If the studies supported their practices they might reply to critics, "You see we were right all along"; if the studies did not they could reply, "Now we have the data and will change our ways. We had no idea." They might also try to "buy science" to show that their critics were hasty and that more study was needed, since the relationship between breast milk substitutes and infant morbidity and mortality is complex. Prudence, long-term planning, or future profits sometimes dictate that companies act like very good citizens advocating for noble causes. In business dealings one ordinarily assumes that competent adults can freely fashion their own arrangements for themselves and their dependents and dispose of their fairly obtained property as they wish. Entrepreneurs presuppose that competent people should bear responsibility for their own choices. The business community assumes that competent people, even those in cultures different from our own, can assess whether or not they want to use their products, including breast milk substitutes. Imagine a poor woman who needs to work outside her home. With supplemental breast milk substitute feedings or early weaning she can keep the job she

126

LORETTA M. KOPELMAN

needs desperately. She might view the advocates' pressure not to allow companies to advertise their milk substitute products as paternalistic and insulting to her intelligence as a rational person. In fact, many women's groups resisted restrictions on their choices, and resented advocacy groups that did not want breast milk substitutes readily available to women. Why should advocacy groups limit their choices "for their own good," if they are competent and informed adults? Advocacy groups believe that they are sometimes needed to show that companies are unethical and unfair. If Popkin is correct, Nestle was, as he said, "identified as the major culprit" in an industry "accused of unethically promoting their product." Competition does not always reveal misinformation, but advocates often can. Freedom to choose means little if consumers are manipulated by biased information. In order for free markets to work in a just society, people need good information. Typically, government must oversee, regulate, or impose controls so people can assess what they ~urchase,including ensuring that people get accurate information. In addition, a just society also seeks fairly to promote children's well-being and opportunities to become self-fulfilled persons. Advocates argued that Nestle and other companies used unethical practices in advertising breast milk substitutes in certain developing countries, thereby causing greater morbidity and mortality among infants. These companies therefore overstepped what was tolerable within the framework of a business ethic. What must an industry identified as culpable in placing profit ahead of truthfulness, social needs, and children's health and well-being do to redeem itself in the eyes of the community? Are industry's gifts forever tainted, or can it do some penance by changing leadership and donating money to good causes-such as funding research? Popkin believes that with suitable safeguards companies can be rehabilitated and investigators can take their money without compromising their integrity or standards.

Multiple Frameworks Science, advocacy, and business activities have different framework beliefs and values guiding practices. Scientists try to avoid partiality, but advocates and businesses do not. Advocates try to protect people from foolish choices, but business people do not. The distinctive framework values and assumptions that mold our different activities overlap as well as conflict in certain ways. For example, honesty is valued in science, advocacy, and business relationships, but for different reasons. Because these activities usually differ and sometimes conflict, it is crucial to keep straight which framework one is using. The same person can act as an advocate, a scientist, or a business person, and our judgment about what they ought to do

BIAS AND CONFLICTS OF INTEREST IN SCIENCE

127

is affected by what framework they adopt. For example, scientists are not supposed to advocate for hypotheses, but they can advocate for thrir departments. Let us consider Popkin's case with respect to the two important ways cited to counteract conflicts of interest: social policies and the investigators' integrity. All agree that Nestle once exceeded what was socially permissible in promoting its product. In accepting Nestle's funds, however, Popkin took measures to ensure that business interests did not affect the design or outcome of his studies. Perhaps advocacy groups like Wellstart and UNICEF should have seen that this was sufficient and should have agreed to have their names associated with research funded by this company. They may, however, fimd whom they wish and stipulate conditions for funding Similarly, investigators and institutions may also set the conditions under which they take funds. From what was presented, no evidence exists that anyone overstepped the socially established minimal standard of what was permissible in supporting or conducting research. The second protection concerns investigators' integrity. Integrity is judged in part by investigators' sense of honor at being asked to do certain things. Researchers' integrity, and the culture that supports and sustains it, may be the best protection from bias and conflict of interest, including commercial interests. Full disclosure of funding sources is onc important part of the response to the now capital-intensive nature of scientific research. Scientists are increasingly seeking support from advocacy and commercial interests, and must carefully protect science's and scientists' primary interests of openness, objectivity, and honesty. Popkin took the money from Wellstart and UNICEF, bowing to their demands, although he firmly believed that full disclosure was necessary to protect research from bias. Should Popkin have accepted those restrictions in order to obtain funding for his research?Arguably not, since he fully supports the scientific community's rule to acknowledge all funding sources and believes it an important moral principle. The investigator's sense of integrity is one of the most important protections against bias.

Notes They also govern other academic modes of inquiry. Of course, many people in business are honest, pursue truth, and are concerned citizens who advocate for good causes. Truthklness is generally more prudent for business in the long run than dishonesty. One could, however, conduct business in a dishonest way; in contrast, one could not conduct science dishonestly by manipulating data and still be doing science. I.

2.

CONTEXT AND COMMUNITY

129

Contexts There are several immediate contexts that may help us to more fully understand the dilemmas inherent in the case. It has become increasingly difficult to find external support for universitybased research. NIH support has, in the last decade, become increasingly difficult to secure. This also is true of foundation funding and other public and private sources of research support. The difficulty of obtaining support for important and expensive research has encouraged scholars like Popkin to seek new funding sources. 2. Corporate funding for research offers one possibility for funding in this environment. As a result of increased corporate hnding for researchespecially researchers holding equity interests in companies sponsoring research-interest in the problem of conflict of interest has grown within research ethics. Further, many universities, and journals, as welt as the federal government, have promulgated rules and regulations governing conflicts of interest. 3. The specific company sponsoring Popkin's research in this case was the focus of a major ethical and political controversy over the sale and promotion of infant formula in the period preceding and during the time the gift was given to Popkin. Popkin's case seems to suggest that the controversy was resolved by the time he started his research, but in fact Nestle continued to be criticized for its infant formula promotion policies throughout the 1980s and early 1990s ("Nestl&' 1989; Sethi 1994).

I.

Interests Confronted with the opportunity to obtain funding from Nestle for a major project on infant mortality, Popkin attempted to deal with a series of potential ethical dilemmas and to reduce possible conflicts of interest. First, he established an advisory panel to assure the scientific integrity of the work. Implicit was the recognition that some peers might inevitably regard work sponsored by Nestle as inappropriately influenced by the funding source. Popkin further sought to isolate Nest16 from the research itself by arranging for the funding to come to the university in the form of a gift. This served to protect the university's reputation and also further assured that the company would not be in a position to shape the research or its findings. Finally, Popkin agreed to acknowledge and clearly disclose Nestle fimding in all publications that arose from the research. 'l'his final principle proved difficult to sustain in practice; according to the case, both Wellstart

130 ALLAN M. BRANDT AND LARA FREIDENFELDS

and UNICEF requested that Popkin delete mention of Nestle support as a quid pro quo for providing funding for related projccts. In these instances, Popkin apparently acceded to their demands. Poplun offers a number of compelling reasons for accepting the support of Nestlt. First, as indicated above, he was convinced that the funding would in no way influence the course of his research or its analysis. Second, conventional funding support is increasingly difficult to obtain and the standards of peer review can be intellectually constraining; further, funds from conventional sources tend to have a number of cumbersome restrictions on their use. He suggests that the unfettered nature of corporate funding may offer opportunities for more interdisciplinary work and allow the kind of "exceptionally complex research design7'that the NIH or other traditional peer-review funders might regard with skepticism. No doubt, there may be some advantages to corporate funding of research, but it is important that the rationale for such funding not become rationalization. Popkin's analysis would be enhanced by more direct inquiry into Nestlt's interests. What does Nestle achieve by funding such research?What is their interest in sponsoring this work and other research like it? While we should acknowledge the notion of corporate altruism, it seems likely that it was Nestle's badly tarnished image that motivated their desire to support university-based infant feeding research. The fact that Nestlt may be using corporate sponsorship of high quality research to enhance its public image and legitimacy is not necessarily a reason to reject their support. In a related example, many university researchers have taken considerable funding from the Tobacco Industry Research Council since its establishment in 1955; typically those who have received such funds have asserted that the industry has not influenced the specific nature of their work. Others, who have rejected industry funding, have argued that they want no part in this legitimating function. This suggests that it might be useful to develop a fuller understanding of any industry's motives and goals in sponsorship. Only then may researchers be in a position to evaluate their role in this legitimating process. In its sponsorship of Popkin's scientificallye~em~larywork, Nestle "purchased goodwill (within a community in which its motives had been suspect); credibility (where it had possessed little); and social legitimation (important to its longterm corporate identity and interest). Sponsoring research was part of a complex marketing and public relations strategy which also included establishingthe Nestle Coordination Center for Nutrition, Inc., and directly contacting church leaders in the United States to gain their support. At the same time, Nestle continued its aggressive and controversial infant formula promotion techniques in the United States and abroad (Sethi 1994). Poplun participated in this legitimating process. We want to emphasize that there is no value judgment here-only that it seems useful, if one is participating in such a process, that it be clearly understood at the

CONTEXT AND COMMUNITY 131

outset and that such an understanding be part of one's evaluation of the pros and cons of participation. In sponsoring Popkin's research, Nestlt was "purchasing" not only social legitimation, but also a set of scientific results, and with them the authority that science carries in social debate. Popkin points out that some of the conclusions from his study probably did not please Nestlt and could be regarded as damaging to the infant food industry. He points this out in order to prove that his research was not influenced by Nestle, and that there are ways to accept money from industry while preventing a company from simply buying favorable results. However, what Popkin does not acknowledge is that Nest16 might still regard his conclusions as valuable, because they make the explanation of infant mortality in lower-income countries more complex, and therefore potentially more confusing. The tobacco industry has been using this strategy ofconfusion for a long time (Hilts 1996). The Tobacco Industry Research Council funds research showing that lung cancer has many potential causes, and that many diseases that might be regarded as smokingrelated in fact frequently have multiple and interrelated causes. Even if the tobacco industry cannot plausibly deny the harms of smoking, it can create confusion and controversy around the diseases that are frequently smoking-related. Nestle can promote this confusion even more plausibly than the tobacco industry can, because infant formula is actually beneficial in many cases. Some of the specific points in Popkin's study may appear threatening to Nestle, but the overall effect of the study is to create a sense of complexity around infant morrality that NestlC can use to its advantage. This is not to denigrate the value of the study; the point is that Nestle may find it useful in ways that the researchers did not expect. In the midst of a sharp sociopolitical controversy, funding elite and complex scientific inquiry no doubt serves Nestle's corporate interest no matter the specific outcome of the research. Popkin says that he was concerned about his colleagues' response to his study. He was pleasantly surprised not to be censured by other scientists for accepting money from industry and proceeding with a study that had not received peer review and approval through the standard channels of grant applications. While Popkin acknowledges the value of peer review, he says the project that he developed after being approached by NestlC was an "exceptionally complex research design" that "few colleagues believed we could carry. . . off ";this is the reason he gives for not applying for N I H funding. Perhaps Nestlt gave the group funding for this project because the company felt that it was important to support "visionary" approaches, which the conservative peer-review system frequently does not support. But it is more likely that Nestle did not need the study to be peerreviewed in order to be useful; in fact, a study that would be considered too complex by the peer-review system might be even more usehl in sowing confusion about infant formula as one of many interrelated causes of infant mortality From

132 ALLAN M. BRANDT AND LARA FREIDENFELDS

sponsoring Popkin's research, Nestlt hoped to gain both social legitimation and the ability to use the authority of science to obscure the debate.

Policy Finally, and perhaps most importantly, it seems appropriate to speculate on the policy-level effects of industry funding of university-based research. Popkin begins his analysis by suggesting that such unfettered funding possesses practical and scientific advantages. He advances the argument that the bureaucratic nature of federal funding programs may inhibit innovative research. A fuller discussion of such advantages, as well as potential disadvantages, would be useful. For example, although peer review may be constraining, is private, non-peer-reviewed research the liberating alternative? What could be the long range implications of increasing the percentage of corporate funding for infant mortality research?Consider the following scenario: Encouraged by their success in giving a substantial gift to infant mortality researchers and being cited as a source of funding in all their papers, Nestlt expands such gifts to most infant mortality research teams. Given the success of Popkin's experience, these groups accept. Soon, other infant food manufacturers begin such funding programs as well, feeling the need to compete with Nestlt for social approbation. Infant mortality researchers in universities throughout the country and abroad celebrate this windfall and expand their programs. Industry funding appears to be a boon to the field and has set new standards of corporate social responsibility. As industry funding expands, however, traditional public and private funding begins to dwindle. Foundations look for other areas in which to have a "needed impact." Their grants appear unimpressive in comparison to industry gifts. Federal funding is radically reduced as budget-cutting senators note the success of replacing tax dollars with corporate funded donations for such research. As corporations dominate the infant mortality research area, they now have the opportunity to shape its agenda, shifting from gifts to grants. Researchers resist such pressures, but may fear that their funding might be jeopardized if their objections are too aggressive; further, voicing publicly their objections to industry influence could compromise the perceived integrity of their work. The result is that much excellent work comes into question and the field of infant mortality research is in disarray. There are, of course, many other scenarios. The one we have outlined, however, is quite realistic. Again, a comparison with the tobacco industry is appropriate.

CONTEXT AND COMMUNITY

133

Though the tobacco industry has not managed to influence funding for medical research to this extreme degree, it has had a great impact in other areas through its support of sports, such as women's tennis and car racing, and arts organizations. For example, in New York in 1994, Philip Morris persuaded many of the arts organizations it supports to petition City Council on the company's behalf. Philip Morris threatened to move its headquarters when City Council was considering strong antismoking legislation, and asked arts organizations to let City Council know how much they relied on Philip Morris's financial support. According to the New York Times,"In the arts world, offending Philip Morris is the equivalent of crossing the National Endowment for the Arts" (Goldberger 1994). Philip Morris has gained a great deal of power over the arts organizations it supports, as city and state funding has withered and the NEA's budget has been cut repeatedly. Though researchers and legislators would presumably guard better against industry influence of medical research than of sports and arts, this example demonstrates the potential power and influence of industry funding. Individual decisions abour the appropriateness of industry research may together have significant policy level implications. The ethics of conflict of interest requires deep investigation of the full range of interests involved in a given case, as well as their long-term implications. Such potential conflicts are aggregative and therefore cannot be analyzed in the context of a single case. The typical approach-a single decision that meets certain "ethical" criteria-may in the aggregate lead tu significant problems. What if all infant mortality researchers followed Popkin's lead? Our concern about conflicts of interest reflects the notion that individual action can ultimately lead to the erosion of integrity and trust . upon which the community of research relies. It is for this reason that perceptions of conflicts are so crucial; most regulatory approaches to conflicts of interest therefore focus on perception rather than on any particular assumptions about individual integrity, bias, andlor fraud.

Conclusions No doubt corporations are in an important position to foster a wide range of socially useful research. As Popkin suggests, such sponsorship holds many opportunities as well as potential costs for the research enterprise. This case suggests that mechanisms that insulate such research from its sponsors, as well as full disclosure of funding sources, offer possibilities for maintaining the integrity of such research in the face of powerful interests. Nonetheless, only a thorough evaluation of the contexts, interests, and possible implications for public policy will make it possible to adequately analyze the complex questions inherent in corporate funding of re-

134

ALLAN M. BRANDT AND LARA FREIDENFELDS

search. It seems inadequate to argue-as Popkin does-that his work is justified by its quality and independence. Researchers must evaluate the full social implications of their funding arrangements in a more systematic manner than demonstrated here. Popkin's decision to accept corporate funds has significance not just for the integrity of his own research, but for the legitimacy and integrity of the broader research community in which he works.

positions have been hotly debated.' This is therefore a particularly important case, not only because it raises the question of the universality of the regulations, but also because of the multidimensionality of the arguments and commentaries that arise from it. Our two commentators, American social scientists with considerable international research experience, offer different cognitive maps for organizing the ethical issues raised by the case. Cynthia Waszak, a psychologist and family planning expert, presents as her framework the classical ethical principles of autonomy and beneficence. Weighing the risks and the benefits, she stresses the importance of informed consent in protecting subjects against the harms that so worry the researchers. Barbara Entwisle, a sociologist and demogapher, is clearly drawing from the same traditions, but she enjoins readers to step back and consider the broader range of "social contexts" that characterize research-which she rightly describes as a social process. Because research contexts vary, the meaning of a behavior under study is subject to interpretation within the local community. Both commentaries agree that the major ethical dilemma of this case lies in the lack of safeguards to protect the interests of potential subjects from the possibility of abuse of power by the researchers. Entwisle argues that the interests of the researcher-as one of the social actors-cannot and should not be excluded from the calculations of risk and benefit, and concludes that without external review, it is difficult to protect parties from inevitable conflicts of interest. Thus, although she calls attention to the complications of social context, in the end it is those very complexities that create the need, in her view, for a third-party review. Balin and Ramos take a different approach. For them, the definition of interest explicitly includes their commitment to social action agendas. Protection of the interests of subjects is seen by them in the context of broader social responsibilities to poor women. In addition, they consider their conduct as researchers to be regulated by a "higher code" of ethics that is derived from the ongoing nature of their relationships with research subjects. As Balin stated at the conference, "The end of the research is not the end of the research." He described himself as involved in projects with no clearly identifiable end-projects that depend upon social relationships built over time. Both commentators locate the source of the protection of subjects in an external review of risks and benefits. For Balin and Ramos, however, the development of this trusting relationship is a morally superior source of protection for subjects, and the regulations may actually interfere with this process. A final issue involves the scientific merit of the study. For Entwisle, the validity of evidence obtained from this exploratory study is a critical determinant of the ethics of the research. But as she also notes, commentaries are constructed within a local context. Hers is the American scientific community. For Balin and Ramos,

INTRODUCTION

139

the frame of "local knowledge" includes n o t only science, but also the meaning given to these nineteen stories of women n o t well served by the Argentine health care system. This political value is part o f the equation.

Note I. Macklin argues in her keynote essay that the principles of autonomy, beneficence, and justice are inviolate and universal, and that all research with human subjects should adhere to these basic values. This position was strongly supported by an editorial in the prestigious New EngLrndJournalofMedicine (Angel11988). O n the other hand, many field researchers-including those who have conscientiously attempted to apply those principleb in a variety of settings-have encountered such difficulty that they question their universality, and instead call for application that takes account of local context (Christakis 1992; Newton 1990). More recently, voices of and for the "locals," from Akican social scientists (Ijsselmuiden and Faden 1992) to Chinese human rights activists (Li 1996). have argued that the call for particularistic interpretation of universal rights is, in itself, a form of oppression.

INDUCED ABORTION IN ARGENTINA

141

Mundial de la Salud/Organizaci6n Panamericana de la Salud 1990). Recent studies have shown that deaths due to complications of induced abortions make up approximately forty percent of all maternal mortality Furthermore, it appears that many of these deaths might have been prevented by providing proper medical attention in hospital settings. Because current legislation restricts abortion to cases where the life of the mother is at risk or pregnancy has resulted from rape of a mentally deficient woman, the large majority of abortions are clandestine. They are therefore more costly and less safe than they would otherwise be. Their high cost increases the probability of complications, since low-income women get cheaper and less safe services (for example, with untrained professionals, in septic environments). The increased incidence of complications in turn imposes high costs not only upon patients but upon available medical services. Post-abortion complications are treated in public hospitals, which are required by law to assist any emergency case. Private facilities usually refuse to admit any patient with an ongoing abortion, whether induced or not. Around forty percent of gynecological in-patient beds in a typical general hospital are currently occupied by such cases. The patients tend to be low-income women who request medical attention with signs of incomplete abortions, such as heavy vaginal bleeding or post-abortion sepsis. Previous studies indicate that delay in seeking andlor obtaining appropriate medical attention is a major factor associated with maternal mortality due to postabortion complications (Direcci6n de Maternidad e Infancia 1992). A principal objective of the current study was to explore whether the fear of indictment under current laws produces delays, both delays in the woman's seeking medical attention and delays in initiating treatment in the hospital after the woman has arrived at the emergency ward. The former problem might be considered trivial as a research question, since obviously women will not be eager to call attention to their illegal, and often severely chastised, behavior. The latter, however, is of greater interest since hospital personnel normally seek to ascertain as soon as possible whether an abortion was induced, and if so, how, in order to follow the prescribed medical procedures. Nurses and doctors are instructed to ask women immediately about induction, before checking for other indicators (for example, through uterus inspection or checking body temperature for possible infection). Patients, however, tend to deny insistently, due to fear of prosecution (although prosecution is very unlikely) andlor due to fear of being denied medical attention (such as curettage). The negotiation process, according to our hypothesis, might be a factor in delaying proper medical attention: women may be sent back home, or curettage may be delayed, until more clear and dangerous signs of an ongoing abortion appear. In order to investigate these issues, we designed an exploratory study of postabortion behavior to learn about (I) how women seek medical assistance for con-

N SILVINA RAMOS 142 JORGE B A ~ AND

sequences of abortion; (2) what role the fear of prosecution and lack of information play in delaying the search for appropriate help; and (3) how medical assistance after abortion is managed and provided in public hospital settings. We chose to work in a hospital ward where our acquaintance with the director and ongoing professional relations with the medical staff made it possible. We did not consider alternative study designs, since we did not want to rely on large samples, and it was not advisable to focus entirely on past events. Only a hospital setting provided us with easy access to women who were experiencing serious physical consequences of induced abortion. This design, however, confronted us with a host of ethical issues, which are the subject of this paper.

Study Design Our study involved, first, systematic observations at the emergency room and hospital wards, supplemented by unstructured interviews with both medical personnel and patients on the spot; and second, the use of open-ended questionnaires applied to a small, nonrandom sample of women who were admitted with a presumed diagnosis of "induced abortion," whether they had recognized it as such or not. Nineteen women were selected for the study while still in the hospital, and were interviewed both in this setting and a month after they returned home. Women granted permission for both of these interviews, but were not explicitly told about the focus of our research. Doctors and nurses knew that we were studying induced abortion, and that we had hospital authorization to do so, but we did not discuss with them the issue of delays. Before actually starting the fieldwork we held two seminars with the medical personnel at the ward, talking about what we, as social scientists, knew about fertility behavior, contraception, and abortion in Buenos Aires. We did not engage in a specific discussion about our research design and major hypothesis, but we did explain the overall nature of the study we wanted to carry out. The hospital does not have a review policy regarding social science research, or an institutional review board. Permission was requested from the director of the gynecological ward, who asked the rest to "collaborate," and professionals were informed as a group of the scope of our study. Printed forms to obtain informed consent are not regularly used in this setting, and we did not use them with either professionals or patients.

The Interview Process Interviewers spent long hours, at least twice a week, at the gynecological ward. They approached possible interviewees after being informed by medical person-

INDUCED ABORTION IN ARGENTINA 143

nel that they were admitted either for an ongoing abortion that was completed at the hospital or for the presence of hemorrhage and other symptoms of an induced abortion. Only two potential interviewees admitted having induced an abortion to medical personnel. Denial is normally a consequence of fear of prosecution. Although nobody at the hospital actually knew ofwomen who had been prosecuted, the physicians are supposed to file a report to the police in any verified case of induced abortionwhich some do, and others do not-and this in itself is acircumstance any woman would try to avoid. At least eleven of the women who received requests from "university students" to "talk about their hospital experiencefor a study" said no. Since the request to talk about their experience came after an initial chat, there were other, uncounted cases ofwomen whose unwillingness to be interviewed became evident to the interviewers even before asking the question, and were therefore not considered to be possible cases. Thosewho accepted received very little initial information, and only at the end of the first interview were they asked for permission to be visited at home some time later. Home visits took place between one and three months after discharge from the hospital. The interviewers were always the same young women who had approachcd thcm while they wcrc still hospitalized. There were no outright refusals at home, but home interviews varied considerably in length and depth according to the willingness of the women to explore the issues posed by interviewers. Interviewers also had a chance to see the medical histories and to talk to medical personnel about the cases. We did not carry out formal interviews with medical personnel; thus we did not get explicit oral informed consent or refusals from those with whom we interacted. Clearly, the collaboration of medical personnel varied considerably. An important, unanticipated topic of our research, however, which was not explicitly discussed with professionals when seeking their collaboration, was the nature of hospital rules and regulations regarding procedures followed in cases of induced abortion. We were led to believe, during the preliminary stages of our project, that such rules were few, simple, and straightforward. A major one was the following: "All cases of induced abortion have to be reported to the authorities." In fact, we found that only a minority of cases is actually reported, and "confessed inductions are to be found in both the reported and unreported categories. Reporting seems to be selective and related to high risk of death. However, another rule was followed in all cases: patients were always asked, often several times, whether they had induced an abortion or not (under the implicit assumption that all of them had actually done so and were seeking assistance to complete the curettage procedure). The rationale for this rule is derived from the presumed differences in medical procedures to be followed (for example, use of antibiotics, whether there are signs

144 JORGE BALAN AND SlLVlNA RAMOS

of an infection or not) when an abortion has been induced. Curettage, according to another unwritten rule, is avoided whenever possible (i.e., when an abortion is not yet in course). Clearly, the policy is to deny abortion upon demand. Women immediately feel under pressure to "confess" an abortion as self-induced, and they tend to deny this. Fear of legal indictment, and of denial of treatment, are major reasons. To ascertain this fact, however, we had to proceed in much the same fashion as medical doctors and nurses did: We asked the women what had happened to them. It was unreasonable to expect that women would dissociate us from medical professionals when interviewed in hospital settings, in spite of our explicit explanation that we were not associated with the hospital. Thus, although we probed this issue while at the hospital, we asked for their permission to interview them later, at their homes, where we might expect better results. In only a few cases were "true" inductions reported to us this way. Also, in at least one case we were told by the interviewee that she had lied to the medical personnel when admitting induction, since she believed this was what they wanted to hear.

Dissemination of Study Findings Based upon our study of nineteen women, we found evidence to support our hypothesis that women's fears of prosecution did, in fact, lead to delays in seeking care. An additional finding was that some delays occurred in the hospital as a result of the process of questioning women and diagnosing their conditions. We observed many cases in which protracted negotiation played a role in the delays. Many months after fieldwork was concluded, but before filing our final report to the granting agency, we presented a verbal report of our major results in a followup seminar for the hospital medical personnel. Later on we did many public presentations in other settings, not allowing the particular hospital to be identified. Professional collaboration with the ward head continued, including a different study that investigated why some women with positive Pap test results do not return for follow-up care. However, this is the only fieldwork we actually conducted in this hospital setting. Although the medical profession is a target audience for this and other studies we have conducted in Argentina, we expect the results to reach a variety of audiences through different means. In this study, we looked for indications that illegality leads to risky induced abortions and also to poor treatment of abortion consequences. We feel that advocacy groups, composed mainly ofwomen, can make good use of these results in trying to influence public opinion and changing the policy environment. We did not feel that coming back to the specific women we interviewed was necessary for their own sake, and since the women who were in-

INDUCED ABORTION IN ARGENTINA

145

terviewed for this study came from a wide geographical area, there was no local "comn~unity"of women to which we specifically wanted to report.

Ethics and Ethical Principles Our study poses a host of ethical problems that appear whenever we study illicit behavior. We could not tell the women about our focus, since to do so would warn them that what they said could put them at risk, and would heighten their concerns about confidentiality, thus potentially making them less willing to speak freely. The project in itself entailed considerable invasion of privacy (we had access to the women's medical records before requesting their permission) of a particularly vulnerable population (low-income women who were involved in illicit, culturally rejected behavior). It is clear to us, however, that such problems cannot be discussed without reference to the local context and to the problem under study. We assured both professionals and patients that our records were confidential. The former did not havc much to fear, evcn if thcir professional behavior was at stake, since they were better informed about our role as researchers. However, scrutiny of their behavior-which could put them at risk as well-became an ethical issue because it was not spelled out to them in the beginning, but only became important in the process of the study. The patients, however, had good reasons for fear, and perhaps preferred to lie to us, as they lie to the doctors about these matters. But do we actually have a right to keep probing on these issues under the assumption that they face no real risk? Is not fear in itself some kind of harm? Is this harm compensated by our ultimate goal of providing a rationale, and a method, for shortening delays in medical attention and thus decreasing the risks for other patients in similar circumstances? Social science research in Argentina is only recently adopting explicit procedures about ethical conduct, and mostly in response to outside funders who request them. Within the field of reproductive health, a major influence in this regard has come from the World Health Organization's program. However, the major procedural element has been informed consent, usually gathered through a written formula. This particular form of informed consent has been attacked locally: most researchers find that low-income women will have dificulties understanding the text and interpret incorrectly the function of a signed statement. In our particular case, we are not aware of any ethical review procedure on the part of the Population Council's program that provided us with funding. The hospital does not have a review board for social science research, and in any case the study was conducted through an informal arrangement with a specific ward. We do not believe the hospital administration ever learned about our work.

N SlLVlNA RAMOS 146 JORGE B A ~ AND

We were not particularly concerned, from an ethical point of view, about using medical records without the knowledge of the women, nor did we feel that women were coerced to collaborate due to their vulnerable status. Women, however, are coerced to incriminate themselves by the medical personnel and administration, and we certainly did as much as we could to avoid doing the same. Nevertheless, it was clear that we often ended up asking very similar questions, and in one way or another assuming that they had lied to the doctors and might lie again to us. Coercion is always based on some kind of power, moral or material. Doctors have such power, because they can refuse to comply with patients' requests, but we did not. Although interviewers tried to be friendly and helpful, this was not in any way misconstrued by women subjects to be in exchange for compliance with requests to collaborate. We did, however, often feel that there was tension inherent in the task of repeating questions that were already weighted with many different concerns about truth-telling. Another source of concern, of a very different nature, was related to the possible harm to women in our research procedures. After all, these women were undergoing difficult times. Perhaps many of them were eager to hide what had happened to them from family, friends, and acquaintances, not just from the police. Why should they trust us? And what might be the negative consequences of their trusting us, even if we deserved to be trusted and kept strictly confidential anything we learned about them? Certainly, we could follow some procedures to preserve the confidentiality of our records, and to avoid contact with friends or relatives that might be embarrassing for the women. But how could we avoid creating fears in our subjects, or the potentially negative effects of recalling recent or past events? Our overall justification, of course, is that we could do some !good to the general class of poor women, since our research results could be used to foster their rights. But this justification was only satisfactorywhen we actually were convinced that talking in itself was seldom harmful, and that our women subjects often expressed positive feelings about sharing their experiences with young, educated, caring women.

148 CYNTHIA WASZAK

harms they face because of their clandestine abortion is an empirical question, but a number of safeguards are available to minimize the risks that exist.

Informed Consent Informed consent is the cornerstone of ethical conduct in research. The two key elements are adequate information and choice. It is clear that permission or consent was obtained by women in the study, but it is unclear what participants thought they were giving consent to and the degree to which they thought they truly could refuse to participate given their highly vulnerable situation. The fact that interviewers documented at least eleven cases in which women refused to participate provides greater confidence that study participants did perceive that they had a choice, however. Information Balin and Ramos state that researchers could not tell women the focus of the study because "to do so would warn them that what they said could put them at risk, and would heighten their concerns about confidentiality, thus potentially making them less willing to speak freely." This assumption bears reconsideration, however, since it is not clear why women would speak less freely, with assurances of confidentiality if they were told the focus of the study than if they were given less information. Study participants form their own personal hypotheses about the reason for a study, which affects the way in which they choose to answer questions. Clues to this include (I) the reason for their admission to the hospital and (2) the questions that are asked of them. Though unrelated to the study, this hypothesis generation is precisely what occurred with the woman who falsely incriminated herself, believing that was what hospital personnel wanted to hear. It seems better that all women be given enough information about the true purpose of the study so that they may answer questions more realistically. While the investigators did not actually deceive participants by telling them something false, they did not fill in all the details, either. More information could have provided the participants with a better basis for making an informed choice they felt comfortable with, and might have reduced stress or conflict at later points of the study. Choice The potential for the participant to perceive a lack of choice can result from the vulnerable position she finds herself in because of her abortion and her status with regard to hospital staff, their relationship to legal authorities, and her de-

INDUCED ABORTION IN ARGENTINA

149

pendence on them in a potentially life-threatening situation. Researchers cannot entirely eliminate this problem-especially when the behavior under question is an illegal one. Though it is difficult to convey, researchers must emphasizeand demonstrate when possible- that there will be no negative consequences for nonparticipation. This is a particular problem in countries where the political system is not democratic and the study is being conducted through governmental agencies. This issue was raised in a study of hospitalized abortion complications in Vietnam. There was criticism that women would feel that participation was mandatory because they live in a country where freedom of choice in many matters is not possible. The potential threat in that study was not self-incrimination in a legal sense, because abortion is legal. There were psychological and social threats, however. It was reported to us by interviewers at all study sites that the most difficult question for women to answer was "Are you married?" Unmarried women were embarrassed to be identified as such because their pregnancies were considered reasons for shame. Women wishing to refuse participation in a study where a number of questions could cause them shame, embarrassment, or guilt might have felt unable to do so. Not only does the political situation cause women to believe they have no choice to refuse participation, but the power or status differences perceived by clients between themselves and researchers often creates this perception. This situation is illustrated by my experience a few years ago while monitoring a study of an ongoing postpartum IUD program in a hospital in sub-Saharan Africa. A woman who was being interviewed as part of the control group-that is, she was not getting an IUD-was very upset and agitated as she talked to the study nurse, looking at me the whole time. When asked about this, the nurse told me that the woman was afraid I was going to force her to have an IUD inserted. I was horrified this perception could exist, but upon reflection it is understandable, considering her dependence on the hospital for her postpartum care and my presence as an outside authority. Having said this, however, I would also make the point that studies should not be criticized solely on the basis of the vulnerability of the study population due to poverty, lack of education, or youth. This vulnerability often necessitates some of the program evaluation we become involved with, and such is the case in Balin and Ramos's study. The goal is not to eliminate the potential threat but to eliminate the actual harm by doing all that is possible to assure women of their right to refuse and to demonstrate this when possible. To the extent possible, involving local community groups in the design of a study and training peer data collectors might reduce the fear of exploitation of a vulnerable population.

150

CYNTHIA WASZAK

Ethical Review While permission was ganted by the hospital administrator to conduct this study, this is not the same thing as ethical review. In the best of all worlds, ethical review is conducted by an institutional review board. B a l h and Ramos reported that none was available and that no official reviewwas required for the hospital. Review might have seemed unnecessary if it was not required. However, when there are ethical questions involved in a study, it is always in the best interest of the investigator to find some outside body to review the study for just this purpose. It would not serve a legal purpose, but it would provide a range of viewpoints that might inform revisions in the design of the study and that might also improve the validity and reliability of the data. At the very least, it would serve to answer criticisms that might threaten the implementation of the study. The kinds of groups that might serve in this capacity include local women's groups-especially any that act as advocates for women and/or provide reproductive health care or counseling. Of particular importance is the inclusion of reviewers who are similar to women who comprise the study population. Though the focus has been on individual risks, there also might be negative social ramifications of the study findings. Selective reporting of induced abortion cases was found in this study; one possible consequence of this finding might be greater pressure on health care workers by legal authorities. If more pressure were applied, providers might avoid asking questions that would provide evidence of illegal activity even if the answers might result in better care for the patient.

Maximizing Benefits Some studies provide women with services they might not receive otherwise, but this is not the case here. Researchers were documenting existing services- though it is difficult to say to what extent the process of documentation changes provider behavior-for better or worse. One possible individual level benefit from the study, however, was the chance for women to talk to an interviewer about their abortion and complications. The training of interviewers to be nonjudgmental and sensitive to the needs of the women in the study increases the likelihood that the interview will be a positive experience for them. Overall, the greatest possible benefit from the study would be changes in policy and/or improvements in service delivery that remove barriers to treatment or promote more timely treatment. This benefit is realized only if the information necessary to influence stakeholders in the process of change is obtained and stakeholders are made aware of it. While it is typical for researchers to write reports of their findings for publication in scientific journals, it also is important to plan

INDUCED ABORTION IN ARGENTINA

151

from the beginning of the study a process for making results known to policymakers and program managers and relevant advocacy groups in a useful way. To the credit of Balin and his colleagues, they have spent much time and energy in advocacy activities. I have recently been working in the Women's Studies Project at Family Health International. The research question driving this project was: "How does family planning affect the lives of women?" One of the project objectives, and in fact a key component characterizing this project, was the dissemination of the study results to policymakers and program managers in such a way that they would be used to improve reproductive health programs at all levels in the six priority countries. Our strategy for fulfilling this objective included the involvement of the community at the design stage of the project. Local advisors defined research questions specific to their political and cultural context; local researchers have conducted the research, and plans for the dissemination through the most effective channels were developed before the study results were available. We feel that community involvement in all stages of research is an important tool in maximizing the likelihood that research results will be used in a beneficial way. It also is important that researchers be willing to draw conclusions, as Balin and Ramos have, to identify the practical implications of their results. Many researchers shy away from advocacy because of a fear that it may compromise their "objectivity." Values are inherent to the research process, though, and it is unethical for a researcher not to ensure that the results are used in the most responsible way. It also is the responsibility of the researcher to describe the limitations of the data obtained, of course. The data from this case are descriptive. While examples of how women experience their complications in light of the legal restrictions are important, this case does not provide any information on the prevalence of the problem or provide estimations of percentages of the population who experience these difficulties. The concept of clinical rather than statistical significance is an important one here. The existence of serious adverse events due to delays in treatment, which might be avoided with a change in policy or procedures, is the important finding. In closing, Balin and Ramos ask whether the fear of incrimination and the risk of indictment are risks worth taking because of the ultimate goal of shortening delays in medical attention and decreasing risks to other patients. A cost /benefit analysis is at best hypothetical because it depends on guesses about how much the research process itself increases women's risks of harm and a faith that findings will be used to improve services. While all the risks involved here are real, they mostly are risks that already are found in the treatment procedures themselves because of the legal environment. Increases in the risk because of the research may seem small in comparison to the potential good in identifying ways in which

152

CYNTHIA WASZAK

women can have better access to services. The ultimate realization of this potential good may be out of the control of the researcher, but rhere is much the researcher can do to facilitate the process. The information dissemination process is an important point of discussion as we consider new paradigms for ensuring ethical practices.

154

BARBARA ENTWISLE

Risks of Disclosure A key feature of the case is its focus on an illegal behavior. It is interesting that the behavior in question is not illegal everywhere, and in particular, is not illegal in the United States. I worry that our views of what should be and what could be will affect our evaluation ofwhat is. What are the legal risks associated with disclosing an illegal abortion in Argentina? Apparently the woman can be prosecuted; I do not know what will happen to her if she is convicted. In their presentation of the case, B a l h and Ramos minimize the risks associated with disclosure, arguing that the chances of prosecution are very small. I would just note that legal risks are a combination of the likelihood of conviction and severity of the penalty not just the risk of prosecution. Further, these objective legal risks associated with disclosure are only a subset of the risks that a woman might face. Loss of privacy, embarrassment, and emotional harm are among these other risks. Balin and Ramos state that ethical problems cannot be discussed without reference to the local context; I agree. In this case, the risks are much greater in the Argentine setting than, say, the American setting. Nor are the risks of disclosure limited to women observed and interviewed as part of the study. Bal6n and Ramos turn up "irregularities" in the reporting of illegal abortions that could prove embarrassing, if not damaging, to the hospital where the research was done and to the medical staffworking there. What might an enterprising reporter do with the results? Imagine the headline: "Public Hospitals Cover Up Illegal Abortions." Perhaps this outcome is so unlikely that it is not even worth considering But perhaps not. Public hospitals receive their support from the same body that creates and enforces the laws pertaining to abortion. An even more explosive headline might read: "Women Trick Doctors into Giving Abortions on Demand." One of the nineteen women included in the study said she had lied about having induced an abortion. Needless to say, such a headline could create a lot of problems for the physician in question. Presumably, his or her identity would be held in confidence. But, if publicized in the medical community, it could exacerbate the treatment delays that the researchers would like to ameliorate. In considering risks, it is important to consider the entire context of the research, in this case the hospital and medical staff as well as the patients. BaMn and Ramos presented their findings in a variety of settings, and apparently there was no bad publicity. However, risks must be evaluated before, not after the fact.

The Self-Interested Researcher At least initially, it appears that Balin and Ramos considered the risks to women of participating in the study to be minimal, especially given the importance of the

THE CONTEXTS OF SOCIAL RESEARCH

155

question they sought to address. There is an interesting contradiction here. While the researchers perceived these risks to be minimal, a central idea of the study is that the women did not. After all, it is precisely this fear that BalPn and Ramos believe is responsible for delaying treatment for complications of illegal abortion. How is this contradiction resolved? I do not know for certain. My guess is that while the researchers recognize that women's fears about disclosing an illegal abortion are critical for understanding their behavior, they believe that these fears are unfounded. They, as well-informed researchers, know better. Further, Balin and Ramos felt that the good they could do for poor women in general justified any small harm to specific women participating in their study. The trouble is that while Balin, Ramos, or indeed anyone can be a well-intentioned researcher, each is also a self-interested researcher. For the case that has been presented, an interest in doing important policy-relevant research has the potential to conflict with the interests of the patients recruited into the study. Even if there is no conflict of interest, there is the appearance of one. There needs to be some kind of external review, a system of checks and balances. The fact that the hospital in which BalPn and Ramos worked did not have a formal review procedure, and the fact that the funding agency did not require such a review, does not obviate the need for one. BalPn and Ramos comment on the inapplicability of standard written consent forms in the context of their research project. I agree that such forms can be inappropriate, and in the present case, written consent might itself constitute a risk, given that illegal behavior was involved. Nevertheless, subjects should have the right to know what the risks of participation are, and to weigh them according to their own criteria. Alternative procedures must be developed, which adequately inform potential participants in social research. The spirit, not the letter, is what is important here. Women consented to participate, but were not told that the subject of the research was abortion. Also, it seems that potential participants may have mistaken the researchers for hospital staff, even though the interviewers were careful to identify themselves as university students. It is possible that women connected their participation in the research with the quality of the care they could expect to receive. These women were probably in pain, frightened about their medical symptoms, confused about hospital procedure, and worried that something bad might happen to them in the hospital. Could they say no? Eleven did, which is reassuring, but it does not guarantee that the nineteen women who agreed to participate felt that it was a choice. Their vulnerability may have been compounded by educational and other social status differences between them and the researchers. Since women were reluctant to report the illegal abortion to a physician in a position to help them directly, it seems unlikely that they would report it to a researcher with no immediate aid to offer. If they had been fully informed about the nature of what they were being asked to do, and had felt completely free to

156

BARBARA ENTWISLE

exercise their choice, would they have agreed to participate? From the presentation of the case, it appears that the answer is no. It seems as if some deception was necessary for the project to be completed. The justification would be that lives are being lost because of an inadequate understanding of the negotiation process between physician and patient. The stakes are high. In the presentation, Bal6n and Ramos have asked us to consider the severity of the problem under study as part of the context of the research. Locally determined risks must be weighed against locally construed benefits. Can we accept such a logic? It seems to me that before we can even consider it, we must first demonstrate that the research project will answer the question it was designed to answer.

Research Design Ultimately, we need to be able to justify research based on human subjects not only on the importance of the questions we ask, but also on our ability to answer them. Consider the design of this study in relation to the central hypothesisthat fear leads to treatment delays. The research was intended to be exploratory. Hospital staff identified potential participants as those likely to be seeking medical help because of an illegal abortion. The study involved systematic observation, unstructured interviews with an unknown number of professionals and patients, and guided in-depth interviews with nineteen women at two points in time, first in the hospital and then later in their homes. With this set-up, is it possible to say that fear leads to treatment delays? It is not unusual for researchers to have some general notion ofwhat they want to study and to do some exploratory research to refine the questions. The procedures mentioned might work very well for this purpose, and it is clear that the research generated a much richer understanding of delays and the interpersonal negotiations responsible for these delays in some selected instances. Is this enough? Given the risks, I am very troubled by the idea of doing such open-ended research with subjects who are not free to make an informed choice about their participation.

Contexts One of the great values of this case is that it demonstrates the importance of context for the ethics of social research. What follows is a brief review of how it does this. First, it shows how a particular behavior has many meanings, depending on context. In the case of abortion, which is legal in some places and illegal in others, a consideration of risks and benefits must be specific to the setting. In this instance, the focus was abortion in Argentina. Further, the research was based mainly in a

THE CONTEXTS OF SOCIAL RESEARCH

157

public hospital. In some ways, the ethical dilemmas of the research were magnified by the hospital setting. It is clear that the ethical dilemmas posed by the present case can only be addressed within the context of the case. Second, the case demonstrates how important it is to think broadly about the "subjects" of the research-to consider relationships with everyone involved in the research process. In this case, the "subjects" were not only patients but also medical staff and even the hospital itself. Third, there is a temporal context to consider. The dilemmas posed by a research project may change over time as the investigator develops his or her own understanding and views, but also as a function of the research process itself. It appears that some of the findings were completely unanticipated. It is an open question as to how the self-interested researcher can appropriately deal with such findings. Exploratory research seems particularly vulnerable to such dilemmas. Certainly, a review by some external body at the beginning of a project is no guarantee against shifting ethical sands in the middle of a project. Fourth, there is a research context to consider. What is the accumulation of research evidence? Apparently, it is already known that delays in seeking andlor obtaining medical treatment for incomplete abortions is a major factor in maternal mortality. As initially designed, would this study have contributed important new knowledge to this research base?Was the project designed in such a way as to produce definitive answers to the questions posed? What alternative research designs might - be feasible? Fifth and finally, the case demonstrates the complexity of multiple overlapping and nonoverlapping contexts for the evaluation of ethical issues. One potential set of rules and procedures comes from the funding agency, in this case the Population Council, based in the United States. As it turns out, the Population Council did not require any review, but this will not always be the case. A second potential set of rules and procedures comes from the public hospital in Argentina where the research was conducted. Again, a formal review was not required in this case, but such a review might have been required. The results were intended to be of immediate policy interest in Argentina, suggesting an additional political context in which the costs and benefits should be weighed. The results were also of interest to a larger international research community, of which I am a member. Ethical issues play out differently in each of these contexts. This might be a reason for not overly worrying about following any given set of procedures-that is, someone who is trying to follow several masters will succeed in following none. However, just as interdisciplinary research is improved by the need to pass muster according to multiple standards, evaluation of ethical issues within multiple contexts also has the potential for greater good.

162 CASE 5

which trust serves as a foundation stone of the researcher-subject relationship. A related issue is the notion of the gift: in conference discussion Runyan emphasized that the LONGSCAN researchers view the information given to them by subject families as a gift that encumbers researchers with special obligations toward the givers. A third is the power of expertise. Child abuse is a matter of expert knowledge; of individual, family, and cultural values; and of public policy. How child abuse is defined and dealt with depends on how these domains arc negotiated, and on whether science or society is accorded the lion's share of authority Runyan wants to see expertise given greater deference; Aracena looks to families and focus groups for definitions, thus acknowledging the power of social norms; and King, a lawyer and ethicist, asks how the context in which the research question arises affects, or should affect, the balance of power in adjudicating abuse. LONGSCAN's unique proposal for dealing with the problem of child abuse reporting in the research context combines deference to experts with acknowledgment of the autonomy of the child. In important ways, this solution depends on the long-term relationships that researchers in this project are developing with subject families, who, by the time it is implemented, will already have been followed for many years. LONGSCAN, of course, also wishes to follow these families for many years to come, and does not seek to introduce anything that could cause any of them to withdraw from the research. This relationship, therefore, like all relationships, is one benveen interested parties. The hard question is how the rights of subjects and the duties of researchers can be discharged in an interested relationship. The new part of the question is admitting the interest. In this case, like the others in this volume, the admission of interest is a means of moving to a richer and more realistic understanding of research relationships.

164

DESMOND K. RUNYAN

in family violence research who admit to having engaged in abusive behavior have a risk of having their children removed from the home by the authorities and being prosecuted because of information learned as a result of their participation in research. O n the other hand, nearly one million children are known to be the victims of child maltreatment each year in the United States (Curtis et al. 1995) and the actual toll may be much higher. A 1995 Gallup poll of U.S. parents suggested that more than four million children were physically or sexually abused in the past year (Gallup 1995). Social service agencies have been widely criticized for failing to provide effective interventions (Newberger 1985). Systematic research into the causes and consequences of child abuse is needed to improve societal responses (National Research Council 1993). Barriers that alter the likelihood that some families will participate in research may produce biased estimates of incidence or impact and thus provide useless information. Balancing the need for valid information and the protection of subjects in human research with the need to protect children from child abuse or neglect presents a difficult problem for society. Improving the appropriateness of the societal response depends upon accurate data about the prevalence of maltreatment, the consequences of maltreatment at different developmental stages, the effectiveness of intervention, and the effectiveness of prevention strategies. There is a clear need for research to guide practice and policy.

A Case Study of the Ethical Conflicts One set of investigators has begun to tackle the conundrum of valid and ethical child maltreatmerit research-but we are finding the development of an appropriate set of protective service policies and the development of a careful research protocol to be very complex. The issue has arisen as part of a longitudinal research project that addresses important social policy considerations, but which will require the development of new approaches to potential ethical conflicts if its mission is to succeed. LONGSCAN is a set of five prospective, multisite studies of the determinants and consequences of child maltreatment (Runyan et al. 1998). The agency that has supported our studies, the National Center on Child Abuse and Neglect in the United States Department of Health and Human Services, has recently awarded us a second period of five years of funding. The impetus for this research came from the field and from the United States Congress. The goals of the research are to examine the impact of maltreatment on the child's subsequent development, with specific explorations of child development, school performance, peer relationships, aggressiveness, later dating and mating behaviors, and attitudes about parenting.

MALTREATMENT IN FAMILIES

165

The studies were designed in 1989, begun in 1990, and planned to continue for twenty years. Investigators in Seattle, San Diego, Chicago, Baltimore, and North Carolina are collaborating in the conduct of the studies. The five studies share common definitions, measures, timing of interviews, training of interviewers, and data analysis. The project is coordinated through an executive committee of the investigators and a coordinating center at the University of North Carolina. The major difference among the different constituent studies is that the sample systematically varies in order to examine the impact of societal intervention on the families. Three of the sites have recruited samples among families reported to the authorities for suspected abuse. These samples include families reported to social services for suspected abuse or neglect (Seattle); families reported and investigated, and then provided either long-term family services or "usual care" by social services (Chicago); and children reported for maltreatment who have been in foster care and have either returned home or remained in foster care (San Diego). Two additional samples were also included. In North Carolina, LONGSCAN has a sample of children in families identified as "high risk" by the public health system because of young parent age, child medical problems, or other social risk factors. In Baltimore, investigators have recruited a sample of children at medical risk because of maternal drug use, maternal HIV status, or because of a diagnosis of failure-to-thrive. Three of the studies (Chicago, Baltimore, and North Carolina) have enrolled control subjects who were not known to have been reported for maltreatment at the start of the study. Each of the five studies has sufficient statistical power to address its own important questions. Comparisons across and between studies, either through replications or by pooling the data, will allow the investigators to determine whether there are systematically different outcomes for children reported versus children not reported, and to explore the impact of interventions. In all of the LONGSCAN studies, the families were recruited before the target children were five years of age. When the children are young, the primary respondent is the child's mother. As the children reach six years of age, more data are collected from the children. The oldest children in the project just reached their twelfth birthdays in 1998. By age twelve the majority of the interview data will be collected from the children. The LONGSCAN interview collects data about family circumstances, stresses, and life events; parent background; use of social and health services; and behavioral functioning of the children. To date, the primary data about the occurrence or recurrence of child maltreatment comes from social service records. The lack of concordance between official social service reports and the occurrence of physical abuse or neglect is well documented. In addition to examining official social service reports of child abuse and neglect in all five states on a regular basis, LONGSCAN measures maltreatment indirectly by asking parents about disci-

166 DESMOND K. RUNYAN

pline practices and ascertaining attitudes about children and punishment. A truncated version of the Conflict Tactics Scale (Straus 1979) is used, in which subjects are asked about specific discipline practices, but are not asked about practices that clearly would constitute reportable abuse (such as "beaten up" or "threatened with a knife or a gun"). Although assessing the impact of child abuse and neglect is the objective of the study, LONGSCAN is not now directly asking either the child or the parent about the occurrence of the independent variable-child maltreatment-and is purposely using other less precise estimators. There are many reasons why official state records underreport both whether a child has been maltreated and whether a maltreated child has been the victim of repeated maltreatment. The instability of families may result in children being given different last names at different times in their lives. In some states, children who have already been reported and who have an "open" case may not have a new report recorded, even for new allegations. Cases for which criminal charges or other legal proceedings have been initiated may be reclassified as "neglect" or "dependency" instead of abuse, because the family was willing to stipulate to the less stigmatized accusation. Further, there are known biases in the likelihood of being reported to authorities. Poor families and minority families are more likely to be reported to authorities, both because of apparent prejudice among mandated reporters (Hampton and Newberger 1985) and also because of the institutional biases that result from the reality that poor families use public institutions for medical care, live in crowded buildings with thin walls and close neighbors, and have a greater likelihood of having societal agents, like police and social workers, present in the neighborhood. Maltreatment is the major independent variable of interest for LONGSCAN. We are collecting large amounts of data about child well-being and family functioning, with the goal of understanding the risks related to maltreatment and the impact of intervention. We must have a method of measuring the independent variable of maltreatment. Because of the problems with official records, LONGSCAN has few choices: The investigators can wait for the children to grow up and ask them about their childhood when they are adults and the knowledge will no longer mandate reports, or the investigators can ask the families and children directly about abuse while the children are still minors. The latter approach has an unknown degree of risk of introducing serious bias into the study by losing subjects through refusals to participate. We know of no data or prior experience to estimate the attrition that will result from first securing informed consent from the parents about the risk of participation and then asking either the parents or the children directly about maltreatment. The original LONGSCAN proposal planned to rely upon indirect measures and official reports; it also planned to ask the children directly when they reached eighteen. However, the investigators have been pushed by scientists and prac-

MALTREATMENT IN FAMILIES

167

titioners in the field to reconsider this decision. The data from the studies are thought to be too important for policy to wait eighteen years for accurate classification of the independent variable. The LONGSCAN investigators have decided that pursuing accurate child maltreatment data before the children reach the age of majority is necessary. LONGSCAN will continue to pursue an incremental approach to maltreatment data. First, every two years the names of all of the children in the study will be checked against state files to see if there have been official reports of maltreatment recorded. Second, at the baseline, age four, and age six interviews, the family is asked about discipline when the child misbehaves, using a version of the Conflict Tactics Scale. As described above, the scale has been truncated by removing those questions about behaviors that are clearly abusive. No current LONGSCAN interview questions solicit responses that directly require reporting the families to the authorities. If the family discloses maltreatment by telling the interviewer more than what was asked by the questions, this is to be reported to the authorities. This risk of report is disclosed in the current informed consent process for the study. When the children are eight years of age, the interviewer of the mother asks her to describe the family's approach to discipline and then asks about specific behavior, using the full Conflict Tactics Scale. The interviewer also asks the parent if the child has ever been sexually abused. The research protocol does not gather the data about perpetrator identity The Child Sexual Behavior Inventory (Friedrich et a1. 1991)is included in the interview battery to look for sexual acting out, which might be an indirect indication that the child has been the victim of sexual abuse. This interview, when the child is age eight years, is preceded by a consent process that makes clear that any revelations about specific child maltreatment will need to be revealed to the authorities. To date this consent process has not been very problematic. Among the first 225 families interviewed for the eightyear-old protocol, a few reports have been made to social services and none has resulted in major social service interventions. The potential for problems is much greater for the interview that LONGSCAN puts into the field when the oldest children reach twelve years of age. We ask the children directly about past experiences with neglect or abuse. We use a computer-administered interview that directly asks the child about abuse and neglect experiences. We believe that the normal child at twelve years of age is old enough to both accurately answer questions of this nature and understand the implications of sharing the information. H e or she can give informed consent after having the risks and benefits of participation explained. One particular wrinkle in this interview is that we give the child the opportunity to encrypt some data so that they are not available to the interviewer or the local investigators. We make these data available only to the Coordinating Center, where our procedures al-

168 DESMOND K. RUNYAN

ready ensure that responses to the instruments are divorced from all identifying information. The child is also given the opportunity to alert the interviewer about his or her response to these questions if he or she would like the maltreatment brought - to the attention of an adult in order to seek assistance. The problem with this approach is that the twelve-year-old interview may result in the child's implicating his or her parents in illegal acts, which must be reported by the interviewer. Since the child abuse statutes do not specify a level of severity needed to trigger the reporting requirement, nor do they contain any criteria of recency, simply following state statutes might result in "nuisance" reports to social services for old or relatively trivial events. Such reports would interfere with the research process, but would not prompt any real concern or even an investigation by the local social service agency. The LONGSCAN investigators have used a federal research "Certificate of Confidentiality" to develop their own reporting guidelines to minimize the number of reports that must be made (Melton 1990). This federal certificate, designed for use in studying illegal behavior such as drug use, is surrounded by controversy in the child abuse research field. It is the current policy of the National Institute of Mental Health and the National Center on Child Abuse and Neglect (Malcolm Gordon, Ph.D., personal communication, 1994) not to award research funding to investigators who propose to avoid child abuse and neglect reporting. Our solution gives the child subjects autonomy by respecting their decisions not to share some data, and "triages" the problems so that only maltreatment that poses a serious or imminent risk is reported. LONGSCAN does not use the certificate to avoid all reporting. Instead we use the certificate to develop our own individual reporting policy. This policy commits the investigators to reporting severe or life-threatening abuse and current serious abuse; they are not required to report old or trivial cases that meet the letter of the current reporting statutes but are not likely to lead to action by social service agencies. We believe it is important to develop our own reporting guidelines to minimize the number of reports that must be made. We believe that the data obtained with this approach will more accurately represent what is happening among the families enrolled in the study. It is our impression that the LONGSCAN informed consent process has become longer and more frightening for the participants with each successive interview. LONGSCAN interviews occur at recruitment and when the child is four, six, eight, twelve, and sixteen years of age. We thought we were clear about the risk of reporting rhat families incur when we began, but we have tried to be even clearer about the risk as the children are older. At age eight, children assent to the interview, and at age twelve they sign their own independent consent. We debated whether children of age twelve are old enough to comprehend the risks involved for themselves in reporting a family member for maltreatment. We have chosen ~

-

-

-

MALTREATMENT IN FAMILIES 169

to honor their decisions for most, but not all, forms of maltreatment, after consultation with our institutional review board. Abuse and neglect reporting have not posed a major barrier to the successful completion of the project for the first years of each child's involvement. However, we have discovered that the most troublesome disclosure from the eight-year-old children is a revelation of suicidal intent. Although we worried about abuse disclosures, we were surprised to discover that suicidal ideation is a more common reported concern. Our interviewer training covered a protocol for reporting maltreatment in detail, but was too circumspect on suicidal statements. We have had to develop new protocols for training and debriefing interviewers. Even interviewer hiring has been problematic. Despite the obvious attraction of hiring trained mental health professionals as interviewers, we have found mental health professionals to have great difficulty separating the roles of interviewer and therapist. Some of the therapeutically trained interviewers report that they want to get involved and help the child or family. This tension also appears among the investigators. Some LONGSCAN sites are directed by clinicians, whereas others are not; as a result, there are significant site differences in attitudes about making referrals for assistance when a child has a problematic score on the instruments used to assess developmental level or behavior problems. All of the investigators have agreed to make some basic help information available to the subjects by identifying hot line numbers or specific resources that the family can call. Beyond this basic level of intervention, it has not been possible to develop a LONGSCAN consensus on the appropriateness of making referrals based on information acquired through study participation. There are site-by-site variations in the frequency of referrals. All referrals are recorded in the data entry system, so that subsequent analyses will be able to examine whether this factor induces significant variation. It is clear, however, that the involvement of families in LONGSCAN is an infrequent experience, which pales in the face of the life events encountered by our subjects on a daily basis.

Conclusions Over one million children per year in the United States are investigated and have social service interventions imposed upon the family as a result of reports of child abuse, and'there are many others in countries such as Canada, the United Kingdom, Australia, the Netherlands, and France. Interventions are now being chosen without the benefit of a body of scientifically collected information regarding their risks and benefits. There is a widely acknowledged need for more and better data about treatment alternatives.

170

DESMOND K. RUNYAN

In LONGSCAN, one of the major longitudinal studies of the impact of maltreatment, we have identified problems related to the process of securing informed consent from parents and children about the risks that might accrue from directly asking children about child maltreatment in the context of longitudinal research. The investigators of this project have committed themselves to working out a process by which children may be directly asked about their past histories of abuse and neglect in the face of legal reporting criteria. Out of respect for persons, we have attempted to give the children (I) protection from "nuisance" reports about old or minor maltreatment and (2) control over whether their revelations are intended to seek help or are for the benefit of the research only. The consent process for parents is more problematic. We do not hide the fact that we ask children about abuse. The parents have to give informed consent. Differential loss to follow-up, with abusing families denying permission for child involvement, may become a serious threat to the validity of the study. We hope that the data collected from other sources will permit an assessment of the loss-to-follow-up problem. In the case of LONGSCAN, we are somewhat less concerned about drop-out rates. We already have a wealth of data about the children from a variety of sources; in addition, the eighteen-year-old follow-up, when the children can be asked with no legal jeopardy to their parents, is expected to provide important information soon. It is our hope that the indirect data will help us understand the forces related to the loss of subjects through the process of securing informed consent. Indeed, data on loss of subjects will itself be a great help to the field. LONGSCAN has developed an innovative approach to research about maltreatment with the protocol for reporting for the twelve-year-old interview. We look forward to the response of the scientific community and the solutions others develop based upon our approach.

Note I. The LONGSCAN approach is the model for the proposed National Study of Child and Adolescent Well-Being, a new study examining the impact of the child welfare system on a sample of nearly 7,000 children in IOO U.S. counties. This model will thus be scrutinized by hundreds of researchers, institutional review boards, and policymakers interested in the complex questions we address in this project.

172

MARCELA ARACENA ALVAREZ

case in the United States. The question is then: Why should two societies that face a similar problem have such different ways of protecting the rights of affected individuals? This commentary examines (I) diversity in definitions of child abuse, (2) conflicts berween confidentiality and the obligation to report, (3) issues of free and informed consent, and (4) cross-cultural variation in the ethics for clinical research. Sociocultural, historical, legal, and clinical factors are explored in order to understand these differences.

What Is Child Abuse? Each society or culture has its own standards for child care and discipline (Korbin 1980; Samuda 1988). In this context, what is understood as "child abuse" substantially differs, not only according to whether the definition is medical, psychological, or legal (Biller and Solomon 1986; Gelles 1982; Starr et al. 1990; Valentine et al. 1984)~but also according to the perspective of the community under study. Kempe (1982) has emphasized that definitions of child abuse vary among cultures and evolve over time. Definitional differences can occur not only among but within countries, and across researchers, states, and agencies (Gelles 1982; Gelles and Cornell 1983; Starr et al. 1990; Valentine et al. 1984). Kamerman (cited in Gelles and Cornell 1983) reports that definitions of child abuse vary from one European country to another, depending upon the cultural perceptions toward chtldren and the perceived extent of child abuse in the respective countries. Korbin (1980) has argued that if society fails to allow a cultural perspective in the definition of child maltreatment, we will be hopelessly locked in an "ethnocentric" position in which our own set of cultural beliefs and practices are to be preferred and considered superior to any other. I have used a qualitative descriptive design to explore how child-rearing practices and child abuse were perceived by Chilean professionals and community representatives, using focus group discussion and in-depth interviews. During the focus group discussions, the concept arose of a "child-rearing patterns-abuse" continuum. This continuum concept synthesizes the value that an adult gives to the strategies used during the socialization process. We found that a strategy can be perceived by different adults as appropriate or not, depending on the specific situation. Whether a strategy is used as a child-rearing practice, and whether it is adequate or not, will depend on different factors, such as (I) the emotional state of the parents, (2) the stress level in the family, (3) the parents' expectations, (4) education level, (5) economic difficulties, and (6) the parents' history. Participants in this study recognized that many of the beliefs and practices related to child abuse are perceived by the community under study more as childrearing patterns than as child abuse. Physical punishment and threatening, for in-

CHILD ABUSE RESEARCH 173

stance, can be perceived as appropriate child-rearing practices, not as abusive ones (Aracena et al. 1994). All focus groups discussed the "breakpoint" between childrearing practices and child abuse, and participants identified criteria that would allow us to determine when a behavior could be called abuse. These criteria were: (I) frequency, (2) intensity, (3) emotional control level of the adult, (4) meaning and context for the child, ( 5 ) psychological, cognitive, and physical stage of development, and (6) cultural beliefs. Similarly, in Hong Kong, a self-completed questionnaire was given to IOO unmarried university students, ages twenty-two to thirty-one years (Samuda 1988). The researchers found that physical punishment was used in ninety-five percent of the respondents' homes. This phenomenon is explained by the author as the way a filial society ensures that children will obey and care for their aged parents. Because of that, he explains, the requirement of the eldest child's obedience is associated with strict discipline. Another example is given by Fontes (1993) when she hypothesizes that certain general ideologies of child-rearing exist within Puerto Rican culture, and certain practices follow from these ideologies. The emphasis on obedience leads many children to comply with adults' sexual advances, and to keep silent if an adult has forbidden disclosure. Do we have the right to consider our own set of cultural beliefs and practices better or worse than others? Korbin (1980) has argued that it is not possible to take a stance of extreme cultural relativism in which all judgments about treatment of children are suspended because of cultural sensitivity or awareness. It is necessary to begin to build definitional issues into a coherent framework so that child maltreatment can be appropriately identified within and across cultural contexts (Finkelhor and Korbin 1988; Korbin 1980). At this point, however, we must admit that American social awareness of child abuse differs from that in Chile. As Gelles (1987) has stated, whether a country recognizes child maltreatment depends on local definitions, traditions, and priorities. Assuming that cultures differ on definitional issues, how do researchers in different societies behave with regard to children who have been abused? I will analyze the elements from this ethical standard comparing data from developed and developing countries regarding (I) the information given to prospective subjects about the possibility of reporting suspected child abuse, and (2) the informed consent process.

Respect for Individuals

The Conjict of ConJidentialiq and the Obligation to Report From a historical perspective, problems exist because each society substantially differs in its awareness of the different types of child abuse over time (Kempe 1978;

174 MARCELA ARACENA ALVAREZ

Kempe 1982). The United States did not recognize the battered child syndrome until 1962 (Kempe et al. 1962). From the 1960s to the present, a number of efforts have been developed in different countries toward the acknowledgment and treatment of child abusc2 But societies like Chile, which have been exposed to an authoritarian system in which human rights have been transgressed, are behind in this process. Child abuse is only widely recognized in democratic regimes. In Latin America, violence has only been considered an important issue in the last eight to ten years. Knowing the background of a particular society, it is possible to understand societies like the Chilean one, where the acceptance of the phenomenon is currently an important issue, but legal aspects such as reporting the cases are only beginning to be addressed. In some legal codes, like Chile's, moderate punishment by parents is recognized as appropriate and legal and does not reach the threshold ofwhat is expected to be reported to authorities (Fernandez 1995).Although to date Chilean laws do not explicitly require researchers to report cases encountered while carrying out studies, once researchers are legally responsible for reporting suspected child abuse, many issues will require answers and bring new dilemmas. For example, who should report abuse? What is the statute of limitations for reporting? DeKraii and Sales (cited by Bussell 1994) have provided guidelines for reading and understanding child abuse reporting statutes that help clarify the researcher's legal and ethical responsibilities in other contexts. The American Psychological Association (APA) suggests that an obligation based on general ethical principles exists for the researcher to disclose information to potential subjects in order to avoid harm (see Bussell 1994).The ethical tension between researchers' responsibilities to serve the best interests of the participants and to avoid harm is not yet resolved. Laws differ in different contexts, requiring specific but different responses to the same behavior. In addition, professionals' adherence to mandatory child abuse reporting laws is not clear. For example, in an exploratory survey conducted to assess whether mental health practitioners possess an adequate knowledge of their legal obligations, it was shown that a significant proportion of psychologists, psychiatrists, and social workers (37 percent) were unaware of two of the most basic laws that apply to their professions: one providing for the right of confidentiality and privileged communications, and another requiring the reporting of child abuse (Swoboda et al. 1978). As to clinical factors, problems exist for researchers for at least four reasons: (I) professionals' variable adherence to mandatory child abuse reporting laws, ( 2 ) different views about what is understood by breach of confidentiality (3) the balance between resources and needs, and (4) the balance between risk and benefits. Many efforts have been made to understand the factors that bear upon professionals' decisions to report suspected child abuse. Even though, as we have de-

CHILD ABUSE RESEARCH

175

scribed before, every state in the United States requires reporting suspected child abuse, several studies have found that many professionals fail to report because of the harm they feel it may cause to the family (Muehleman and Kimmons 1981; Swoboda et al. 1978). Studies have shown that many factors influence the decision to report or n0t.j It has been thought that educational programs increase the identification and reporting of maltreatment on the part of professionals (Nightingale and Walker 1986). The right to privacy implies that individuals have control over information and communication about themselves. Marshall (1992) notes that a problem arises when researchers experience a conflict of interest between the traditional obligation to protect a subject's confidentiality and a broader social obligation to protect a third party who may be in danger. Bussell (1994) describes three predominant views. The first view considers it best to omit the possibility of breach of confidentiality from the consent procedure; thus, if a breach were to occur, it would be explained to subjects as falling into the category of an unanticipated emergency. A second position argues for disclosure of the potential breach of confidentiality without giving details of specific risks. Finally, a third view holds that a family's trust in the interviewer would be enhanced by a complete specification of the circumstances requiring breach of confidentiality. This last position has received attention because it has been argued that only a full disclosure of information fulfills the researcher's ethical responsibility to respect the dignity and rights ofrhe participants (Bussell 1994, citingAPA 1982, Principle J, andAPA1992, Principle D). Another clinical issue that should be addressed is the balance between cases reported and the resources the state has to meet these needs. In Latin American countries, like Chile, the resources to intervene and treat people who have suffered abuse are limited. Sexual abuse and severe physical abuse would be the kinds of abuse professionals would be willing to report if they were sure about the occurrence of abuse and if they were sure that state institutions could undertake the treatment of the child and family. It should also be acknowledged that the evaluation professionals make of risks and benefits tends to be inclined toward risks. Sometimes, risks to family integrity can be perceived much more clearly than the benefits the social system can offer. Because of that, professionals are against reporting cases because they do not want either the minor or the family to suffer. As we have seen, the reporting obligation constitutes an important dilemma for any research team. Ethical considerations like those described in this paper need to be addressed in the research design phase. Internal policies must be developed that are consistent with legal duties in every state and country in which child abuse research is conducted.

176 MARCELA ARACENA ALVAREZ

Free and Informed Consent Regarding sociocultural factors, problems exist for at least three reasons: "Autonomy" itself is a culturally determined construct, which may be inappropriate or difficult to recognize in some cultural contexts (Streiner et a1. 1995). The power differential between patient and physician is much greater in some countries than others, which may obviate the notion of freedom for some participants in studies directed by physician-researchers (Streiner et al. 1995). Prospective subjects' education and intellectual capacity may not always enable free and informed consent (La Vertu and Linares 1990). Autonomy is rooted in the concept of personhood. In some cultures, however, there is little perception of conflict between self and society. In some parts of South America, the relationship of the subject with the society is not viewed the same way as it is in the United States. Rather, each person is considered a participant in the common effort of a collective whole. Accordingly, subjects are expected to participate in projects supported by the community (La Vertu and Linares 1990). In addition, Gostin (1991) notes that in the Indian subcontinent and West Africa, great deference is given to clinicians / healers /elders. If permission has already been given by a community or family representative, the idea of an informed refusal by the individual may not even arise. The second issue is that of the power difference between doctor and patient. Not until 1947 did the Nuremberg Code state that the voluntary consent of human subjects was absolutely essential in research (Thomas 1978). Since that time much effort has been devoted to following this guideline; still, human rights continue to be an issue in many societies. As a result of Chilean history, for example, where human rights were transgressed, even today the community is prepared to respond to state authorities without questioning and without feeling that privacy rights are being trampled. The right to individuality was undermined in a system where individual rights were not respected and the authorities had the right to enter into private places (such as homes) without any permission. In addition, in Chile there is an unequal hierarchical relationship between doctor and patient, especially for persons of low and middle socioeconomic levels. Thus, physician-researchers can request almost anything and subjects are likely to show no opposition. Informed consent requires that participants have enough intellectual capacity, give sufficient thought to the matter, and know enough about risks, benefits, and options available to provide effective consent (La Vertu and Linares 1990). Linguistic barriers can also be obstacles to informed consent (Marshall 1992). In Latin American countries such as Chile, educational standards are low and illiteracy very high (De la Pefia and Gonzilez 1989). From this perspective, in many devel-

CHILD ABUSE RESEARCH 177

oping countries, informed consent sometimes can be more a bureaucratic procedure than an ethical issue. Regarding clinical factors, problems exist for at least two reasons: psychological development of the child and family, and family awareness about their own level of violence. The psychological development of the child and the educational level of the family will not always allow either the child or the child's parents to give a free and informed consent to participate in a study. Hunter (1995) asks researchers to think about the following questions: (I) what do children understand about promises not to tell? and (2) if children tell researchers that they are being hurt, will they not expect researchers to help them even though researchers have promised confidentiality? In addition, Hunter asks whether parents or children know what the law considers to be child abuse, and thus know what disclosures might place them at risk. Second, family awareness about their level of violence could differ from the researchers' perspective. For example, Browning and Dutton (cited in Aldarondo and Straus 1994) report that men in treatment for spouse abuse have been found to report less violence than their partners do. From that perspective, could parents be insufficiently aware of the implications of the consent they are signing? Are they giving information about behavior that members of their community do not perceive as "wrong or illegal," without realizing that these disclosures are considered reportable by researchers?

Cross-CulturalVariation in the Ethics of Human Subjects Research Some researchers and scholars question whether it is proper for Americans to insist that their ethical standards be applied to human subjects research performed in other countries (Angel1 1988; Barry 1988). Should ethical standards be substantially the same everywhere, or is it inevitable that they differ from one region to another, reflecting local beliefs and customs? Christakis (1992) has suggested four different models engaging cross-cultural variation in the ethics of research with human subjects. The first model suggests that no transcultural research can be done. An alternative model is to require transcultural research independently to meet the ethical requirements of the two cultures involved. A third model suggests that an absolutist system of ethics could be developed around the observed common topics. Finally, the fourth model suggests that Western research ethics should be applied at a global level. As we have seen in this paper, ethical decisions about how to conceive a research study are undoubtedly influenced by the society involved. In this context, is the Chilean position less ethical than that of the United States? The question that should be asked is this: Which ethical considerations should be taken into

178 MARCELA ARACENA ALVAREZ

account, keeping in mind the idiosyncratic aspects of each nation but without failing to take care of children exposed to abuse? Perhaps we should focus on selecting some basic ethical precepts with a global scope, such as respect for persons, and discover how to apply them in each society from a cross-cultural perspective. Certainly, Chilean researchers could agree upon some widely accepted criteria concerning which cases should be reported immediately, including those which could be very harmful to the emotional well-being of the children because of the severity, intensity, or consequences of the abuse. From this point of view, we think that the proposal made by the LONGSCAN team concerning the development ofspecia1 policies for those cases that should be reported is promising. LONGSCAN must also develop a risk-management plan to be implemented in the event that clinically significant cases are observed (Bussell 1994). As for informed consent, LONGSCAN has set out to develop creative means of ensuring that the decisions of children and adults to participate in research are freely taken, that the research has a realistic potential for good outcomes, and that the conduct of the research is truly fair and honest (Streiner et al. 1995).As Streiner and colleagues suggest, the goal of informed consent policy in cross-cultural settings is to design alternatives that guarantee protection of individual rights and promote autonomy within particular cultural contexts. Finally, the case presented by the LONGSCAN team has given us the opportunity to think about issues of cross-cultural difference in the ethics for clinical research. We agree that systematic research about the causes and the consequences of child abuse is needed, and that the process of balancing the need for valid information and the protection of human subjects in research with the need to protect children from child abuse or neglect presents a very difficult problem for society. However, ethical considerations in research must take into account cultural differences. To illustrate this, consider the following: If current American guidelines were followed with UNICEF data in Chile, six out of ten parents should be reported to the legal authorities. No system could withstand such an onslaught of cases, nor could any research project.

Notes I.

Factors include the 1985 and 1990 resolutions concerning violence at home by the

U.N. General Assembly, laws passed in Latin America and the Caribbean against intrafamily violence (Puerto Rico in 1989, Costa Rica in 1990, Bahamas in 1991, and Barbados in 1992),and the acceptance and signing ofthe Children's Human Rights Agreement in 1990, among others. 2. See Agathonos (1983), Agathonos et al. (1982), Arnold (1982), D e Silva (1981), Dubanoski and Snyder (1980), Farinatti et al. (1990)~Fraser and Kilbride (1980), Gelles (1987), Haffejee (1991), Ikeda (1982), Jinadu (1986), Korbin (1981), Liakopoulou and Xypolyta

CHILD ABUSE RESEARCH

179

(1989)~Nathan and Hwang (@I), Okeahialam (1984), Payne (r989), Robertson and Hayward (1976), and Samuda (1988). 3. Factors such as the age of the child, the child's behavior during a clinical interview, situational factors, clinic experiences (Kalichman and Craig 1991), the accused parents' acceptance or denial of abuse, clinicians' expectations of what effect reporting would have on continued therapy (Kalichman et al. 1989)~and confidence in the occurrence of abuse (Kalichman et al. 1988, 1990) influence professionals' decisions to report suspected child abuse. In addition, the child's life and secrecy (Muehleman and Kimmons 1981), the type of event, staff commitment (Rindfleisch and Bean 1988), and personal and professional characteristics can influence the decision to report or not (Haas et al. 1988). It is clear from all these studies that professionals' adherence to mandatory child abuse reporting laws has not been adequate in the United States.

RESEARCH IN DISTRESSED FAMILIES

181

in conflict with the interests of their children. What if some parents refuse to allow their children to participate because they are concerned about overreporting of abuse, and are reluctant to risk exposing their families to social services investigations that do not ultimately uncover abuse?2O r perhaps some parents may refuse because the disclosure causes them to realize that their children may be asked questions about issues that, in their view, are potentially upsetting or inappropriate for their age.? In examples like these, the researchers, parents, and society may each have a somewhat different perspective on what constitutes responsible advocacy for children. Unlike some child abuse researchers, Kunyan does not wish to take advantage of either of two widely discussed ways to maintain the statistical validity of the research. Those two ways are failing to disclose the reporting requirement, and failing to report. Some researchers argue that in most such research, the likelihood of uncovering reportable abuse is so low as to be unduly alarming to families when disclosed. This may certainly be true, but researchers must also consider the magnitude of the impact of even an extremely unlikely occurrence. Thus, a very small chance of the occurrence of a very great harm should be disclosed. This is a standard informed consent requirement.* How we measure the magnitude in this particular instance depends upon our views about the impact of reporting and the effectiveness of the interventions that follow reporting-and, as we have seen, a variety ofviews can and do exist. Does the system for reporting suspicion and substantiating abuse deal appropriately with "false positive" reports? Does the system for intervening when abuse is substantiated do good, or does it do more harm than good? Other researchers have also sought to design studies in which, although reportable abuse is in fact disclosed, the disclosed information cannot be linked to an identified subject. Examples include anonymous telephone surveys conducted through random digit dialing, and studies gathering information not through interviewing techniques but on paper questionnaires or computer-assisted surveys, without maintaining linkage files. Similar techniques have been used in medical research, particularly for large-scale surveys of the prevalence of sexually transmitted diseases, such as AIDS. Since sexually transmitted diseases must be reported, and since, therefore, subjects' consent is required for blood testing that can be linked to individual subjects, only anonymous blood sampling can avoid both consent and reporting. However, researchers then find themselves in the position of determining that some blood samples are HIV positive, without having the ability to inform the owner of the blood. An assiduously cultivated relationship of anonymity does not necessarily obviate the researcher's feeling of relationship with, and obligation toward, the owner of infected blood or the anonymous child who acknowledges abuse. But here as well, the magnitude of the impact of not reporting is placed at issue. In these in-

182

NANCY M. P. KING

stances, nonreporting should not be viewed only in terms of the possibility of legal sanction; the justifiability of not reporting depends fundamentally on researchers' judgments that not reporting is not so bad. And the researchers' judgments, in the case of child abuse, appear to be at odds with society's. In medicine and science, we usually think of research as that which brings new technologies and other advances into common practice. From this perspective, if you cannot find an ethical way to conduct research, you cannot introduce the new thing. This may be undesirable if no available technology effectively addresses a serious problem, but it may not be so bad if all it means is that yet another expensive and not very effective treatment, or yet another so-so headache remedy, is kept off the market. But innovation does not always work that way. Many new technologies are simply introduced into use. Sometime later, after they have become entrenched and have begun to provide conflicting evidence of effectiveness, someone comes along and says, "Hey, wait a minute-we need to do some research to see if this new thing really helps, because we think it might not!" Many new health care technologies are introduced in this way, and research into their effectiveness is often undertaken only after they have become accepted (see, for example, King and Henderson 1991). In this instance, if you cannot ethically do research, you are not keeping something unproven off the market; you are doing the oppositekeeping something that is unproven, and potentially detrimental, in use. Runyan wants to argue that the existing system of child abuse intervention may be like the technology that was adopted without testing, and that if reliable research about child abuse cannot be done, we will stay stuck with a system that does more harm than good. In other words, what needs to be tested is a series of assumptions that underlie child abuse policy, including what should be reported and investigated; when intervention is appropriate; and what interventions are effective. This is a very different way to look at public policy, and at the researcher's all-too-common claim that science needs to be above the law. Here the claim is that research must inform the law. But law and public policy are informed by a variety of factors in addition to research. As Aracena points out, policy is influenced not only by cultural norms but also by what a society can afford to be concerned about. Certainly it is true that, as Runyan has noted, priorities in investigating and intervening in child abuse differ not only from state to state but also, within each state, from county to county, depending on the budget, staff size, training, and other priorities ofcounty social service departments. Aracena suggests that despite these variations, there may be universal or universalizable notions about what counts as child abuse. Whether or not this is so, there are some questions that must be addressed. How do definitions of child abuse reflect cultural norms, national priorities, or universal concerns? For instance, do we decide that something is to be called child

RESEARCH IN DISTRESSED FAMILIES

183

abuse when we know or suspect that children are harmed by it? If harm is the principal criterion, then not reporting suspected child abuse discovered in the research context may look something like withholding treatment from the subjects in the Tuskegee syphilis experiment. O n this basis, we might decide that exempting researchers from reporting is dangerous, even when they are conscientious and have no desire to see children harmed. Even so, to ground a reporting requirement, we must have some agreement on how harm is measured, what kinds of harms we should be concerned about, and how certain we must be that harm has resulted or will result. What kind of harm is the "worst" kind-death and physical injury? Psychological harm? Impairment of the victim's ability to function appropriately in society, and to effectively raise his or her own children? O r perhaps we identify abuse when we think we have an intervention that will be effective, or when we have enough money to address it. Does our view of abuse then change when we change our minds about what is affordable and what works, the way medicine defines diseases according to whether we have treatments, and health insurers define medically necessary treatment according to its cost-effectiveness?Runyan and other child abuse researchers have noted that child abuse interventions have seemed to go in and out of fashion. For example, whether it is better to keep the family intact or to place children in foster care depends on two factors: (I) on whether the state has a sufficient budget to supply foster placement for all children who have been determined to have been abused, and (2) on currently prevailing theories of child development and family psychology. According to this view, what constitutes abuse and how it should be dealt with depends on how expert theory is translated into public policy. But how is public policy to be determined in American democracy? It might be argued that child abuse is, almost by definition, behavior that falls outside of society's established consensus on what kind of parent-child relationship should be taught and encouraged by the state. How do our determinations ofwhat is not acceptable in the parent-child relationship, and of how that unacceptable behavior should be identified and dealt with, help us to understand the values we see as underlying that relationship? And how do these determinations help to draw the boundaries and intersections between that relationship and others, particularly the state's relationships with child abuse researchers and with families? How important is the freedom of parents to exercise personal judgment and values about raising children? How important is the responsibility of someoneparents, and if not parents, then the state-to protect children from harm? How important is it that states also be able to exercise judgment and promote particular values? How important are the judgments of professionals and experts, what constitutes expertise, and what values underlie expert judgments about child abuse and the issues it raises? Disclosure, parental consent, and confidentiality in research seem at this point

184

NANCY M. P. KING

to be notions that do not help us much in addressing those rather cosmic questions. But in the move toward an ethics of relationships, these are the terms by which the relationship between researchers and public policy about child abuse is negotiated. Runyan's proposed model in which researchers, rather than social service agencies, make cost-benefit judgments about what constitutes potential abuse requiring investigation and intervention has some obvious benefits, in that the guidelines used can be systematically developed and applied, and the results can be memorialized for study, discussion, and refinement. It also has at least one obvious risk, in that its explicit goal is to minimize the intrusion of reporting upon research. This goal necessarily privileges research over intervention. Runyan argues that social services agencies are not neutral public policy enforcers but interested parties, whose widely varying standards, training, budgets, and case loads are not fair to parents or to children. He seeks better data and believes that better data will lead to more uniform interventions that will work better to meet children's needs. It is worth noting that the child abuse reporting system was originally intended to provide more uniform intervention. Physicians and others who often encountered child abuse did not deal with it consistently, and laws requiring reporting were thought likely to promote the same treatment for every child regardless of the family's position or relationship to the health care and social services systems. Runyan presenrs persuasive evidence that differential trearment is still common, and sees research such as LONGSCAN, which appears to ask for special treatment for researchers, as a means of establishing - a sounder scientific basis for uniform treatment of child abuse. Of course, Runyan and other child abuse researchers are also interested parties, and it is appropriate to be skeptical of their claims that only they can determine what is best for abused children. In fact, Runyan himself and the LONGSCAN team are willing to be skeptical in this way, and offer a radical alternative. Instead of just choosing between reporting rules designed by the state and reporting rules designed by LONGSCAN, Runyan proposes that once they reach twelve years of age, LONGSCAN; child subjects should be able to decide whether they want to report or not. Many questions have yet to be answered about this novel approach, but one seems especially critical. Is adolescent autonomy an adequate alternative to uniformity in child abuse reporting? Put another way, is this a means for adolescents to adopt or reject the values and priorities within their families, or is it a way for LONGSCAN researchers to pass the buck to their adolescent subjects? From the diehard autonomist's perspective, the parens patride role of the state often appears paternalistic; thus, giving children some say in how they want to live makes more and more sense as they gain psychosocial maturity and approach the age ofmajority. However, ifpreventing harm to children is the overriding con-

RESEARCH IN DISTRESSED FAMILIES

185

cern, permitting children to shield their parents from the consequences of harming them seems like a significant overvaluing o f a disordered parent-child relationship. A key difficulty i n answering these questions is as simple a n d profound as determining whether the child o r the family is the principal focus of interest, scrutiny, a n d protection, in child abuse policy o r i n child abuse research. I n relationship terms, LONGSCAN's unique proposal allows the twelve-year-old child t o determine how the intertwined parent-child-researcher-state nexus of relationships should be untangled. But L O N G S C A N has to figure o u t h o w t o explain this t o subject parents a n d children, as well as t o state a n d federal agencies a n d IRBs. A n d the explanation depends o n making explicit some of the values that underlie past, current, a n d future public policy o n child abuse. I d o n o t envy them the task, b u t I a m glad they are taking it on.

Notes I. "Let us not forget that progress is an optional goal, not an unconditional commitment, and that its tempo in particular, compulsive as it may become, has nothing sacred about it. Let us also remember that a slower progress in the conquest of disease would not threaten society, grievous as it is to those who have to deplore that their particular disease be not yet conquered, but that society would indeed be threatened by the erosion of those moral values whose loss, possibly caused by too ruthless a pursuit of scientific progress, would make its most dazzling triumphs not worth having" (Jonas 1969). 2. The standard for reporting is suspicion, which is generally undefined and can be interpreted in avariety ofways. Child protective agencies are required to substantiate abusealso a term susceptible to various interpretations, but necessarily representing a higher standard than suspicion of abuse. Thus, one thing of which families can be certain is that more abuse is suspected (and therefore reported and investigated) than is substantiated. 3. One manifestation of the recent resurgence of interest in "family values" is a renewed call for "family privacy," a call that includes proposed legislation-for example, the Family Privacy Protection Act of 1995 (H.R. 1271, 104th Gong.)-that would have, among other things, ensured that parents retain control over their children's participation in research. 4. For example, in the important case of Canterbury v. Spence (464 Ezd 776, D.C.Cir. 1972)~the court held that the defendant surgeon was required to inform the patient that the exploratory back surgery he recommended carried a one percent chance of paralysis.

disclose; in others, on who has veto power over research. But only in this case are we even tempted to dispense with consent-and only in this case are we confronted with immediately life-threatening, emergency circumstances. If informed consent is important, here is where it must stand or fall. As Kraybill and Bauer note, IRBs charged with reviewing and approving human subjects research will take on some unfamiliar tasks under the consent waiver rule, including that of identifying the communities associated with emergency research and supervising their notification about it. The role of the community in this research is rather different from the community's role in our first two cases. Browder, a physician and ethicist who serves on several IRBs, further problematizes the concept of "community" and its relationship to the autonomy of potential subjects. The rule itself appears to gloss over the difficulty of this concept, to conflate reasoning about beneficence in research with reasoning about beneficence in treatment, and to marginalize concerns about subjects' autonomy and preferences. King, a lawyer and ethicist, points out that we should not be surprised at the principlist paradigm's capacity to shift weight from one principle to another. She claims that a relationships paradigm would force into the open some of the contentious issues that have remained unarticulated in the debate over this rule. Finally, she argues that because these hidden relationships will govern the rule's application, we had better get to work on it with the new p a r a d i p . With this last case, and its detailed examination of the federal regulatory role in ensuring the ethics of human subjects research, we have come to the logical conclusion of this volume's academic function and the logical beginning of its practical utility. In teaching about research ethics, in designing research projects, in preparing to go before the IRB, in deciding how to explain research to potential subjects, and in serving on IRBs, researchers must use everything this volume can teach them and more, if human subjects research is to remain worthy of being called a moral enterprise.

192 ERNEST N. KRAYBILLAND B. SUSAN BAUER

than minimal risk to subjects-a condition not met by the National Acute Brain Injury Study. IRBs of centers participating in the Acute Brain Injury Study were required to use new criteria for approving the study. One requirement was to find and document that there had been "consultation with representatives of the communities from which the subjects will be drawn." Individual IRBs used differing interpretations of this requirement in approving the study. Some cited the presence on the IRB of members who were not otherwise affiliated with the institution and represented "the community." Others brought together individuals already serving their institution as community representatives to fulfill this requirement. The Committee on the Protection of the k g h t s of Human Subjects-the IRB serving the University of North Carolina at Chapel Hill's School of Medicine and University of North Carolina Hospitals-had approved the Acute Brain Injury Study earlier, with a requirement of informed consent by a surrogate. When asked to approve the waiver that had been authorized by the secretary, the committee debated the merits of such a provision but approved it. There was little attention given to, and virtually no discussion about, the specific new requirements that had been imposed. Rather, the committee relied on the traditional guidelinesminimizing risks, balancing risks against anticipated benefits, and equitable selection of subjects. When asked later, by the Study Coordinating Center, to disclose how the requirement of "community consultation" had been met, the committee could only cite membership on the committee of individuals not otherwise associated with the institution. The Study Coordinating Center replied that this was an unacceptable method of satisfying the requirement of community consultation. The ensuing informal discussions among IRB staff members, and with members of other IRBs, failed to identify a method that would truly constitute community consultation. The conditions that IRBs were required to find and document in order to allow the waiver for subjects in the Acute Brain Injury Study were similar to those that had been proposed in a consensus statement that emerged from a January 1995 Public Forum sponsored by the Food and Drug Administration (FDA) and the NIH on Informed Consent in Clinical Research Conducted in Emergency Circumstances (Biros et al. 1995).The stage for that meeting had been set by a 1993 ruling by the Director of the Office for Protection From Research Risks (OPRR) that "deferred consent" did not constitute informed consent as required by Department of Health and Human Services (DHHS) regulations (OPRR Reports 1993). Deferred consent had been proposed by investigators involved in earlier studies of comatose patients (Abramson et al. 1986). Under that procedure, incompetent subjects would be enrolled without consent; when a family member became available, or the subject became competent, consent would be sought for the subject's participation.

COMMUNITY CONSULTATION AND INFORMED CONSENT 193

About two months after the announcement that DHHS Secretary Shalala had waived consent requirements for the National Acute Brain Injury Study, the FDA published, for comment, notice of a proposed rule that would extend the waiver provisions to other kinds of emergency research (DHHS 1995b). Over a year later, the final rule was published, effective November I, 1996 (DHHS 1996b). The final rule amends current FDA regulations in order to permit a waiver of informed consent if the IRB finds and documents each of the following: The human subjects are in a life-threatening situation, available treatments are unproven or unsatisfactory, and the collection of valid scientific evidence, which may include evidence obtained through randomized placebo-controlled investigations, is necessary to determine the safety and effectiveness of particular interventions. z. Obtaining informed consent is not feasible because: i) The subjects will not he able to give their informed consent as a result of their medical condition; and ii) The intervention under study must be administered before consent from the subjects' legally authorized representatives is feasible; and iii) Therc is no reasonable way to identify prospectively the individuals likely to become eligible for participation in the clinical investigation. 3. Participation in the research holds out the prospect of direct benefit to the subjects because: i) Subjects are facing a life-threatening situation that necessitates intervention; ii) Appropriate animal and other preclinical studies have been conducted, and the information derived from those studies and related evidence support the potential for the intervention to provide a direct benefit to the individual subjects; and iii) Risks associated with the investigation are reasonable in relation to what is known about the medical condition of the potential class of subjects, the risks and benefits of standard therapy, if any, and what is known about the risks and benefits of the proposed intervention or activity. 4. The clinical investigation could not practicably be carried out without the waiver. 5. The proposed investigational plan defines the length of the potential therapeutic window based on scientific evidence, and the investigator has committed to attempting to contact a legally authorized representative for each subject within that window of time and, if feasible, to asking the legally authorized representative contacted for consent within that window rather than ~roceeding without consent. The investigator will summarize efforts made to contact legally authorized representatives and make this information available to the IRB at the time of continuing review.

I.

194 ERNEST N. KRAYBILL AND B. SUSAN BAUER

6. The IRB has reviewed and approved informed consent procedures and an informed consent document. . . .These procedures and the informed consent document are to be used with subjects or their legally authorized representatives in situations where use of such procedures and documents is feasible. The IRB has reviewed and approved procedures and information to be used when providing an opportunity for a family member to object to a subject's participation in the clinical investigation. . . . 7. Additional protections of the rights and welfare of the subjects will be provided, including, at least: i) Consultation (including, where appropriate, consultation carried out by the IRB) with representatives of the communities in which the clinical investigation will be conducted and from which the subjects will be drawn; ii) Public diiclosure to the communities in which the clinical investigation will be conducted and from which the subjects will be drawn, prior to initiation of the clinical investigation, of plans for the investigation and its risks and expected benefits; iii) Publit disclosure of sufficient information following completion of the clinical investigation to apprise the community and researchers of the study, including the demographic characteristics of the research population, and its results; iv) Establishment of an independent data monitoring committee to exercise oversight of the clinical investigation; and v) If obtaining informed consent is not feasible and a legally authorized representative is not reasonably available, the investigator has committed, if feasible, to attempting to contact within the therapeutic window the subject's family member who is not a legally authorized representative, and asking whether he or she objects to the subject's participation in the clinical investigation. The investigator will summarize efforts made to contact family members and make this information available to the IRB at the time of continuing review. Additional requirements for approving a waiver of informed consent pertain to notifjring the subject, legally authorized representative, or family, after the fact, about the subject's enrollment in a research study, and seeking permission, from the subject or family, for continued participation of the subject in the research study. DHHS regulations (45 CFRPart 46) and FDA regulations (21 CFRPart yo) are brought into harmony by a DHHS waiver of informed consent for research activities that are subject to the new FDA rule and that have been approved by the IRB according to the rule's emergency research exception (DHHS 1996~).

COMMUNITY CONSULTATION AND INFORMED CONSENT

195

Comment The three ethical principles set forth in the Belmont Report as "generally accepted in our cultural tradition" are respect for persons, beneficence, and justice (National Commission 1978). The rule permitting waiver of informed consent suggests a presumptive "weighting" of these principles. In the lengthy supplemental information that accompanied the notice of the proposed rule, beneficence and justice were claimed in support of the waiver; both principles were given priority over respect for persons. The Belmont Report emphasizes that beneficence is seen as an "obligation." John M. Luce states: "The most compelling principle of medical ethics has always been beneficence: acting to benefit patients by sustaining life, treating illness and relieving pain" (Luce 1990). Traditionally, beneficence has been the basis of the fiduciary relationship existing between the physician and the individual patient. The Belmont Report cites Claude Bernard's extension of this concept from treatment to research, saying that one should not injure a person regardless of the benefits that might come to others. The beneficence claimed in support of the waiver seems directed toward society in general rather than toward individual research subjects, inasmuch as placebo-controlled trials are specifically included as research for which consent may be waived. In the announcement of the proposed waiver, justice was given priority over respect for persons through the following argument. After acknowledging that surrogate consent is less likely to be obtained from families of minorities and persons of lower income, the notice portrayed the more equitable distribution of the burdens and benefits of emergency research that would result from waiving surrogate consent as a desirable consequence-thereby ignoring disproportionately the wishes of poor minority families. In the preamble to the final rule, the FDA responded to public comments critical of this reasoning by arguing that its concern was simply that it is more difficult to locate legally authorized representatives of potential subjects in poor and minority communities. IRBs overseeing research in communities where potential subjects and their families are highly likely to refuse participation are free to refuse to authorize the use of the waiver exception. Nonetheless, research could be approved for the waiver exception by the IRB even though many potential subjects of the research would refuse participation if asked. Despite the FDA's assurances, the principle of respect for persons is still at risk under the final rule. The Belmont Report (National Commission 1978) explains, "Respect for persons incorporates at least two ethical convictions: first, that individuals should be treated as autonomous agents, and, second, that persons with diminished autonomy are entitled to protection." The first ethical conviction, autonomy, underlies

196 ERNEST N. KRAYBILLAND B. SUSAN BAUER

the requirement for obtaining informed consent from those who serve as research subjects. Jay Katz (1993a) writes, "Whenever we use human subjects for the sake of others, using them as means for our ends, the principle of autonomy alone must guide such practice." Luce (1990, 697) further elaborates this principle in the research setting. Informed consent is called for especially when patients are given innovative treatments or are recruited for research. Physicians frequently are tempted to try new therapies in critically ill patients either on a one-time basis or as part of ongoing trials. This is ethically supportable only after consent has been obtained from patients or surrogates or, when consent cannot be obtained, when the treatments are known to be safe and of great potential value. In keeping with the fiduciary relationship, physician obligation is to present-day patients; such patients should not be experimented on solely because future patients might benefit from what has happened to them. The waiver emphasizes the second ethical conviction that is part of respect for persons, namely, that those with diminished autonomy require greater protections. That conviction is acknowledged by requiring that additional protections be provided for subjects unable to give consent. Traditionally, these additional protections have been applied to vulnerable subject populations, such as persons who are cognitively impaired. The application of these protections has tended to focus upon the individual research subject, by making provisions such as the appointment of a consent auditor-an uninvolved third party who can intervene with the investigator on the subject's behalf. The additional protections in the FDA's rule include the requirement that there be consultation with representatives of the communities in which the research will be conducted and from which the subjects will be drawn, and public disclosure about the research before and after it is carried out. Such community consultation would, it is presumed, be designed to inform the members of the community that their next trip to the local emergency room might land them in a research protocol without their consent. Consultation would, it is presumed, also seek to elicit consent from the community, although this is not stated explicitly. In a sense, the investigator would be seeking a form of substituted judgment from members of the community on behalf of the as-yet-unidentified subject. One can imagine focus groups in which the investigator andlor IRB invites participants to consider what they, as reasonable persons, would want to happen if they were critically ill or injured, and were being considered as potential subjects in the study. Respect for persons was also said, in the FDA notice, to be satisfied "if in circumstances of clinical equipoise either the test therapy or its historic alternative is provided, even without specific consent" (DHHS 1996a). The notice quoted

COMMUNITY CONSULTATION AND INFORMED CONSENT

197

(approvingly) testimony presented at the January 1995 FDAINIH Public Forum, saying that clinical equipoise would exist "whenever at least a reasonable minority of medical professionals believe the experimental treatment would be as good as, or better than, the standard treatment." A "reasonable minority" could be construed to consist of the investigators and the sponsor. The preamble to the final rule elaborates on this unusual reasoning, arguing that clinical equipoise is compatible with the anticipation of benefit needed to justiEy proceeding without consent-and thus that such anticipation can outweigh respect for persons in emergency research. The FDA's waiver of consent for the participation of noncompetent subjects in medical research involving more than minimal risk is a sharp departure from current practice and may violate the principle of respect for persons. This is because respect for persons incorporates both convictions: that informed consent must be obtained from subjects or their surrogates, and that there must be additional protections for vulnerable subjects. The waiver satisfies the second conviction but not the first. Efforts to shore up this weakness include invoking other ethical principles (beneficence and justice) but rely mainly on the substitution of community consultation and disclosure for individual autonomy. Apart from these difficult questions about the ethical basis for the waiver of consent in emergency medicine research, a number of practical questions arise: Will the rule establish a precedent that principles may be shaped to accommodate research considered so important that it must be done (the slippery slope), or are there some kinds of research that are impossible to carry out because of ethical considerations? 2. In Western, individualistic societies, can community judgment effectively replace individual consent? Will individual subjects andlor their families accept having unconscious individuals become unwitting and involuntary research subjects on the basis of "community approval"? 3. Will the international research community accept the waiver as within the principles enunciated in the Nuremberg Code ("The voluntary consent of the human subject is absolutely essential"), the Helsinki Declaration ("The physician should then obtain the subject's freely given informed consent, preferably in writing"), and the Belmont Report ("Respect for the . . . incapacitated may require protecting them while they are incapacitated. Some persons are in need of extensive protection, even to the point of excluding them from activities which may harm them")? 4. What will be the final (research) status of an individual, enrolled in an emergency research protocol under the waiver, for whom no legally authorized representative is ever found, and who never regains consciousness?

I.

198

ERNEST N. KRAYBILL AND B. SUSAN BAUER

5. How will investigators deal with subjectslfamilieswho not only refuse continuing participation but object to the individual's already having been made a research subject? 6. Will investigators, IRBs, and their parent institutions be potentially liable for injuries caused by research procedures for which there has been no prior consent? 7. How can the "community" be defined, relative to individual emergency research projects? 8. How will the legitimate representatives of that "community" be identified? 9. Are consultation and disclosure sufficient to satisfy the requirements of the rule, or must community agreement be obtained? How will such agreement be defined? If community agreement is not a requirement, how does community consultation and disclosure constitute "additional protection" of subjects? 10.Can I M s , as currently constituted, fulfill (through their existing "community members") the requirement of community consultation and disclosure?

200

J.PAT BROWDER

answer in the Nuremberg trials: There are some kinds of research that should be prohibited on ethical gounds. The legacy of the Nazi doctors, and our own sad history of the Tuskegee syphilis experiment and the Department of Defense radiation studies, provide concrete examples of such research, which had two characteristics that justified condemnation: (I) the research, while it might have been of benefit to society, caused harm to the subjects with no commensurate benefit to the individual subject; and ( 2 ) the research was conducted without informed consent. According to Jay Katz (1993a), it is the second of these two characteristics that is the more important to avoid: "Whenever human subjects of research serve the ends of others, the morality of protecting their physical integrity, which continues to preoccupy the thinking of physician-investigators, is not the central issue. It is, first of all, the rnoralzty with which the invitation to participation in research

is extended so that the rights of subjects to be secure in their person and body remain sacrosanct" (Katz's emphasis). Katz acknowledges that the doctrine of informed consent, borrowed from the law of torts, is not easily transplanted into therapeutic settings, much less research settings, at least not without a thorough reappraisal of its function. More important than the legal doctrine of informed consent, however, is the idea of informed consent, that physicians and patients, investigators and subjects, must make decisions jointly, with patients' and subjects' choices ultimately receiving the utmost respect. Without implementation of this idea, informed consent becomes a charade, especially in the arena of clinical research. Katz bases his plea for respect for persons in part on a premise articulated by Daniel Callahan (1984), that it is "good as a bodyguard against moral bullies," and as such can protect patient-subjects from coercion by investigators too invested in the morality of their scientific pursuits. The practice of medicine and the conduct of research are different enterprises, a distinction that Katz feels is too often lost on Congress and federal regulators. Policymakers, he says, must confront their responsibility to formulate a distinctly separate societal mandate for the conduct of research that specifies the lengths to which a democratic society like ours can go in compromising citizen-subjects' rights to autonomy and physical integrity for the sake of medical science. To that end, in the wake of the awful disclosures of the Tuskegee study, Katz proposed that Congress establish a permanent body, a National Human Investigation Board (NHIB), with the authority to regulate all federally supported research involving human subjects. T h e proposal was incorporated into a bill submitted by Senator Edward Kennedy to the Senate but never passed. Katz believes that one of the reasons for this failure is that the Senate was reluctant to expose to public view the value conflicts inherent in the conduct of research-specifically, the problem ofwhen, if ever, inadequately informed subjects can serve as a means to society's and science's ends.

COMMUNITY CONSULTATION AND INFORMED CONSENT

201

These value conflicts are the subject of the last two questions posed by Kraybill and Bauer that I wish to consider: In Western, individualistic societies, can community judgment effectively replace individual consent? And can IRBs, as presently constituted, fulfill the requirements of community consultation and disclosure? Each of these questions begs the prior question of what is the proper way, in a democratic society, to resolve issues in which promotion of the welfare of the society is sought at the expense of the welfare of the individual. The new regulations offer two mechanisms for resolving the issue at hand, community consultation and public disclosure, both to fall under the auspices of individual IRBs (Department of Health and Human Services 1995~). The specific language calls for (I) consultation with "representatives of the communities" in which the research will be conducted and from which the subjects will be drawn, (2) public disclosure prior to beginning the study "of plans for the investigation and its risks and expected benefits," and (3) public disclosure after the study is completed "to apprise the community and researchers of the study, including the demographic characteristics of the research population, and its results" (21 CFR %5o.q(a)(7)).In addition it calls for the establishment of an independent data monitoring committee. I have serious problems with the ethical and political theory that informs this language as well as with its practical implementation. Its use of the word "community" is emblematic of both these problems, in that it presumes to assign univocal meaning to a word whose usage in Western society is theoretically ambiguous and full of practical conflict (Mariner 1995). To put it concisely, in our pluralistic society there is no "community" to consult as these regulations propose, at least no community constituted to resolve the kind of fundamental ethical conflict that these regulations propose to deal with. The premise of the democratic process is that it serves a plurality, not a community, and attempts through procedures agreed upon in advance to be fair to form policy binding on the whole in the absence of community. To speak of community as these regulations do is to assume the existence of a sociopolitical entity that may exist in certain sectarian enclaves, such as the more conservative Christian denominations, but evaporates when it is forced to encompass a society whose credo is individualism and cultural pluralism. It is completely specious to pretend that the new regulations involve nothing more than an attempt to resolve a bureaucratic squabble between the DHHS and the FDA, as was suggested in earlier criticism of regulating research in CPR (Olson 1994). It is equally specious to suggest, as do Biros et al. (1995) and Levine (1995), that what is at stake here is the well-being of individuals who are being deprived of the benefits of participation in emergency research. While it is undoubtedly true that future individuals might benefit from research on emergency medical conditions, it is in no sense true that the individual subjects of the research - -

202

J.PAT BROWDER

would benefit, since by definition the benefits of experimental therapy are unknown and may very well carry considerable risk beyond that afforded by standard therapy. There is no benefit to participating in properly constituted therapeutic research that can exceed that of standard therapy, other than the rewards of altruism-at least no benefit that could justify, without very careful consideration, forgoing informed consent. What is at stake here is something of geater importance than bureaucratic obstruction or even the purported benefits of participating in research. What is really at issue here is the process by which our democratic society decides when it is appropriate to compel individuals to sacrifice some of their own welfare for the benefit of the whole society. What should the process be, in this instance: to compel individuals to risk their well-being in a clinical trial, without their consent, so that others in the future may benefit from what is learned? One answer could be to take a poll. If ninety-five percent or more of potential research subjects or their families say they would consent to an experimental therapy if given the chance, as a recent study of the relatives of head injury patients discovered (Abramson and Safar 1990), then it can be assumed in the vast majority of these cases that failure to offer them the experimental therapy would be contrary to their wishes and do them an injustice. The problem with this approach is that it fundamentally misunderstands the democratic process as it applies to conflicts ofwelfare between the individual and society. The issue to be decided is not what the majority wants; if that were the case, then only fifty-one percent would be needed to decide the issue. The issue, instead, is when it is acceptable to benefit the majority, no matter how great, at the expense of a minority, no matter how small. No poll can decide that. Our society does have established mechanisms for deciding this kind of issue. The closest analogue to the case at hand is the procedure for compulsory military service, the draft. It is particularly apt given the role that the regulations propose for individual IRBs (Manvick 1995). If these regulations were applied, not to emergency research, but to compulsory military service, then individual draft boards would be empowered to authorize, in specific instances, the imposition of the draft. And while this could be made a very open process with full disclosure of the risks and benefits of the draft, including consultation with representatives of the groups most likely to be subject to its compulsion, it is ludicrous to think that our society would entrust to a local body with its necessarily limited perspective a decision of such fundamental importance. To say that IRBs are not qualified to undertake this role is not to disparage the work they do or the dedication of those who serve in them. I do not share the cynical opinion of some that IRBs as currently constituted do not protect research subjects as much as they protect the institution and the institution's investigators (Annas 1988). It is not a question of forbidding the fox to guard the chicken house; -

COMMUNITY CONSULTATION AND INFORMED CONSENT

203

individual IRBs are very capable at their appointed task of protecting the rights of human subjects in accordance with the fundamental ethical principles of biomedical research. The issue here is much more difficult, involving not the application of ethical principles to research in emergency conditions, but the far more profound task of deciding whether to violate one principle for the sake of other principles. It is, to use Katz's phraseology, the critical question of whether the rights of subjects to be secure in their person and body are to remain sacrosanct, or whether this right, in this instance, is to be sacrificed for reasons of beneficence and justice. No individual IRB can or should be asked to decide this. Kraybill and Bauer have done us the service of pointing out the critical issues and asking the right questions. The answers that I have offered compel me to reject the FDA regulations, and instead to suggest the formation of a national review board along the lines suggested by Katz. I, as an individual IRB member with more than fifteen years of service in that role, could not in good conscience agree to take part in a process such as these regulations propose. If a case that fell under the terms of these regulations were to come before me, I would have to abstain, since I could not consider myself qualified to make the kind of decision called for by the regulations. I have a very strong feeling I would not be alone. Those who feel strongly that this kind of research should be permitted to go forward are not doing themselves a service by insisting on this kind of regulatory authority for IRBs, since in fact it authorizes IRBs, for "relevant ethical concerns" other than those stipulated in the regulations, not to approve the research (21 CFR $50.24(e)). They may wish to take their chances with the consciences of individual IRB members, but they would be far better off taking this debate where it belongs-to the Congress and to a national review board, who would have the real authority to decide the fate of this research in keeping with accepted democratic processes. Doing so would not preempt the authority of local IRBs. Assuming that a national review board were to approve the waiver of informed consent for certain kinds of emergency research, individual protocols would still have to pass IRB scrutiny to see if they met the criteria stipulated by the national board. And while I might still have qualms of conscience about approving research protocols of this kind, at least I would know that the decision before me as an IRB member was within my proper area of competence, since the larger question of how to balance these fundamental ethical principles would have been decided. I would also have had the benefit of listening to and even participating in the public discussion of this issue and would come to my decision better informed and, I hope, well prepared to accept its consequences. As it stands now, that discussion has not even started.

MEDICAL RESEARCH

205

and Human Services 1991b). Informed consent may be waived by IRBs under appropriate circumstances. What constitute appropriate circumstances? Minimal risk to subjects in the course of necessary research that cannot otherwise be carried out-in other words, circumstances are appropriate when risks and benefits are balanced and benefits come out ahead. The new rule simply extends the riskbenefit calculus to circumstances of minimal relative risk, when there is no definitively effective treatment for a serious condition and no reason to believe, from what is known, that the intervention to be tested will be significantly riskier than standard treatment. Of course, as Kraybill, Bauer, and Browder have pointed out, this formulation begs a number of questions, such as how the need for the research is determined, how benefit is projected, and how uncertainty, lack of information, and the possibility of unanticipated risks are factored in. Rut all this is secondary to the central matter: whereas the Nuremberg Code held that the voluntary consent of the subject is absolutely essential in every instance, every set of precepts related to research ethics since then has come to a different conclusion. Both the federal regulations and the Declaration of Helsinki permit consent from someone other than the subject, namely a "legally authorized representative," even though this is not the same as consent from the subject. The federal regulations currently permit participation without subjects' consent in research of minimal risk that holds the promise of societal benefit, thus weighing both the risk of harm and the prospect of societal benefit as more important than the wrong of failure to respect autonomy. Why has only the N u r e m b q Code held the line on informed consent? Only the code takes a position of skepticism regarding the underlying moral justification for research (Glantz 1992). All other sets of moral and policy precepts affecting research have accepted the proposition that research is a beneficial enterprise, and few if any systematic moral justifications for research have even been offered (Annas 1996). Nonetheless, the possibility of abrogating informed consent can only arise where there has been a weighing of research against autonomy, and research has won out. In this regard, it is worth mention that in recent years both the magnitude of the research enterprise and public confidence in the beneficence of research have grown considerably. Whereas research used to be viewed as something from which potential subjects needed protection, nowadays persons suffering from a variety of conditions clamor for quicker access to untested interventions (Annas 1989; Levine 1994). Most significantly, when patients become subjects it becomes difficult to separate treatment from research, and to separate intent to benefit future patients from intent to benefit current patient /subjects-especially when there is no good standard therapy available. When these confusions crop up, subjects are thought of as patients, and exceptions to the informed consent requirement become easier

206

NANCY M. P. KING

to justify on the ~aternalisticbasis of beneficence (King 1995; King and Henderson 1991). This last point unearths a crucial justification for abandoning informed consent in the proposed rule: it is popularly believed that biomedical research is different from other categories of research and therefore deserving of different treatment. O n the one hand, some have argued that research that does not put subjects at risk of physical or psychological harm, but only risks wronging subjects through violation of their rights-for example, much social science research-does not require the same level of protections for subjects that biomedical research does. This argument would make informed consent more important in biomedical research than in social science research, to protect subjects from harm. However, others have made an opposite argument about the difference between biomedical and other research. This argument holds that biomedical research carries with it significantly greater potential for benefiting both society and, ultimately, its individual members. Thus, especially where the condition under study has no satisfactory standard treatment, biomedical research should be made more readily available to patient-subjects who might conceivably benefit from it. To ensure that availability would mean to relax protections for subjects of biomedical research more than for subjects of social science research-protections like informed consent. Is biomedical research different? Proponents of the consent waiver rule would probably argue that emergency medicine research is indeed different, by reason of the lack of time to obtain consent in emergencies and the lack of effective treatment for the conditions that could be studied if the informed consent requirement were waived. This research is also different in some other respects, however, such as in the need to enroll large numbers ofsubjects relatively quickly in order to qualify for the large amounts of funding being made available for such studies by the manufacturers of drugs and devices, and in the increasing dependence of large academic medical centers on research grant income (Scofield 1995). It appears also to have been assumed that only biomedical research can be considered potentially beneficial to subjects; the image of medicine as working miracles has certainly influenced current popular views about the value of medical research. Nonetheless, many research interventions that provide subjects or their communities with benefits of some sort as a quid pro quo for research participation thereby offer much greater benefit to subjects than the chance of benefit from the tested intervention. The role confusion that occurs when researchers are also service providers thus occurs in both biomedical and nonbiomedical research. It seems fairly clear how we came to the possibility of dispensing with informed consent in emergency research. It also seems clear that the same tools can be employed in support of the possibility and against it, using the current prin-

MEDICAL RESEARCH

207

ciplist paradigm. But can a relationships paradigm tip the balance of the argument either way?

The Consent Relationship

A patient who is brought to the Emergency Department (ED) after an injury comes into a setting where the physician-patient relationship, as we often idealize it, does not fit at all. A sudden, unpredictable event has precipitated the encounter with physicians; even patients who are accustomed to long-term physician-patient relationships are unlikely to have preexisting relationships with ED physicians; and, although persons who are seriously injured may well develop new relationships with physicians as a result, those will begin on an inpatient service rather than in the ED. Thus, it is quite easy to view the patient's time in the ED as a slice of time. Human subjects research-biomedical or otherwise-that takes a slice-oftime perspective is somewhat resistant to a relationships-based perspective. A great deal of traditional biomedical research fits this description: subjects show up, are put through their paces, get paid, and leave. The encounter is transient and no relationship is built, let alone maintained. Still, there are other ways of looking at even these transient research encounters, and it can be persuasively argued that our perceptions about biomedical research are changing. First, even if the researcher-subject encounter is indeed a slice of time, in many studies the data obtained from subjects are stored and used for many years. The storage of information of many types, from abstracted medical records data to blood and tissue samples, for long periods gives rise to many questions about ownership, control, and subsequent use, and suggests the need for a relationships paradigm in order to take account of the dilemmas that may arise in the future. Second, as information begins to be understood by the public at large as carrying more and more "social r i s k (as mentioned in the Blanchard case), even biomedical researchers who attempt to protect themselves from developing relationships with subjects may have to contend with relationship issues. Many studies may uncover information that suggests the need for medical or other follow-up. A standard blood test or blood pressure test may reveal the risk of diabetes or hypertension, or the presence of HIV. A standard medical screening interview may raise the possibility of family violence or suicidal ideation. A search for a "cancer gene" may reveal inconclusive information that could affect subjects' future health care decisions (or even their employability or insurability). Researchers may feel relational obligations to provide such information to subjects, or to oth-

208

NANCY M. P. KING

ers on their behalf, even in a study designed to discourage disclosure or make it impossible. It is thus becoming clear that biomedical research is deeply embedded in the relationships of social and professional life, and that even so-called "hard data" depend upon those relationships for context and meaning. The apparently straightforward application of the traditional paradigm of research regulation represented by the federal Common Rule can be made more meaningful, and more responsive to the felt needs of both subjects and researchers, if the relationships paradigm is also considered.

Relationships in Emergency Research Questions about the balance between risks and benefits in research, the need for information that can only be gathered through research, and the entanglement of research with the provision of services are as pressing in traditional biomedical research as they are in other research contexts. In the many instances where researchers need data about treatments for serious medical conditions, questions about relationships can help to highlight the issues. For this problem of consent in emergency research, there are four relevant questions: I. Should researchers view themselves as doctors helping patients or as investigators enrolling subjects, and can they be both? The federal regulations and the Belmont Report distinguish research from treat-

ment by addressing the primary goal of the intervention and the likelihood of its success. "Research" has as its goal the accumulation of generalizable knowledge; "treatment" is intended to benefit its recipient and must have a reasonable likelihood of succeeding (Department of Health and Human Services 1991a; Levine 1988, 3; National Commission 1978). The problem is that these goals often overlap in health care, and the overlap is often compounded by the possibility that what is a "reasonable" chance of success may be very small when no effective accepted treatment exists for a condition (Katz 1993b).Emergency research is a perfect example of this overlap and confusion, where at least some physicians and patients would consider the research intervention to be the best available therapy, despite the paucity of available data about its risks and success. If it is possible for researchers to see themselves simultaneously as physicians, it is necessary for them to have a similarly layered relationship with their patientsubjects. The question then is whether the layers blend well, or mix like oil and water. And the risk is that when these two pairs of roles are blended, the patient part and the physician part tend to take precedence-privileging what the doctor thinks best over the subject's free choice, or setting data-gathering above the pa-

MEDICAL RESEARCH

209

tient's treatment needs, or even making both kinds of errors at once. My fear is that emergency research carries all of these risks.

I f a project?funding depends on the rapid enrollment of subjects (with targets set in order to meet drug approval deadlines,for example), when does the relationship with funders compromise the relationship with subjects and their rights and needs? 2.

Emergency research, including the protocols that helped to generate the new rule for waiver of informed consent, is unquestionably important. It is also a big money matter, with large grants, great potential for income from new drugs and devices, and the need, therefore, to enroll substantial numbers of subjects. The informed consent requirement does not prevent researchers from conducting emergency research-it merely precludes the rapid enrollment of enough subjects to provide statistical power for the results. When patient-subjects are unable to provide consent, and family members must be contacted about study enrollment, more time is generally needed than for initiating emergency treatment. If there is a "window of opportunity" after which initiating the research intervention will not be useful, some subjects will be lost. (Even when family decisionmakers are available, some will say no to the inclusion in research. The decision has more than one component, after all-enrollment is not just the opportunity to try an unproven intervention. It also means being a subject, which itself is both a contribution and a burden.) If the reality that completing enrollment will take longer, and that some people will say no, fuels the decision to forgo consent, we need to think again about which relationships are most important, and about whether a researcher-subject relationship that is created unilaterally, without even the chance to say no, is a relationship at all. 3. What are the characteristics of rehtionships between physician-researchers and

potential subjects that affect deliberation about the needfor informed consent? The long-standing rule that consent to treatment may be forgone when necessary in emergencies is based upon an assumption that the physician and the patient have common goals for treatment, and on the assumption of trust between them. Society assumes, in emergencies, that people want to be treated in order to preserve their lives, relieve their pain, and prevent further harm. That assumption has limits. Patients may refuse treatment. We all know that Jehovah's Witnesses often refuse treatments involving blood transfusions, and that some people, especially those who are seriously ill, refuse cardiopulmonary resuscitation by means of do-not-resuscitate orders. Patients also must be involved in the decisionmaking process, either directly or through family members able to speak for them, as soon as it is feasible to do so. Whether the assumptions of common goals and mutual trust are justified in the context of emergency research is, at best, a very open question. In emergency

210 NANCY M. P. KING

research, physician-researchers and patient-subjects do indeed share one goalhealing-but many patient-subjects and their families may have only a vague understanding ofwhat medical research is and what its goals are. We cannot assume altruism on the part of patient-subjects. Nor can we assume that self-interest for patient-subjects and beneficence for physician-researchers go hand in handunless we afford patient-subjects, or their families, the opportunity to examine the dual goals of emergency research (that is, data-gathering and healing) and weigh them for themselves. Moreover, trust is something that physicians, too used to assuming, forget that they need to earn. As Kraybill and Bauer discuss, in early discussion of the need for the emergency research rule, the FDA noted that the groups most likely to refuse to participate in medical research are the poor and minorities, especially African Americans. It is important to include members of these populations in research and to ensure that they share in the fruits of research; proponents of the rule apparently reasoned that the best way to ensure their enrollment was to disenfranchise them. From a relationships-based perspective this reasoning is preposterous, and, as Kraybill and Bauer point out, the FDA to its credit backed away from it in its discussion of the final rule. Minorities and the poor have real and justified concerns about their relationships with the health care system and the medical research establishment-concerns grounded in history as well as in current experience. Many members of poor and disadvantaged communities have as their principal source of health care the urban teaching hospital: short on amenities, long on waiting lines, and especially long on eager, inexpert physicians-in-training, with extra questioning and extra procedures seen as part of the price of admission. A relationships perspective would focus instead upon the need to build relationships with minorities and the poor, through education about the research enterprise and frank admission of the need to establish trust. Persuading families that emergency research might in fact serve the needs of their injured members with a minimum of burden and risk means beginning a relationship, with individual patient-subjects and their families as well as with communities of potential subjects. It is far preferable to avoiding one. 4. What is the relationship between the community consultation andpublic disclosure requirements and the waiver of informed consent? The community consultation and public disclosure requirements, vague as they currently are, are intended somehow to substitute notice to relevant communities for consent from individuals or their substitute decisionmakers. Whatever ultimately is held to satisfy these requirements, two fundamental problems will remain.

MEDICAL RESEARCH

211

First, what in the world are the relevant communities for emergency research? Browder has already eloquently addressed this question from a philosophical perspective. My perspective, responding to Kraybill and Bauer, is more practical. Should we look to the people who live near the hospital? O r to the people who live in the hospital's catchment area (which for some hospitals might include several states)? O r perhaps to the people who are likely to suffer the injuries of interest in a given study-say, head injuries? What if the hospital emergency department in South Carolina or Georgia that is undertaking head injury research receives most ofits admissions from highway accidents on Interstate 95: New Yorkers who come to gief on their way to winter in Florida. What communities represent them? Second, even ifwe can pick the right community and find the right representative members to notify, and even ifwe were to (wrongly) assume that consultation and disclosure are somehow the same as "community consent," why should we presume that communities can or should speak for their members? In international and cross-cultural research, it is often assumed that individual informed consent is at odds with social practices in societies less developed and more oriented toward family and community than our own. It is often suggested that village elders should consent on behalf of their communities to the involvement of individual members in research. These assumptions about the relationships between communities and their members have been forcefully disproved, however. While consultation with representatives of the community is an important and often indispensable prerequisite to individual informed consent, in most cultures it in no way replaces informed consent. After reading Carel Ijsselmuiden and Ruth Faden's powerful argument to this effect about research in Africa (199z), I had the opportunity to ask the question of epidemiological researchers at an international meeting. They had read most of the cases in this volume, and when I asked them about community leaders and individual informed consent, I received replies from researchers who had enrolled subjects from rural and urban areas in the Philippines, Thailand, Burma, China, South America, and all over Africa. All of them said that they needed the village leader's consent to talk with individuals-because that gave them the credibility they needed to begin the informed consent process with individual community members-but all individuals in those communities expected to make their own decisions and, in every respect, did so. Moreover, especially in urban areas, diversity and mobility often made it less likely that those identified as community leaders held authority in the eyes of all or most of the community's members. The influence of community leaders and the importance of consultation with spouse and family is thus undeniable, and in some settings in developing coun-

212

NANCY M. P. KING

tries it may be considerable. But it provides the context for the individual's choice, while in no way substituting for the choice itself. If this is true in the developing world, how can it be any less true in the many communities to which each of us in the United States belongs?

Conclusion: Using the New Paradigm

I do not like the consent waiver rule any better than do Kraybill, Bauer, and Browder, and I think the arguments are all on our side. But I recognize that the current principlist paradigm, as it is played out in the federal regulations themselves, also supports the waiver of informed consent in some research. This is at least in part because the current paradigm permits a slice-of-time examination of an atomistic encounter between patient-subject and physician-researcher. When we use the new paradigm, simply by widening the frame of our point of view, it includes a lot that could change the picture. Now we have to consider the encounter in the emergency department as the beginning of a set of relationships with researchers and health care providers. We have to see the research project as embedded in a nest of relationships, not only between researchers and subjects, but between researchers and communities of subjects, researchers and the institutions that sponsor and fund the research, researchers and their institutional homes and academic peers who will judge the results of their research. All of these relationships affect the argument about the appropriateness of the rule, but only in the relationships paradigm are they openly considered. Only the relationships paradigm recognizes the need to rank and prioritize relationships and to consider their effect on each other. To be sure, adding all these new elements to what has up to now been a rather simple picture can make it harder to focus on what is really important. It forces us to name those elements and argue about their significance, instead of allowing us to push them out of view. In particular, in the case of emergency research, the relationships paradigm raises issues about differences in power and viewpoint between researchers, funders, and potential subjects. These extraordinarily difficult issues have been hinted at throughout most of the recent history of human subjects research, and trumpeted by an articulate and vocal few, such as Jay Katz, who have remained in the minority for many years. The social changes of the last several decades, from civil rights to political correctness, have brought questions about poverty, power, and perspective to the table but have not improved our ability to address them, in the health care context or otherwise. But we are struggling more openly with those questions these days, and as our paradigm shift continues, I hope we may get closer to some answers.

214

CONCLUSION

mentaries in this book. They also have seen, however, that the authors who do not attempt to use a relationships paradigm do not all use principles, either-and rarely focus on the regulations. Instead, the moral reasoning used in this bookas in contemporary research ethics and bioethics generally-is eclectic. Has this book made a case for relationships over principles? No; attempting to do so is probably premature. Principles and regulations are not enough. Sophisticated principlism, however, does add contextual seasoning-sometimes a great deal of it-to stone soup, and especially thoughtful principlists like Ruth Macklin thereby recognize the need to bring to light many of the issues that the relationships paradigm places in the foreground. When it is used well, the principlist paradigm can be rich and expansive; but even when expertly employed, it does not obviate the need for the relationships paradigm. By emphasizing connections, differences in viewpoint, and narrative perspectives, the relationships paradigm provides access to dimensions of moral reasoning that are relatively underemphasized in the principlist paradigm. When used to complement and expand principlism, the relationships paradigm most closely resembles some of the newer ethical perspectives, like narrative and feminist ethics. But when it is fully developed and well applied, the relationships paradigm is normative in its own right, standing independent of principlism as a model of moral reasoning. Whether, so employed, it should render principlism wholly unnecessary is far from settled; moral reasoning may be irreducibly eclectic, and it may be desirable always to employ and emphasize different models for different needs and circumstances. If this is so, the principlist paradigm may always be needed, but should never be employed in a way that excludes or subordinates other modes of moral reasoning. A brief look back at the cases may serve to illustrate how the relationships par. . adigm can serve as an especially promising means of opening up moral argument about difficult issues. (A great deal more could be said about these cases, individually and collectively, but our hope for this book is that readers will be doing this for themselves.)

The Cases Revisited Recall the relationships table in Sue Estroff's commentary to Case I. The relationships paradigm does not tell field researchers like Alan Benjamin how to prioritize among the relationships with one's peers, one's field, and one's subjects. It does, however, identify the demands of those potentially competing relationships, so that they are included in consideration of the moral aspects of the research. If we start from the recognition that a balance must be found among these relation-

REGULATIONS AND RELATIONSHIPS

215

ships, Rente Fox's argument to protect academic freedom represents one way to balance them in this case. Benjamin is acutely aware of differences between his sense of his own power at the bottom of the pecking order of academia and his subjects' sense of his power as part of that academic pecking order. The fact that Benjamin's dissertation was readily approved by the Directiva, and that, as he predicted, he has thus not been at any real risk of violating the contract, could be viewed as either a good relationship (as he and Estroff might characterize it) or self-censorship (as Fox might have it). This discrepancy in perceptions about power is a significant issue, which comes up again in a different form in Case 2. Case 2 adds two issues to the analysis of research relationships presented in Case I: (I) the question of representativeness (the Directiva had acknowledged authority to represent the congregation; the CAB'S authority and role are much more ambiguous); and (2) the need to include funders and sponsors in the balancing of relationships. Adding the latter to our calculation amplifies the prioritization problem this case shares with Case I. We could examine the role of the CAB in terms of its relationships, in a fashion analogous to Estroff's relationships table, by observing that the CAB could understand itself as in one of two primary relationships: I.

2.

Relationship with the researcher(s). Relationship with community and potential subjects.

Each of these potentially serves different functions. If the CAB sees itself as primarily in relationship with the researchers, then it is likely to see itself as having responsibilities to help the researchers to properly follow the federal regulations. Attending to the rights and interests of subjects is crucial, but not primary, as it derives from the requirements of the regulations. In contrast, if the CAB sees itself as primarily in relationship with subjects and the community, it has a representative function and responsibility-representing the community to and for the researchers. Helping the researchers speak effectively with the community would, on this view, derive from the CAB'S duties to the community. The CAB'S role is thus subtly different depending on which relationship is considered primary. Taking a step beyond the first two cases, Case 3 shifts the primary relationship focus to that between researchers and their funders/sponsors. It is tempting to apply standard conflict-of-interest analysis to a case like this; Popkin points out that the relationships paradigm shows that conflicts of interest and other funding related issues arise with all funders, including governmental funding and peer review of research proposals. Although conflicts of interest arise in all research, it is especially important to

216

CONCLUSION

address them in this case by including consideration of the researchers' relationships with their human subjects. For what purposes may subjects be used, and to what purposes may they consent? The disclosure model of handling conflicts of interest applies not just to disclosure to editors and readers of publications, but also to the informed consent process. Yet is mere disclosure sufficient to address potential conflicts, if they are as subtle and far-reaching as Brandt and Freidenfelds suggest in their commentary? Should the relationships paradigm stretch as far as they propose, to constrain individual researchers because of the harms that could result from the aggregation of many future individual choices? The authors of Case 4 describe themselves as advocates for women's reproductive health. Their sincerity is unassailable and their history of social activism is impressive. We might then ask whether the researcher-subject relationship in Case 4 is an example of the advocacy model that Loretta Kopelman describes in her commentary on Case 3. Case 4 raises questions like those raised in Case I about the need for researchers to weigh potentially competing relationships. The researchers felt significant obligations to a social and political constituency much broader than the groups of subjects they studied. Here, though, the influence ofpower and class in the researcher-subject relationship is, with respect to the women subjects, more like the relationships in Cases 2 and 3 than in Case I. What if (as the commentators suggest) the researchers had decided to set up a CAB for this study?Who should be on it? Does the researchers' sense of activist long-term commitment to women and reproductive health issues outweigh, or at least reframe, their relationship with the actual subjects here (who were not clearly a "community")? Can anything be finally said when the subjects themselvesboth women patients and hospital personnel-are so imperfectly heard? Should we ask, as in Case 3, for what purposes subjects may be used? O r should we go further, and require the researchers to provide subjects with the information they need to allow them to collaborate in the research's purposes, even if that would require substantial alterations in the research design or even make it very difficult to recruit subjects? This last concern leads directly into Case 5 and the following question: Does the relationships paradigm require us to examine the relationship of research to the socjety that supports it? In Case 5, the researcher states that the LONGSCAN project researchers had first planned to wait until the child subjects turned eighteen to collect direct data about abuse they suffered as children, in order to avoid the reporting problem; but researchers and clinicians told them that the data would be too important to wait for that long. Is it in society's interest to modify consent to research in order to obtain valuable data more rapidly? This significant public policy question also relates directly to the importance of the data-gathering in Case 6. The principlist paradigm clearly supports exceptions to the informed consent requirement in research, though such exceptions

REGULATIONS AND RELATIONSHIPS 217

are nonetheless controversial. In contrast, because the relationships paradigm assumes the desirability of some degree of relationship between researchers and subjects, it may be that this model simply cannot support the social judgment that subjects may be enrolled in research without an adequate informed consent process, regardless of the need for results. To circumvent this conclusion, one would need an informed social judgment that such research is necessary-a sort of advance directive from subjects-which would then permit emergency research on those terms.

Blending Paradigms: Can They Coexist? The fundamental differences between the principlisr paradigm and the relationships paradigm are simple. As the outline in the introduction shows, these two paradigms invite those faced with questions or problems of research ethics to ask different kinds of questions and to consider and weigh different information in analyzing issues. As can be seen from the cases and commentaries, these kinds of differences in inquiry can give rise to subtly different analyses and perhaps therefore to changed priorities in thinking about the ethics of human subjects research. The questions become more complex and the problems are acknowledged as more troubling, as narrative, context, and power come to be more readily included in moral analysis. In some instances, shifting paradigms might lead to substantial differences in both analysis and conclusions, but in other instances the differences may be less dramatic. People can reach agreement about what the right thing to do is while disagreeing significantlyabout why it is right (AdvisoryCommittee 1995, chap. 4). Thus, we should expect to see differences in the process of analysis when the new relationships paradigm is used, without necessarily expecting differences in the products or outcomes. However, if using the relationships paradigm only affected the process but never the result, we would have to be concerned that the new paradigm was being subordinated to the old. It therefore remains to be seen whether these two paradigms can coexist effectively, either temporarily or permanently, as suggested by the figure in the introduction. It is easy to see that principlisrs would tend to resist including the strong contextual focus of a relationships paradigm because they fear it will result in relativism; likewise, proponents of the relationships paradigm would tend to resist strong principlism, believing it to lead to moral imperialism, paternalism, and absolutism. Perhaps the worthiest aim, at least for now, is to further both a relationship-oriented discussion of principles and a principle-oriented discussion of relationships in the ethics of research with human subjects. But if each paradigm is independently normative, it may be difficult to blend them in a way that

218

CONCLUSION

preserves the moral authority of both. The cases provide ample illustration that single-paradigm analyses do not serve us nearly as well as using two paradigms. The challenge is to use each paradigm appropriately and both paradigms together. The special value of principles in ethics is that they perform an adjudicatory function. We use them in moral analysis to justify, or to condemn, a course of action; thus, we use them to judge and regulate research. Principlism is reasonably challenged by the force of the accretion of context. Yet principlist approaches to ethics find it difficult to give place to competing systems of moral theory, like virtue ethics, narrative, and casuistry, and to determine whether and how they too can be used to judge and govern moral behavior. Perhaps as a consequence of this very difficulty, part of what is currently contested in human subjects research is the nature of adjudication itself. As we learned in our conference, the problem of adjudication raises several important issues. No one questions the need to monitor and oversee-to adjudicate-the conduct of research. Nonetheless, the regulatory, principlist paradigm can be challenged in at least three ways. First, when principles are primary, they stand superior to other systems of moral theory. Whether such systems-universal or local, competing or contextdependent-should replace or simply supplement principlism appears to be an open question, as is how such changes would in fact alter how research is considered and conducted. Second, because principlism is still ascendant, when different models are put forward as ways of understanding a problem, one "winsn-and one "loses." In our first case, the validity of the research (as adjudicated "scientifically" by the researcher's academic peers) was questioned, as was the integrity of the researcher and the researcher's trust in subjects (a non-adjudicatory virtue ethics paradigm). Thus, when a scientific expertise model is offered along with a virtue-based model for examining this case, validity is placed in opposition to integrity and trust. This adjudicatory opposition is an ~nsatisf~ingly limited viewpoint, apparently dooming the results of genuine collaboration between researchers and subjects to the status of second-class data. If the principlist paradigm implies that relationships in research are vertical, then the researcher-subject relationship can only be seen as either dominated by researchers, or dominated by subjects-not as a horizontal relationship of equals. Finally, when the language that makes common discourse possible is adjudicatory in nature, language itself makes horizontal, collaborative relationships difficult, perhaps impossible. The principlist paradigm might be said to be inherently colonial, and the colonized have always seen the imposition of a common language as an imposition of power. This is no less true when the colonizers genuinely believe themselves to be leveling the playing field. One of the pervasive concerns of this book is the use of language: what we understand in common;

REGULATIONS AND RELATIONSHIPS

219

what we thinkwe share but perhaps do not; the poverty of language when it comes to capturing ideas that are changing and new; and the ways in which language depends upon and reinforces the social, historical, and political contexts that shape human endeavors like research.

Examining Research We use language to speak to each other, but also, necessarily, to talk to ourselves. One of the most difficult and rewarding things that scholars or scientists can do is turn the critical gaze upon themselves. Part of the work of this book is therefore to critically investigate the research enterprise itself. This task leads in several important directions. First, using the relationships paradigm to consider research as a process of knowledge production-particularly as a process that is appropriately examined in terms of power-highlights many relationships that have not often been in the foreground of discussion, such as those between researchers and their funders, their institutions, their audiences, their peers. Once these relationships are featured, as they are in many of the cases in this volume, we recognize that they have different weights. For example, Professor Balin's research (Case 4) has been closely questioned on a variety of issues about which we in the United States are well versed. But when discussion at the conference addressed the meaning and importance of academic freedom, Professor Balin pointed out that academic freedom from interference by the state is newly won in Argentina and still nonexistent in many places. Thus he claimed the right to consider some relationships as more important and more in need of monitoring than others. For many researchers, their relationship with government (whether because of its political agendas or its funding priorities) is at least as important in how they think about and do their work in good conscience as their relationships with their subjects, audiences, and academic communities. Explicit discussion of these relationships and the moral priorities that appear to follow from them has heretofore been rare in human subjects research. Second, at the same time, other cases, particularly Alan Benjamin's research (Case I), as well as the concept of participatory research, turn the critical gaze fully on the research enterprise and ask what it is, what it is good for, and why researchers do it. Certainly it is the relationships paradigm more than the principlist paradigm that encourages us to examine the role of research in society and the meaning and value of progress in science. One way to ask the question might be: "Is this human subjects research or is it social change?" Participatory researchers might be the first to answer "Yes-to both questions!" But we all should consider that answer.

220

CONCLUSION

Perhaps it can be said that the principlist paradigm fails to highlight background questions about power and responsibility, in society and in ethics. Principlism at least appears to take certain things as given, including the priority of research, the segregating authority of expertise, and a hierarchical relationship between governments and researchers that permits researchers to maintain that they are relatively powerless, that they just do science, not politics, or that preserving their funding and their academic freedom are their principal priorities. Those outside the researcher-government relationship find this excuse unsettling. All of the groups of subjects in this book would be skeptical of the researcher's claim of relative powerlessness. However, power analysis can be unsatisfying, precisely because it is such an effective analytical tool without necessarily providing clear guidance for individual moral action. Principlism, virtue ethics, even casuistry at least give the principal actors-again, according to these models, usually the principals ate the researchers-the sense that they can act effectively. By incorporating considerations of power, the relationships paradigm, like feminist ethics and even narrative thics, can influence viewpoints so that effective action becomes not only difficult but sometimes even difficult to identify. This may be especially clear in Case 3. Those who would argue that Popkin and his colleagues should not have accepted industry funding are not only bucking the modern trend; they will be asked, by Popkin, "But then where do my responsibilities end? What would you have me do?" Third, it is important to acknowledge that despite this, the relationships paradigm provides better practice in moral imagination than all but a few extremely sensitive principlists can exercise. It is avision tool, not necessarily an action guide. It gives the IRB member the ability to go back and ask questions that might seem already settled by virtue of the project's having gotten to the IRB at all. But the lone voice on the IRB will probably have less immediate effect than that same voice would have by raising the same questions earlier and elsewhere. The relationships paradigm has independent moral authority, even though it currently has no regulatory standing. This relative lack of authority should not, however, be a barrier to its use, particularly in teaching, as well as in aspects of the design and conduct of research that both precede and follow the researcher's necessary preoccupation with regulatory compliance. There is a fourth result from using the relationships paradigm to examine research. Researcher-subject relationships that have dimension and duration can develop into rich and complex exchanges distinguished by reciprocity and trust. One of the changes potentially represented by the relationships paradigm is the legitimation of the attempt to work toward mutuality between researchers and subjects. If morality in human subjects research is negotiated and created through the coming together of multiple moral voices, then the researcher's true peers are

REGULATIONS AND RELATIONSHIPS

221

those who chime into the discussion about what research is good for and whose goods should be sought.

Enlarging Discussion An example from outside the research setting shows that using new tools of ethical analysis can affect outcomes. There are many critics of the advocacy posture of feminist ethics; but used as a lens in policymaking, feminist ethics can make significant differences. When the Maryland legislature was revising its statutes governing advance directives for health care, health lawyers in the University of Maryland's Law and Health Care Program were able to spot several major unexamined assumptions and suggest significant improvements in the statutory scheme by applying a feminist perspective. By asking questions that a standard principlist paradigm of policymaking had neglected, their viewpoint prompted the legislative drafters to recognize two issues: (I) gender-neutral statutory language, when used in the health care context, can conceal a disparate impact on women, who are often the primary caregivers for children, the sick, and the elderly; and ( 2 ) a legislative scheme that establishes decisionmaking safepards designed to protect individuals from family members acting in bad faith also necessarily burdens goodfaith family decisionmaking by virtue of those same safepards (Rothenberg 1996). The draft statute was revised to minimize those unintended (and hitherto unnoticed) effects without unduly sacrificing the intended protections. Similarly, no one would advocate that every research project that rakes place in a community should be turned into participatory action research, but if the possibility of genuinely collaborative relationships between researchers and subject communities is routinely considered when research is designed, opportunities for useful collaboration may arise that would otherwise go unnoticed. The relationships paradigm lays open for critical examination many components of the regulatory, principlist paradigm, including simplification, linear development, and decontextualization. It focuses on multiple viewpoints, complexity, contradiction, change, local factors, and, above all, on connections and interactions of all types. Because we no longer have available, nor can we expect, a linear model and a set of accumulating truths from this new paradigm, it is very difficult to know where it can take us, and it is natural to regret the loss of the sense of certitude and direction that the old paradigm provides. Nonetheless, the cultures to which researchers, research audiences, funders, and communities of subjects belong are already changing in response to the effects of the new paradigm. When we talk about ethics and research we are no longer talking only about what IRBs do, but about the entire domain of research-purpose, problem formulation, funding, research populations, methods, use and reporting

222

CONCLUSION

of results. As the cases and commentaries in this book illustrate, ethics are part of each of these, and each decision made about research must now consider not only relationships, but the role played in those relationships by power and trust. Can these new paradigm considerations inform research policy? This is a question this book only begins to touch on. We know from Professor Macklin's essay, as well as from many of the cases and commentaries, that considering relationships, particularizing inquiries, and using context to foster understanding of culture can improve oversight of research. We discovered this with confirming force when two of us (Henderson and King) brought some of the cases in this book to an international meeting of clinical epidemiologists from many developed and developing countries. The importance of relationships, culture, and context was apparent to them. So was the meaningfulness of autonomy, beneficence, and justice, all three ofwhich were significant in their experiences with their funders and governments as well as with their subjects and communities of subjects. Much as Professor MacMin argues, this international audience of researchers suggested that individual autonomy is meaningful everywhere, but its meaning is particular and local. In many respects, then, the relationships paradigm is currently used by thoughtful researchers attempting to do the best work they can while satisfying the principlist paradigm and its regulatory strictures. Thus, the relationships paradigm can and does inform the application of policy to research. But as we have seen, this use almost necessarily subordinates the relationships paradigm and denies it independent moral standing Current research regulations and guidelines are products exclusively of the principlist paradigm. To influence policy more directly and pervasively, the concerns of the relationships paradigm must be reflected in its language, focus, and structure, not just in its application. How might human subjects research policy be amended to include the relationships paradigm? Readers are invited to examine the regulations and guidelines in Appendices B, C, and D, and to speculate about any necessary changes and the form they might take. The relationships paradigm enlarges the frame of the discussion, in order to help the involved parties to better see the scope and nature of the question or disagreement. It focuses more attention on identification of the historical, cultural, and social determinants of the choices made by those in disagreement, so that we can see what values each choice serves and why. In this respect it resembles narrative ethics, as well as the general view that moral reasoning cannot be understood apart from its social and cultural context (Churchill 1997). This is one of the ways in which the relationships paradigm promises to influence human subjects research policy-by fostering the mutual exploration and exercise of moral imagination that permit the parties in disagreement to understand the sources of the disagreement and the implications of the conflicting choices. We could conclude that the principlist paradigm does indeed still "work," but

REGULATIONS AND RELATIONSHIPS

223

cannot stand alone. Using the considerations of the new paradigm to supplement it is extremely important, and certainly neglected. Nonetheless, such supplementation is not the only practical outcome to anticipate from the paradigm shift. As we have argued, the policy implications of merely supplementing the principlist paradigm with the relationships paradigm are rather narrow. If the moral analysis of research is indeed changing or should change to draw significantly from both paradigms, policy should change more fundamentally as well. And certainly, change must come in how scholars and researchers think and teach. If incremental change continues, then the gaps between the regulations and reality might keep getting larger, and more of the relationships paradigm might be packed in to fill the gaps; at some future point, the regulations will begin to take on a less prominent role, like that of the stone at the bottom of the soup pot. Yet even after the existing and ongoing complexities of relationships in research are acknowledged, there is not going to be a road map for navigating them (see for example Cornwall and Jewkes 1995; Israel et al. 1998). Instead, we are in the midst of laying down the road we are traveling. And even after the mapping of that road is completed, we will need to determine the usefulness of each portion of the existing maps, however enlarged, for each journey. Take the current and pressing example of differing perspectives in collaborative research between the developed and developing world, when the research in question cannot be conducted in the developed world and someone proposes to conduct it in the developing world. One such controversy recently arose over trials oflow-dose intrapartum AZT compared to placebo, in order to reduce motherto-infant transmission of HIV (Angell 1997; Lurie and Wolff 1997; Varmus and Satcher 1997). In the developed world, an intensive peripartum AZT regimen has been shown to significantly reduce the likelihood of transmission, and has become the standard of care. Thus, doing anything less in the research setting becomes unethical. In developing countries, however, this standard of care is effectively unavailable, and researchers argued that the possibility of finding effective, affordable lower doses and less complicated regimens justified such trials, especially since subjects receiving placebos were not being deprived of treatment they could otherwise obtain. Some commentators described this as a deliberate decision to allow the infection of many infants who could be spared with standard treatment; some even made analogies to the U.S. Public Health Service's Tuskegee syphilis study. Researchers in the developing countries responded that such comparisons are paternalistic; they have ethical standards, and so do their governments. Sponsors of the trials noted that the regulatory model applies the same rules to all countries involved, taking appropriate account of local differences, such as the standard of care available. Researchers from the developed world argued that they would certainly prefer to be able to conduct the same kinds of studies in developing countries that -

-

224

CONCLUSION

they do at home, but they are powerless to affect the funding priorities of their sponsors and governments. And finally, a few voices pointed out that the history of relationships between governments in the developed and developing worlds has implications for this debate. The legacy of colonialism, governmental involvement in complex ideological alliances, issues of poverty, resource management, and development aid complicate the relationships between countries and raise moral questions about responsibility at many levels. Such considerations in studies like these international AZT trials affectthough they cannot define-not only the relationships between researchers and their governments and those between researchers from different countries, but also the relationships between researchers, groups and communities of potential subjects, and subjects themselves, which might otherwise be obscured. Using the relationships paradigm to examine the ethics of this research thus changes the analytical process, broadens the definition and discussion of the problem, and hints at conclusions that are less polarized than the ideological oppositions that principlism has produced. This is not to claim that the problem is easily solved. We can, however, legitimately hope that the relationships paradigm might lead to a solution that, if implemented, is more lasting for being less narrow, less abstract, and better gounded in human realities than the principlist paradigm has shown itself to be. An important question is often raised by researchers who talk together in the attempt to bring some of these ideas into clearer focus: "To whom do we turn for moral argument?" How shall we constitute the community, or communities, to examine these things together? The language of the question is significant. It means, "With whom are we in a moral relationship of equals?" Not "Who will adjudicate this for us?Who will tell us the rules?" but "With whom can we talk?With whom can we work toward an answer?" It is significant that the question, which has meaning to all researchers, takes shape only in the relationships paradigm. This is only the beginning of the discussion, and it is both creative and passionate. Our authors are all researchers andlor academics. If we follow our new paradigm where it leads, we must now open the discussion to others-those individuals and communities whom we have studied or hope to study, and the funders and users of research. We anticipate more changes. The questions addressed here will not, finally, be resolved. Each one of us must make and live with our own moral decisions. But the more we turn to others for moral argument, the more we prepare ourselves by exposure to ethical issues and dilemmas in all their dimensions, and the more we glean from the struggles and wisdom of others, the better our decisions will be.

226

APPENDIX A

6. The degree of risk to be taken should never exceed that determined by the humanitarian importance of the problem to be solved by the experiment.

7. Proper preparations should be made and adequate facilities provided to protect the experimental subject against even remote possibilities of injury, disability, or death. 8. The experiment should be conducted only by scientifically qualified persons. The highest degree of skill and care should be required through all stages of the experiment of those who conduct or engage in the experiment.

9. During the course of the experiment the human subject should be at liberty to bring the experiment to an end if he has reached the physical or mental state where continuation of the experiment seems to him to be impossible. During the course of the experiment the scientist in charge must be prepared to terminate the experiment at any stage, if he has probable cause to believe, in the exercise of the good faith, superior skill and careful judgment required of him that a continuation of the experiment is likely to result in injury, disability, or death to the experimental subject. 10.

["Permissible Medical Experiments." Trials of War Criminals before the Nuremberg Military Tribunals under Control Council Law No. 10: Nuremberg October 1946-April 1949. Washington, D.C.: U.S. Government Printing Office, n.d., vol. 2, 8 181-82.1

228 APPENDIX B

. . .

. . . .

118 Applications and proposals lacking definite plans for involvement of human subjects 119 Research undertaken without the intention of involving human subjects 120 Evaluation and disposition of applications and proposals for research to be conducted or supported by a federal department or agency 121 [Reserved] 122 Use of federal funds 123 Early termination of research support; evaluation of applications and proposals 124 Conditions

. i o i To What Does This Policy Apply? Except as provided in paragraph (b) of this section, this policy applies to all research involving human subjects conducted, supported or otherwise subject to regulation by any federal department or agency which takes appropriate administrative action to make the policy applicable to such research. This includes research conducted by federal civilian employees or military personnel, except that each department or agency head may adopt such ~roceduralmodifications as may be appropriate from an administrative standpoint. It also includes research conducted, supported, or otherwise subject to regulation by the federal government outside the United States. (I) Research that is conducted or supported by a federal department or agency, whether or not it is regulated as defined in $ . ~ o z ( e )must , comply with all sections of this policy. (2) Research that is neither conducted nor supported by a federal department or agency but is subject to regulation as defined in 4-.roz(e) must be reviewed and approved in compliance with $ 1 1 , . 1 0 2 , and S . 1 0 7 through g . 1 1 7 of this policy, by an institutional review board (IRB) that operates in accordance with the pertinent requirements of this policy. (b) Unless otherwise required by department or agency heads, research activities in which the only involvement of human subjects will be in one or more of the following categories are exempt from this policy: (I) Research conducted in established or commonly accepted educational settings, involving normal educational practices, such as (i) research on regular and special education instructional strategies, or (ii) research on the effectiveness of or the comparison among instructional techniques, curricula, or classroom management methods. (2) Research involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures or observation of public behavior, unless: (i) Information obtained is recorded in such a manner that human subjects can be identified, directly or through identifiers linked to the subjects; and (ii) any disclosure of the human subjects responses outside the research could reasonably place the subjects at risk of criminal or civil liability or be damaging to the subjects' financial standing, employability, or reputation.

(a)

FEDERAL COMMON RULE 2 2 9

Research involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures, or observation of public behavior that is not exempt under paragraph (b)(2) of this section, iE (i) The human subjects are elected or appointed public officials or candidates for public office; or (ii) federal statute(s) require(s) without exception that the confidentiality of the personally identifiable information will be maintained throughout the research and thereafter. (4) Research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects. ( 5 ) Research and demonstration projects which are conducted by or subject to the approval of department agency heads, and which are designed to study, evaluate, or otherwise examine: (i) Public benefit or service programs; (ii) procedures for obtaining benefits or services under those programs; (iii) possible changes in or alternatives to those programs or procedures; or (iv) possible changes in methods or levels of payment for benefits or services under those programs.

(3)

(6) Taste and food qualiry evaluation and consumer acceptance studies, (i) ifwhole-

(c) (d)

(e) (f)

(g) (h)

some foods without additives are consumed or (ii) if a food is consumed that contains a food ingredient at or below the level and for a use found to be safe, or agricultural chemical or environmental contaminant at or below the level found to be safe, by the Food and Drug Administration or approved by the Environmental Protection Agency or the Food Safety and Inspection Service of the U.S. Department of Agriculture. Department or agency heads retain final judgment as to whether a particular activity is covered by this policy. Department or agency heads may require that specific research activities or classes of research activities conducted, supported, or otherwise subject to regulation by the department or agency but not otherwise covered by this policy, comply with some or all of the requirements of this policy. Compliance with this policy requires compliance with pertinent federal laws or regulations which provide additional protections for human subjects. This policy does not affect any state or local laws or regulations which may otherwise be applicable and which provide additional protections for human subjects. does not affect any foreign laws or regulations which may otherwise be This applicable and which provide additional protections to human subjects of research. When research covered by this policy takes place in foreign countries, procedures normally followed in the foreign countries to protect human subjects may differ from those set forth in this policy. [An example is a foreign institution which complies with guidelines consistent with the World Medical Assembly Declaration (Declaration of Helsinki amended 1989) issued either by sovereign states or by an organization whose function for the protection of human research subjects is internationally recognized.] In these circumstances, if a department or agency head determines that the

230 APPENDIX B

(i)

procedures prescribed by the institution afford protections that are at least equivalent to those provided in this policy, the department or agency head may approve the substitution of the foreign procedures in lieu of the procedural requirements provided in this policy. Except when otherwise required by statute, Executive Order, or the department or agency head, notices of these actions as they occur will be published in the Federal Register or will be otherwise published as provided in department or agency procedures. Unless otherwise required by law, department or agency heads may waive the applicability of some or all of the provisions of this policy to specific research activities or classes of research activities otherwise covered by this policy. Except when otherwise required by statute or Executive Order, the department or agency head shall forward advance notices of these actions to the Office for Protection from Research Risks, Department of Health and Human Services (HHS), and shall also publish them in the Federal Register or in such other manner as provided in department or agency procedures.

. l o 2 Definitions

Department or agency head means the head of any federal department or agency and any other officer or employee of any department or agency to whom authority has been delegated. Institution means any public or private entity or agency (including federal, state, and other agencies). Legally authorized representative means an individual or judicial or other body authorized under applicable law to consent on behalf of a prospective subject to the subject's participation in the procedure(s) involved in the research. Research means a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge. Activities which meet this definition constitute research for purposes of this policy whether or not they are conducted or supported under a program which is considered research for other purposes. For example, some demonstration and service programs may include research activities. Research subject to regulation and similar terms are intended to encompass those research activities for which a federal department or agency has specific responsibility for regulating as a research activity (for example, Investigational New Drug requirements administered by the Food and Drug Administration). It does not include research activities which are incidentally replated by a federal department or agency solely as part of the department's or agency's broader responsibility to regulate certain types of activities whether research or non-research in nature (for example, Wage and Hour requirements administered by the Department of Labor). Human subject means a living individual about whom an investigator (whether professional or student) conducting research obtains (I) data through intervention or interaction with the individual, or

FEDERAL COMMON RULE

231

identifiable private information. Intervention includes both physical procedures by which data are gathered (for example, venipuncture) and manipulations of the subject or the subject's environment that are performed for research purposes. Interaction includes communication or interpersonal contact between investigator and subject. "Private information" includes information about behavior that occurs in a context in which an individual can reasonably expect that no observation or recording is taking place, and information which has been provided for specific purposes by an individual and which the individual can reasonably expect will not be made public (for example, a medical record). Private information must be individually identifiable (i.e., the identity of the subject is or may readily be ascertained by the investigator or associated with the information) in order for obtaining the information to constitute research involving human subjects. (g) IRB means an institutional review board established in accord with and for the purposes expressed in this policy. (h) IRB approvalmeans the determination of the IRB that the research has been reviewed and may be conducted at an institution within the constraints set forth by the IRB and by other institutional and federal requirements. (i) Minimal risk means that the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests. (j) CertiJiation means the official notification by the institution to the supporting department or agency, in accordance with the requirements ofthis policy, that a research project or activity involving human subjects has been reviewed and approved by an IRB in accordance with an approved assurance. (2)

. l o 7 IRB Membership (a) Each IRB shall have at least five members, with varying backgrounds to promote complete and adequate review of research activities comn~onlyconducted by the institution. The IRB shall be sufficiently qualified through the experience and expertise of its members, and the diversity of the members, including consideration of race, gender, and cultural backgrounds and sensitivity to such issues as community attitudes, to promote respect for its advice and counsel in safeguarding the rights and welfare of human subjects. In addition to possessing the professional competence necessary to review specific research activities, the IRB shall be able to ascertain the acceptability of proposed research in terms of institutional commitments and regulations, applicable law, and standards of professional conduct and practice. The IRB shall therefore include persons knowledgeable in these areas. If an IRB regularly reviews research that involves a vulnerable category of subjects, such as children, prisoners, pregnant women, or handicapped or mentally disabled persons, consideration shall be given

232 APPENDIX B

(b)

(c) (d)

(e)

(f)

to the inclusion of one or more individuals who are knowledgeable about and experienced in working with these subjects. Every nondiscriminatory effort will be made to ensure that no IRB consists entirely of men or entirely of women, including the institution's consideration of qualified persons of both sexes, so long as no selection is made to the IRB on the basis of gender. No IRB may consist entirely of members of one profession. Each IRB shall include at least one member whose primary concerns are in scientific areas and at least one member whose primary concerns are in nonscientific areas. Each IRB shall include at least one member who is not otherwise affiliated with the institution and who is not part of the immediate family of a person who is affiliated with the institution. No IRB may have a member participate in the IRB's initial or continuing review of any project in which the member has a conflicting interest, except to provide information requested by the IRB. An IRB may, in its discretion, invite individuals with competence in special areas to assist in the review of issues which require expertise beyond or in addition to that available on the IRB. These individuals may not vote with the IRB.

. i o g IRB Review of Research (a) An IRB shall review and have authority to approve, require modifications in (to secure approval), or disapprove all research activities covered by this policy. (b) An IRB shall require that information given to subjects as part of informed consent is in accordance with g . 1 1 6 . The IRB may require that information, in addition to that specifically mentioned in g . 1 1 6 , be given to the subjects when in the IRB's judgment the information would meaningfully add to the protection of the rights and welfare of subjects. (c) An IRB shall require documentation of informed consent or may waive documentation in accordance with g . 1 1 7 . (d) An IRB shall notifj investigators and the institution in writing of its decision to approve or disapprove the proposed research activity, or of modifications required to secure IRB approval of the research activity. If the IRB decides to disapprove a research activity, it shall include in its written notification a statement of the reasons for its decision and give the investigator an opportunity to respond in person or in writing. (e) An IRB shall conduct continuing review of research covered by this policy at intervals appropriate to the degree of risk, but not less than once per year, and shall have authority to observe or have a third party observe the consent process and the research. (Approved by the Office of Management and Budget under Control Number 9999-0020.)

. i i o Expedited Review Procedures for Certain Kinds of Research Involving No More Than Minimal Risk, and for Minor Changes in Approved Research (a) The Secretary, HHS, has established, and published as a Notice in the Federal Register, a list of categories of research that may be reviewed by the IRB through an ex-

FEDERAL COMMON RULE

233

pedited review procedure. The list will be amended, as appropriate after consultation with other departments and agencies, through periodic republication by the Secretary, HHS, in the Federal Register. A copy of the list is available from the Office for Protection from Research Risks, National Institutes of Health, HHS, Bethesda, Maryland 20892. (b) An IRB may use the expedited review procedure to rcvicw either or both of the following: (I) Some or all of the research appearing on the list and found by the reviewer(s) to involve no more than minimal risk. (2) Minor changes in previously approved research during the period (of one year or less) for which approval is authorized. Under an expedited review procedure, the review may be carried out by the IRB chairperson or by one or more experienced reviewers designated by the chairperson from among members of the IRB. In reviewing the research, the reviewers may exercise all of the authorities of the IRB except that the reviewers may not disapprove the research. A research activity may be disapproved only after review in accordance with the nonexpedited procedure set forth in $ . 1 0 8 ( b ) . (c) Each IRB which uses an expedited review procedure shall adopt a method for keeping all members advised of research proposals which have been approved under the procedure. (d) The department or agency head may restrict, suspend, terminate, or choose not to authorize an institution's or IRB's use of the expedited review procedure.

. i l l

(a)

Criteria for IRB Approval of Research

In order to approve research covered by this policy the IRB shall determine that all of the following requirements are satisfied: (I) Risks to subjects are minimized: (i) By using pocedures which are consistent with sound research design and which do not unnecessarily expose subjects to risk, and (ii) whenever appropriate, by using procedures already being performed on the subjects for diagnostic or treatment purposes. (2) Risks to subjects are reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected to result. In evaluating risks and benefits, the IRB should consider only those risks and benefits that may result from the research (as distinguished from risks and benefits of therapies subjects would receive even if not participating in the research). The IRB should not consider possible long-range effects of applying knowledge gained in the research (for example, the possible effects of the research on public policy) as among those research risks that fall within the purview of its responsibility. (3) Selection of subjects is equitable. In making this assessment the IRB should take into account the purposes of the research and the setting in which the research will be conducted and should be particularly cognizant of the special problems of research involving vulnerable populations, such as children, prisoners, preg-

234 APPENDIX B

nant women, mentally disabled persons, or economically or educationally disadvantaged persons. (4) Informed consent will be sought from each prospective subject or the subject's legally authorized representative, in accordance with, and to the extent required by 4 . 1 1 6 . (7) Informed consent will be appropriately documented, in accordance with, and to the extent required by 4 . 1 1 7 . (6) When appropriate, the research plan makes adequate provision for monitoring the data collected to ensure the safety of subjects.

(7) When appropriate, there are adequate provisions to protect the privacy of subjects and to maintain the confidentiality of data. (b) When some or all of the subjects are likely to be vulnerable to coercion or undue influence, such as children, prisoners, pregnant women, mentally disabled persons, or economically or educationally disadvantaged persons, additional safepards have been included in the study to protect the rights and welfare of these subjects.

. i i 3 Suspension or Termination of IRB Approval of Research An IRB shall have authority to suspend or terminate approval of research that is not being conducted in accordance with the IRB's requirements or that has been associated with unexpected serious harm to subjects. Any suspension or termination of approval shall include a statement of the reasons for the IRBk action and shall be reported . promptly to the . investigator, appropriate institutional officials, and the department or agency head. (Approved by the Office of Management and Budget under Control Number 9999-0020.)

. i i 4 Cooperative Research Cooperative research projects are those projects covered by this policy which involve more than one institution. In the conduct of coopkrative research projects, each institution is responsible for safeguarding the rights and welfare of human subjects and for complying with this policy. With the approval of the department or agency head, an institution participating in a cooperative project may enter into a joint review arrangement, rely upon the review of another qualified IRB, or make similar arrangements for avoiding duplication of effort.

. i i 5 IRB Records (a) An institution, or when appropriate an IRB, shall prepare and maintain adequate documentation of IRB activities, including the following: (I) Copies of all research proposals reviewed, scientific evaluations, if any, that accompany the proposals, approved sample consent documents, progress reports submitted by investigators, and reports of injuries to subjects.

FEDERAL COMMON RULE

235

Minutes of IRB meetings which shall be in sufficient detail to show attendance at thc mcctings; actions takcn by the IRB; the votc on these actions including the number of members voting for, against, and abstaining; the basis for requiring changes in or disapproving research; and a written summary of the discussion of controverted issues and their resolution. Records of continuing review activities. Copies of all correspondencebetween the IRB and the investigators. A list of IRB members in the same detail as described in % . r o ? ( b ) ( 3 ) . Written procedures for the IRB in the same detail as described in 4.103(b)(4) and §--103(b) (5). Statements of significant new findings provided to subjects, as required by §-.II~(~)(Y). (b) The records required by this policy shall be retained for at least 3 years, and records relating to research which is conducted shall be retained for at least 3 years after completion of the research. All records shall be accessible for inspection and copying by authorized representatives of the department or agency at reasonable times and in a reasonable manner. (Approved by the Office of Management and Budget under Control Number 9999-0020.)

. 1 1 6 General Requirementsfor Informed Consent Except as provided elsewhere in this policy no investigator may involve a human being as a subject in research covered by this policy unless the investigator has obtained the legally effective informed consent of the subject or the subject's legally authorized representative. An investigator shall seek such consent only under circumstances that provide the prospective subject or the representative sufficient opportunity to consider whether or not to participate and that minimize the possibility of coercion or undue influence. The information that is given to the subject or the representative shall be in language understandable to the subject or the representative. No informed consent, whether oral or written, may include any exculpatory language through which the subject or the representative is made to waive or appear to waive any of the subject's legal rights, or releases or appears to release the investigator, the sponsor, the institution or its agents from liability for negligence. (a) Basic elements of informed consent. Except as provided in paragraph (c) or (d) of this section, in seeking informed consent the following information shdI be provided to each subject: (I) A statement that the study involves research, an explanation of the purposes of the research and the expected duration of the subject's participation, a description of the procedures to be followed, and identificationof any procedures which are experimental; (2) A description of any reasonably foreseeable risks or discomforts to the subject; (3) A description of any benefits to the subject or to others which may reasonably be expected from the research; (4) A disclosure of appropriate alternative procedures or courses of treatment, if any, that might be advantageous to the subject;

236 APPENDIX B

( 5 ) A statement describing the extent, if any, to which confidentiality of records identifying the subject will be maintained; (6) For research involving more than minimal risk, an explanation as to whether any compensation and an explanation as to whether any medical treatments are available if injury occurs and, if so, what they consist of, or where further information may be obtained; (7) An explanation ofwhom to contact for answers to pertinent questions about the research and research subjects' rights, and whom to contact in the event of a research-related injury to the subject; and (8) A statement that participation is voluntary, refusal to participate will involve no penalty or loss of benefits to which the subject is otherwise entitled, and the subject may discontinue participation at any time without penalty or loss ofbenefits to which the subject is otherwise entitled. (b) Additional elements of informed consent. When appropriate, one or more of the following elements of information shall also be provided to each subject: (I) A statement that the particular treatment or procedure may involve risks to the subject (or to the embryo or fetus, if the subject is or may become pregnant) which are currently unforeseeable; (2) Anticipated circumstances under which the subject's participation may be terminated by the investigator without regard to the subject's consent; (3) Any additional costs to the subject that may result from participation in the research; (4) The consequences of a subject's decision to withdraw from the research and procedures for orderly termination of participation by the subject; ( 5 ) A statement that significant new findings developed during the course of the research which may relate to the subject's willingness to continue participation will be provided to the subject; and (6) The approximate number of subjects involved in the study. (c) An IRB may approve a consent procedure which does not include, or which alters, some or all of the elements of informed consent set forth above, or waive the requirement to obtain informed consent provided the IRB finds and documents that: (I) The research or demonstration project is to be conducted by or subject to the approval of state or local government officials and is designed to study, evaluate, or otherwise examine: (i) Public benefit or service programs; (ii) procedures for obtaining benefits or services under those programs; (iii) possible changes in or alternatives to those programs or procedures; or (iv) possible changes in methods or levels of payment for benefits or services under those programs; and (2) The research could not practicably be carried out without the waiver or alteration. (d) An IRB may approve a consent procedure which does not include, or which alters, some or all of the elements of informed consent set forth in this section, or waive the requirements to obtain informed consent provided the IRB finds and documents that;

FEDERAL COMMON RULE

237

The research involves no more than minimal risk to the subjects; The waiver or alteration will not adversely affect the rights and welfare of the subjects; (3) The research could not practicably be carried out without the waiver or alteration; and (4) Whenever appropriate, the subjects will be provided with additional pertinent information after participation. (e) The informed consent requirements in this policy are not intended to preempt any applicable federal, state, or local laws which require additional information to be disclosed in order for informed consent to be legally effective. (f) Nothing in this policy is intended to limit the authority of a physician to provide emergency medical care, to the extent the physician is permitted to do so under applicable federal, state, or local law. (Approved by the Office of Management and Budget under Control Number 9999-0020.) (I)

(2)

. i i 7 Documentation of Informed Consent Except as provided in paragraph (c) of this section, informed consent shall be documented by the use of a written consent form approved by the IRB and signed by the subject or the subject's legally authorized representative. A copy shall be given to the person signing the form. (b) Except as provided in paragraph (c) of this section, the consent form may be either of the following: (I) A written consent document that embodies the elements of informed consent required by $ . 1 1 6 . This form may be read to the subject or the subject's legally authorized representative, but in any event, the investigator shall give either the subject or the representative adequate opportunity to read it before it is signed; or ( 2 ) A short form written consenr document stating thar the elements of informed consent required by O . 1 1 6 have been presented orally to the subject or the subject's legally authorized representative. When this method is used, there shall be a witness to the oral presentation. Also, the IRB shall approve a written summary of what is to be said to the subject or the representative. Only the short form itself is to be signed by the subject or the representative. However, the witness shall sign both the short form and a copy of the summary, and the person actually obtaining consent shall sign a copy of the summary. A copy of the summary shall be given to the subject or the representative, in addition to a copy of the short form. (c) An IRB may waive the requirement for the investigator to obtain a signed consent form for some or all subjects if it finds either: (I) That the only record linking the subject and the research would be the consent document and the principal risk would be potential harm resulting from a (a)

238 APPENDIX B

breach of confidentiality. Each subject will be asked whether the subject wants documentation linking the subject with the research, and the subject's wishes will govern; or (2) That the research presents no more than minimal risk of harm to subjects and involves no procedures for which written consent is normally required outside of the research context. In cases in which the documentation requirement is waived, the IRB may require the investigator to provide subjects with a written statement regarding the research. (Approved by the Office of Management and Budget under Control Number 9999-0020).

240

APPENDIX

C

that the investigators be competent both to conduct the research and to assure the wellbeing of the research subjects. NonmaleJcence ("Do no harm") holds a central position in the tradition of medical ethics, and guards against avoidable harm to research subjects. Justice requires that cases considered to be alike be treated alike, and that cases considered to be different be treated in ways that acknowledge the difference. When the principle of justice is applied to dependent or vulnerable subjects, its main concern is with the rules of distributivejustice. Studies should be designed to obtain knowledge that benefits the class of persons ofwhich the subjects are representative: the class of persons bearing the burden should receive an appropriate benefit, and the class primarily intended to benefit should bear a fair proportion of the risks and burdens of the study. The rules of distributive justice are applicable within and among communities. Weaker members of communities should not bear disproportionate burdens of studies from which ail members of the community are intended to benefit, and more dependent communities and countries should not bear disproportionate burdens ofstudies from which all communities or countries are intended to benefit. General ethical principles may be applied at individual and community levels. At the level of the individual (microethics), ethics governs how one person should relate to another and the moral claims of each member of a community. At the level of the community, ethics applies to how one community relates to another, and to how a community treats each of its members (including prospective members) and members of other groups with different cultural values (macroethics). Procedures that are unethical at one level cannot be justified merely because they are considered clinically acceptable at the other.

Ethical Principles Applied to Epidemiology Informed Consent INDIVIDUAL CONSENT

I. When individuals are to be subjects of epidemiological studies, their informed consent will usually be sought. For epidemiological studies that use personally identifiable private data, the rules for informed consent vary, as discussed further below. Consent is informed when it is given by a person who understands the purpose and nature of the study, what participation in the study requires the person to do and to risk, and what benefits are intended to result from the study. 2. An investigator who proposes not to seek informed consent has the obligation to explain to an ethical review committee how the study would be ethical in its absence: it may be impractical to locate subjects whose records are to be examined, or the purpose of some studies would be frustrated-for example, prospective subjects on being informed would change the behavior that it is proposed to study, or might feel needlessly anxious about why they were subjects of study. The investigator will provide assurances that strict safeguards will be maintained to protect confidentiality and that the study is aimed at protecting or advancing health. Another justification for not seeking informed consent may

ClOMS EPIDEMIOLOGICAL RESEARCH GUIDELINES

241

be that subjects are made aware through public announcements that it is customary to make personal data available for epidemiological studies. 3. An ethical issue may arise when occupational records, medical records, tissue samples, etc. are used for a purpose for which consent was not given, although the study threatens no harm. Individuals or their public representatives should normally be told that their data might bc uscd in ~pidemiolo~ical studies, and what means of protecting confidentiality are provided. Consent is not required for use of publicly available information, although countries and communities differ with regard to the definition of what information about citizens is regarded as public. However, when such information is to be used, it is understood that investigators will minimize disclosure of personally sensitive information. 4. Some organizations and government agencies employ epidemiologists who may be permitted by legislation or employees' contracts to have access to data without subjects' consent. These epidemiologists must then consider whether it is ethical for them, in a given case, to use this power of access to personal data. Ethically, they may still be expected either to seek the consent of the individuals concerned, or to justify their access without such consent. Access may be ethical on such grounds as minimal risk of harm to individuals, public benefit, and investigators' protection of the confidentiality of the individuals whose data they study. COMMUNITY AGREEMENT

5. When it is not possible to request informed consent from every individual to be studied, the agreement of a representative of a community or group may be sought, but the representative should be chosen according to the nature, traditions and political philosophy of the community or group. Approval given by a community representative should be consistent with general ethical principles. When investigators work with communities, they will consider communal rights and protection as they would individual rights and protection. For communities in which collective decision-making is customary, communal leaders can express the collective will. However, the refusal of individuals to participate in a study has to be respected: a leader may express agreement on behalf of a community, but an individual's refusal of personal participation is binding. 6. When people are appointed by agencies outside a group, such as a department of government, to speak for members of the group, investigators and ethical review committees should consider how authentically these people speak for the group, and if necessary seek also the agreement of other representatives. Representatives of a community or group may sometimes be in a position to participate in designing the study and in its ethical assessment. 7. The definition of a community or group for purposes of epidemiological study may be a matter of ethical concern. When members of a community are naturally conscious of its activities as a community and feel common interests with other members, the community exists irrespective of the study proposal. Investigators will be sensitive to how a community is constituted or defines itself, and will respect the rights of underprivileged groups. 8. For purposes of epidemiological study, investigators may define groups that are composed of statistically, geographically or otherwise associated individuals who do not normally interact socially. When such groups are artificially created for scientific study, group

242

APPENDIX C

members may not readily be identifiable as leaders or representatives, and individuals may not be expected to risk disadvantage for the benefit of others. Accordingly, it will be more difficult to ensure group representation, and all the more important to obtain subjects' free and informed consent to participate. SELECTIVE DISCLOSURE O F INFORMATION

9. In epidemiology, an acceptable study technique involves selective disclosure of information, which seems to conflict with the principle of informed consent. For certain epidemiological studies nondisclosure is permissible, even essential, so as to not influence the spontaneous conduct under investigation, and to avoid obtaining responses that the respondent might give in order to please the questioner. Selective disclosure may be benign and ethically permissible, provided that it does not induce subjects to do what they would not otherwise consent to do. An ethical review committee may permit disclosure of only selected information when this course is justified. UNDUE INFLUENCE 10. Prospective subjects may not feel free to refuse requests from those who have power or influence over them. Therefore the identity of the investigator or other person assigned to invite prospective subjects to participate must be made known to them. Investigators are expected to explain to the ethical review committee how they propose to neutralize such apparent influence. It is ethically questionable whether subjects should be recruited from among groups that are unduly influenced by persons in authority over them or by community leaders, if the study can be done with subjects who are not in this category.

INDUCEMENT T O PARTICIPATE

Individuals or communities should not be pressured to participate in a study. However, it can be hard to draw the line between exerting pressure or offering inappropriate inducements and creating legitimate motivation. The benefits of a study, such as increased or new knowledge, are proper inducements. However, when people or communities lack basic health services or money, the prospect of being rewarded by goods, services or cash payments can induce participation. To determine the ethical propriety of such inducements, they must be assessed in the light of the traditions of the culture. 12. Risks involved in participation should be acceptable to subjects even in the absence of inducement. It is acceptable to repay incurred expenses, such as for travel. Similarly, promises of compensation and care for damage, injury or loss of income should not be considered inducements. 11.

Maximizing Benefit COMMUNICATION O F STUDY RESULTS

13. Part of the benefit that communities, groups and individuals may reasonably expect from participating in studies is that they will be told of findings that pertain to their health. Where findings could be applied in public health measures to improve community health, they should be communicated to the health authorities. In informing individuals of the findings and their pertinence to health, their level of literacy and comprehension must be considered. Research protocols should include provision for communicating such information to communities and individuals.

ClOMS EPIDEMIOLOGICAL RESEARCH GUIDELINES 243

Research findings and advice to communities should be publicized by whatever suitable means are available. When HIV-prevalence studies are conducted by unlinked anonymous screening, there should be, where feasible, provision for voluntary HIV-antibody testing under conditions of informed consent, with pre- and post-test counseling, and assurance of confidentiality. IMPOSSIBILITY O F COMMUNICATING STUDY RESULTS

14. Subjects of epidemiological studies should be advised that it may not be possible to inform them about findings that pertain to their health, but that they should not take this to mean that they are free of the disease or condition under study. Often it may not be possible to extract from pooled findings information pertaining to individuals and their families, but when findings indicate a need of health care, those concerned should be advised of means of obtaining personal diagnosis and advice. When epidemiological data are unlinked, a disadvantage to subjects is that individuals at risk cannot be informed of useful findings pertinent to their health. When subjects cannot be advised individually to seek medical attention, the ethical duty to do good can be served by making pertinent health-care advice available to their communities. RELEASE OF STUDY RESULTS

15. Investigators may be unable to compel release of data held by governmental or commercial agencies, but as health professionals they have an ethical obligation to advocate the release of information that is in the public interest. Sponsors of studies may press investigators to present their findings in ways that advance special interests, such as to show that a product or procedure is or is not harmful to health. Sponsors must not present interpretations or inferences, or theories and hypotheses, as if they were proven truths. HEALTH CARE FOR T H E COMMUNITY UNDER STUDY

16. The undertaking of an epidemiological project in a developing country may create the expectation in the community concerned that it will be provided with health care, at least while the research workers are present. Such an expectation should not be frustrated, and, where people need health care, arrangements should be made to have them treated or they should be referred to a local health service that can provide the needed care. TRAINING LOCAL HEALTH PERSONNEL 17. While studies are in progress, particularly in developing countries, the opportunity should be taken to train local health workers in skills and techniques that can be used to improve health services. For instance, by training them in the operation of measuring devices and calculating machines, when a study team departs it leaves something of value, such as the ability to monitor disease or mortality rates.

Minimizing H a m CAUSING HARM AND DOING WRONG

18. Investigators planning studies will recognize the risk of causing harm, in the sense of bringing disadvantage, and of doing wrong, in the sense of transgressing values. Harm

244 APPENDIX C

may occur, for instance, when scarce health personnel are diverted from their routine duties to serve the needs of a study, or when, unknown to a community, its health-care priorities are changed. It is wrong to regard members ofcommunities as only impersonal material for study, even if they are not harmed. 19. Ethical review must always assess the risk of subjects or groups suffering stigmatization, prejudice, loss of prestige or self-esteem, or economic loss as a result of taking part in a study. Investigators will inform ethical review committees and prospective subjects of perceived risks, and of proposals to prevent or mitigate them. Investigators must be able to demonstrate that the benefits outweigh the risks for both individuals and groups. There should be a thorough analysis to determine who would be at risk and who would benefit from the study. It is unethical to expose persons to avoidable risks disproportionate to the expected benefits, or to permit a known risk to remain if it can be avoided or at least minimized. 20. When a healthy person is a member of a population or sub-group at raised risk and engages in high-risk activities, it is unethical not to propose measures for protecting the population or sub-group. PREVENTING HARM T O GROUPS 21. Epidemiological studies may inadvertently expose groups as well as individuals to harm, such as economic loss, stigmatization, blame, or withdrawal of services. Investigators who find sensitive information that may put a group at risk of adverse criticism or treatment should be discreet in communicating and explaining their findings. When the location or circumstances of a study are important to understanding the results, the investigators will explain by what means they propose to protect the group from harm or disadvantage; such means include provisions for confidentiality and the use of language that does not imply moral criticism of subjects' behavior.

HARMFUL PUBLICITY 22. Conflict may appear between, on the one hand, doing no harm and, on the other, telling the truth and openly disclosing scientific findings. Harm may be mitigated by interpreting data in a way that protects the interests of those at risk, and is at the same time consistent with scientific integrity. Investigators should, where possible, anticipate and avoid misinterpretation that might cause harm.

RESPECT FOR SOCIAL MORES 23. Disruption of social mores is usually regarded as harmful. Although cultural values and social mores must be respected, it may be a specific aim of an epidemiological study to stimulate change in certain customs or conventional behavior to lead through change to healthful behavior-for instance, with regard to diet or a hazardous occupation. 24. Although members of communities have a right not to have others impose an uninvited "good" on them, studies expected to result in health benefits are usually considered ethically acceptable and not harmful. Ethical review committees should consider a study's potential for beneficial change. However, investigators should not overstate such benefits, in case a community's agreement to participate is unduly influenced by its expectation of better health services.

ClOMS EPIDEMIOLOGICAL RESEARCH GUIDELINES

245

SENSITIVITY TO DIFFERENT CULTURES 25. Epidemiologists often investigate cultural groups other than their own, inside or outside their own countries, and undertake studies initiated from outside the culture, community or country in which the study is to be conducted. Sponsoring and host countries may differ in the ways in which, in their cultures, ethical values are understood and applied-for instance, with rcgard to autonomy of individuals. Investigators must respect the ethical standards of their own countries and the cultural expectations of the societies in which epidemiological studies are undertaken, unless this implies a violation of a transcending moral rule. Investigators risk harming their reputation by pursuing work that host countries find acceptable but their own countries consider offensive. Similarly, they may transgress the cultural values of the host countries by uncritically conforming to the expectations of their own.

Confidentiulity 26. Research may involve collecting and storing data relating to individuals and groups, and such data, if disclosed to third parties, may cause harm or distress. Consequently, investigators should make arrangements for protecting the confidentiality of such data by, for example, omitting information that might lead to the identification of individual subjects, or limiting access to the data, or by other means. It is customary in epidemiology to aggregate numbers so that individual identities are obscured. Where group confidentiality cannot be maintained or is violated, the investigators should take steps to maintain or restore a group's good name and status. Information obtained about subjects is generally divisible into: Unlinked information, which cannot be linked, associated or connected with the person to whom it refers; as this person is not known to the investigator, confidentiality is not at stake and the question of consent does not arise. Linked information, which may be: anonymous, when the information cannot be linked to the person to whom it refers except by a code or other means known only to that person, and the investigator cannot know the identity of the person; non-nominal, when the information can be linked to the person by a code (not including personal identification) known to the person and the investigator; or nominal or nominative, when the information is linked to the person by means of personal identification, usually the name. Epidemiologists discard personal identifying information when consolidating data for purposes ofstatistical analysis. Identifiable personal data will not be used when a study can be done without personal identification-for instance, in testing unlinked anonymous blood samples for HIV infection. When personal identifiers remain on records used for a study, investigators should explain to review committees why this is necessary and how confidentiality will be protected. If, with the consent of individual subjects, investigators link different sets of data regarding individuals, they normally preserve confidentiality by aggregating individual data into tables or diagrams. In government service the obligation

246 APPENDIX C

to protect confidentiality is frequently reinforced by the practice of swearing employees to secrecy.

Conjict of Interest IDENTIFICATION OF CONFLICT OF INTEREST

27. It is an ethical rule that investigators should have no undisclosed conflict of interest with their study collaborators, sponsors or subjects. Investigators should disclose to the ethical review committee any potential conflict of interest. Conflict can arise when a commercial or other sponsor may wish to use study results to promote a product or service, or when it may not be politically convenient to disclose finding. 28. Epidemiological studies may be initiated, or financially or otherwise supported, by governmental or other agencies that employ investigators. In the occupational and environmental health fields, several well-defined special-interest groups may be in conflict: shareholders, management, labor, government regulatory agencies, public interest advocacy groups, and others. Epidemiological investigators may be employed by any of these groups. It can be difficult to avoid pressures resulting from such conflict of interest, and consequent distorted interpretations of study findings. Similar conflict may arise in studies of the effects of drugs and in testing medical devices. 29. Investigators and ethical review committees will be sensitive to the risk of conflict, and committees will not normally approve proposals in which conflict of interest is inherent. If, exceptionally, such a proposal is approved, the conflict of interest should be disclosed to prospective subjects and their communities. 30. There may appear to be conflict when subjects do not want to change their behavior and investigators believe that they ought to do so for the sake of their health. However, this may not be a true conflict of interest, as the investigators are motivated by the subjects' health interests. SCIENTIFIC OBJECTIVITY AND ADVOCACY

31. Honesty and impartiality are essential in designing and conducting studies, and presenting and interpreting findings. Data will not be withheld, misrepresented or manipulated. Investigators may discover health hazards that demand correction, and become advocates of means to protect and restore health. In this event, their advocacy must be seen to rely on objective, scientific data.

248 APPENDIX D

any foreseeable risks or discomfort to the subject, associated with participation in the research; any alternative procedures or courses of treatment that might be as advantageous to the subject as the procedure or treatment being tested; the extent to which confidentiality of records in which the subject is identified will bc rnaintaincd; the extent of the investigator's responsibility, if any, to provide medical services to the subject; that therapy will be provided free of charge for specified types of research-related injury; whether the subject or the subject's family or dependents will be compensated for disability or death resulting from such injury; and that the individual is free to refuse to participate and will be free to withdraw from the research at any time without penalty or loss of benefits to which he or she would otherwise be entitled.

3. OBLIGATIONS OF INVESTIGATORS The investigator has a duty to:

GUIDELINE

REGARDING INFORMED CONSENT

communicate to the prospective subject all the information necessary for adequately informed consent; give the prospective subject full opportunity and encouragement to ask questions; exclude the possibility of unjustified deception, undue influence and intimidation; seek consent only after the prospective subject has adequate knowledge of the relevant facts and of the consequences of participation and has had sufficient opportunity to consider whether to participate; as a general rule, obtain from each prospective subject a signed form as evidence of informed consent; and renew the informed consent of each subject if there are material changes in the conditions or procedures of the research.

4. INDUCEMENT T O PARTICIPATE Subjects may be paid for inconvenience and time spent, and should be reimbursed for expenses incurred, in connection with their participation in research; they may also receive free medical services. However, the payments should not be so large or the medical services so extensive as to induce prospective subjects to consent to participate in the research against their better judgment ("undue inducement"). All payments, reimbursements and medical services to be provided to research subjects should be approved by an ethical review committee. GUIDELINE

5. RESEARCH INVOLVING CHILDREN Before undertaking research involving children, the investigator must ensure that:

GUIDELINE

children will not be involved in research that might equally well be carried out with adults;

ClOMS BIOMEDICAL RESEARCH GUIDELINES

249

the purpose of the research is to obtain knowledge relevant to the health needs of children; a parent or legal guardian of each child has given proxy consent; the consent of each child has been obtained to the extent of the child's capabilities; the child's refusal to participate in research must always be respected unless according to the research protocol thc child would receive therapy for which there is no medically acceptable alternative; the risk presented by interventions not intended to benefit the individual childsubject is low and commensurate with the importance of the knowledge to be gained; and interventions that are intended to povide therapeutic benefit are likely to be at least as advantageous to the individual child-subject as any available alternative.

6. RESEARCH INVOLVING PERSONS WITH MENTAL OR BEHAVIORAL DISORDERS Before undertaking research involving individuals who by reason of mental or behavioral disorders are not capable of giving adequately informed consent, the investigator must ensure that: GUIDELINE

such persons will not be subjects of research that might equally well be carried out on persons in full possession of their mental faculties; the purpose of the research is to obtain knowledge relevant to the particular health needs of persons with mental or behavioral disorders; the consent of each subject has been obtained to the extent of that subject's capabilities, and a prospective subject's refusal to participate in nonclinical research is always respected; in the case of incompetent subjects, informed consent is obtained from the legal guardian or other duly authorized person; the degree of risk attached to interventions that are not intended to benefit the individual subject is low and commensurate with the importance of the knowledge to be gained; and interventions that are intended to provide therapeutic benefit are likely to be at least as advantageous to the individual subject as any alternative. 7. RESEARCH INVOLVING PRISONERS Prisoners with serious illness or at risk of serious illness should not arbitrarily be denied access to investigational drugs, vaccines or other agents that show promise of therapeutic or preventive benefit.

GUIDELINE

8. RESEARCH INVOLVING SUBJECTS IN UNDERDEVELOPED COMMUNITIES Before undertaking research involving subjects in underdeveloped communities, whether in developed or developing countries, the investigator must ensure that: GUIDELINE

persons in underdeveloped communities will not ordinarily be involved in research that could be carried out reasonably well in developed communities; the research is responsive to the health needs and the priorities of the community in which it is to be carried out;

250

APPENDIX D

every effort will be made to secure the ethical imperative that the consent of individual subjects be informed; and the proposals for the research have been reviewed and approved by an ethical review committee that has among its members or consultants persons who are thoroughly familiar with the customs and traditions of the community.

9. INFORMED CONSENT I N EPIDEMIOLOGICAL STUDIES For several types of epidemiological research individual informed consent is either impracticable or inadvisable. In such cases the ethical review committee should determine whether it is ethically acceptable to proceed without individual informed consent and whether the investigator's plans to protect the safety and respect the privacy of research subjects and to maintain the confidentiality of the data are adequate. GUIDELINE

Selection of Research Subjects GUIDELINE 10. EQUITABLE DISTRIBUTION O F BURDENS AND BENEFITS

Individuals or communities to be invited to be subjects of research should be selected in such a way that the burdens and benefits of the research will be equitably distributed. Special justification is required for inviting vulnerable individuals and, if they are selected, the means of protecting their rights and welfare must be particularly strictly applied. GUIDELINE 11. SELECTION OF PREGNANT OR NURSING (BREASTFEEDING) WOMEN AS RESEARCH SUBJECTS

Pregnant or nursing women should in no circumstances be the subjects of nonclinical research unless the research carries no more than minimal risk to the fetus or nursing infant and the object of the research is to obtain new knowledge about pregnancy or lactation. As a general rule, pregnant or nursing women should not be subjects of any clinical trials except such trials as are designed to protect or advance the health of pregnant or nursing women or fetuses or nursing infants, and for which women who are not pregnant or nursing would not be suitable subjects.

Con$dentiality of Data GUIDELINE 12. SAFEGUARDING CONFIDENTIALITY

The investigator must establish secure safepards of the confidentiality of research data. Subjects should be told of the limits to the investigators' ability to safeguard confidentiality and of the anticipated consequences of breaches of confidentiality.

Compensation of Research Subjectsfor Accidental Injury GUIDELINE 13. RIGHT OF SUBJECTS T O COMPENSATION

Research subjects who suffer physical injury as a result of their participation are entitled to such financial or other assistance as would compensate them equitably for any temporary or permanent impairment or disability. In the case of death, their dependents are entitled to material compensation. The right to compensation may not be waived.

ClOMS BIOMEDICAL RESEARCH GUIDELINES

251

Review Procedures 14. CONSTITUTION AND RESPONSIBILITIES OF ETHICAL REVIEW COMMITTEES All proposals to conduct research involving human subjects must be submitted for review and approval to one or more independent ethical and scientific review committees. The investigator must obtain such approval of the proposal to conduct research before the research is begun. GUIDELINE

Externally Sponsored Research GUIDELINE 15. OBLIGATIONS OF SPONSORING AND HOST COUNTRIES

Externally sponsored research entails two ethical obligations:

An external sponsoring agency should submit the research protocol to ethical and scientific review according to the standards of the country of the sponsoring agency, and the ethical standards applied should be no less exacting than they would be in the case of research carried out in that country. After scientific and ethical approval in the country of the sponsoring agency, the appropriate authorities of the host country, including a national or local ethical review committee or its equivalent, should satisfy themselves that the proposed research meets their own ethical requirements.

CONTRACT BETWEEN BENJAMIN AND THE CONGREGATION

253

pological Association, Alan Benjamin abides by its Code of Ethics, which places the welfare of the people studied as the first responsibility of anthropologists. 3. Alan Benjamin accepts that The Congregation can evaluate better than he whether information will have a clear, adverse effect to the reputations of The Congregation or its members. The Congregation accepts that Alan Benjamin can evaluate better than they whcther, and which, information, interpretations, and conclusions will be significant to a scholarly audience, thus, worthy of publication. THEREFORE EACH PARTY: DECLARES T O HAVE AGREED AS FOLLOWS: I. As regards information gathered concerning United Netherlands Portuguese Congregation Mikvk Israel-Emanuel, The Congregation will lend its cooperation to Alan Benjamin in the aforementioned project by encouraging its membership to give information to him, provided such information can and is allowed to be made public. 2. Alan Benjamin shall, in the use of such information, adhere, but not be limited, to the data as provided by members of The Congregation, and shall not include any data in his writings which are not verified as correct. 3. A. The draft of any publication shall, as far as it concerns United Netherlands Portuguese Congregation Mikvk Israel-Emanuel and its members, be submitted for approval to a Review Committee of five ( 5 ) representatives of The Congregation. The Congregation's Rabbi will be one of these five ( 5 ) . R. The Review Committee shall, within a period of two ( 2 ) weeks after the receipt of a maximum of forty (40) pages of double-spaced text, decide whether approval for release should be given or not, and Alan Benjamin will be informed of this decision in writing. Should approval be refused, then Alan Benjamin will be given the reason therefore. The two (2) week period can be extended directly in proportion to a larger number of pages having been submitted at one time. C. The Congregation's right of prior approval does not imply a right to academic or intellectual censorship of Alan Benjamin's writings. D. The Review Committee shall not withhold its approval in matters regarding Alan Benjamin's interpretations andlor conclusions drawn from factual information, provided those facts are verified as correct per Article two (2) above, and provided that these interpretations andlor conclusions will not, according to the Review Committee, adversely affect the reputation of any individual, organization or company which is directly or indirectly connected with The Congregation. E. The Review Committee is not required to approve term papers, proposals andlor drafts of the dissertation which Alan Benjamin will defend before an academic committee, provided that such material submitted to these parties andlor committee, shall explicitly carry the instruction that the material is "FOR YOUR EYES ONLY.'' F. Any published or publicly distributed material (including the final, bound version of Alan Benjamin's dissertation) will, however, need the prior written approval of the Review Committ-ee of The Congregation, according to the terms of this article. 4. Disputes arising between parties as to whether approval rightly has been refused or not shall be submitted, solely in last resort, to binding arbitration per the following procedure: A. Each party will, within two (2) weeks of being invited to do so, appoint one (I) ar-

25.4

APPENDIX E

bitrator. A third (3rd) arbitrator, who will also function as presiding arbitrator, will be app i n t e d by the two ( 2 ) arbitrators. B. The presiding arbitrator shall preside at all hearings and will ensure that all proceedings are e r i e d o n in an orderly fashion. C. The arbitrators will render their decision within six (6)weeks of the time the third (3rd) arbitrator was appointed. 5. Should any data obtained about United Netherlands Portuguese Congregation Mikvi Israel-Emanuel on C u r a ~ a obe published by Alan Benjamin in violation of the terms of this contract, then he shall thereby forfeit a fine to the amount of NAf. y.000,-per day payable to Congregation Mikvk Israel-Emanuel. This fine will be computed from the day of publication up to and including the day the last copy of the publication has been withdrawn. 6. This agreement shall be governed by the laws of the Netherlands Antilles and any disputes arising in connection with the present agreement and its implementation shall be submitted, to the exclusion of anyone else, to the cognizance of the competent court in the Netherlands Antilles. Thus agreed and signed in duplicate o n Curaqao, on

President (signature) and Secretary (signature) on behalf of, United Netherlands Portuguese Congregation Mikvk Israel-Emanuel, Curaqao, Netherlands Antilles

Alan Benjamin

256 APPENDIX F

For the individual, an ethnic identity can provide a sense of belonging, customs and traditions by which to regulate one's life, and systems of value, justice, and spirituality. Attachment to an erhnic idenrity can provide a sense of one's place in a complex world. Furthermore, every person experiences ethnic identity in their own way. Collective ethnic practices do not preclude the existence of a variety of individual experiences. Those of us who live in large, complex societies often can claim more than one ethnic identity. Within limits, we can even change to a new ethnic identity. This ethnic intermingling produces changes in the collective practices of the ethnic groups. Ethnic intermingling also produces people with multiple ethnic identities. For such people, there will be variation in the importance of the ethnic identities. Field research in anthropology, called "ethnography," centers around a method called "participant-observation." The method of parricipant-observation entails a process of socializarion into the group one is studying. Socialization beyond merely superficial practices requires a long period of time, which is why anthropological research commonly lasts for a year or more. Participation in group life over a long period of time provides opportunities to learn the unspoken-taken for granted-knowledge, values, and assumptions of the group. In addition, by living on location over an extended period of time, the participant-observer can share some of the experiences of the group being studied. Then, writings can be situated in the context of everyday activities, and in the ebb and flow of daily life. The "observation" side of participant-observation implies a process of note-taking. Thus, a major part of research data will be the notes I take o n my daily life, conversations, and activities. Participant-observation will be supplemented by a variety of interviews with members of the Mikvt Israel-Emanuel community, members of the Shaare Tsedek community, and non-Jewish Curaqaoans. The natural outcome of this research project is that some of the things I learn eventually will be published. I can assure you that I will not ~ u b l i s hreal names or material that I know to be personal and private. In addition, I will try to mask identifying data as much as possible. Your participation in this effort is, of course, entirely volunrary. The Mikvk Israel-Emanuel community seems to have several advantages as a place for research into the blending of ethnic identities. Both Curagaoan and Sephardi Jewish identities (I) are emotionally salient in relatively equivalent degrees to members of the Community; ( 2 ) involve specific and fairly well differentiated culiural expertnesses; (3) are relatively equally prestigious locally; and (4) are not mutually exclusive. In addition, the members of Mikvk Israel-Emanuel and I have enough in common so that some collective practices will be familiar to me, yet we are different enough that our differences will make apparent some of our taken-for-granted cultural conventions. Many members of Mikvk Israel-Emanuel descend from a long line of ancestors who lived on Curagao. It is rare ro find in the United States a Jewish community as rooted in its locale. I have the impression that both Jewish and Curaqaoan identities are deeply important to many ofyou. I assume that in the course of daily life you blend these two identities. Moreover, I anticipate, although such blending m ~ seem y mundane, that explicaring the processes of blending ethnic identities will be a significant contribution to the under-

LETTER FROM BENJAMIN TO THE CONGREGATION

257

standing of social processes. Research into the processes by which multiple ethnic identities are blended is broadly applicable both to Jews and to other ethnic groups. It is relevant at a time of worldwide ethnic strife; it is relevant to a world in which widely diverse peoples frequently come into contact; finally, it is relevant to those who experience in their own lives the complexity of adapting to multiple cultural patterns and expectations. Thank you for your coopcration in this research project. If you have any questions, comments, or criticisms please bring them to me. I wish you to understand my purpose as well as possible. I also want to come up with the most well-informed and well-thought piece that I can, so your criticisms are as important as your agreement. 1 look forward to an enjoyable and productive year. Sincerely, Alan F. Benjamin, M.A.

References

Abramson, N. S., A. Meisel, and P. Safar. 1986. "Deferred consent: A new approach for resuscitation research on comatose patients." Journal of the American MedicalAssociation 255: 2466-71. Abramson, N. S., and P. Safar. 1990 "Deferred consent: Use in clinical resuscitation research." Annals ofEmergency Medicine 19 (7): 781-84. Adair, L. S., and B. M. Popkin. 1992. "Prolonged lactation contributes to depletion of maternal energy reserves in Filipino women." Journal ofNutrition 122: 1643-55. Adair, L. S., B. M. Popkin, and D. K. Guilkey. 1993. "The duration of breast-feeding: How is it affected by biological, sociodemographic, health sector, and food industry factors?" Demography 30: 63 - 80. Adair, L. S., B. M. Popkin, J. VanDerslice et al. 1993. "Growth dynamics during the first two years of life: A prospective study in the Philippines." EuropeanJournal of Clinical Nuhition 47:42-51. Advisory Committee on Human Radiation Experiments. 1995. Final Report. Washington, D.C.: U.S. Government Printing Office. Also published as The Human Radiation Experiments, Oxford University Press, 1996. Agathonos, H. 1983. "Institutional child abuse in Greece: Some preliminary findings." Child Abuse and Neglect 7: 71-74. Agathonos, H., N. Stathacopoulou, H. Adam et al. 1982. "Child abuse and neglect in Greece: Sociomedical aspects." ChildAbuse and Neglect 6 : 307-11. Airhihenbuwa, C. O., R. J. DiClemente, G. M. Wingood et al. 1992. "HIVIAIDS education and prevention among African-Americans: A focus on culture." AIDS Education and Prevention 4 (3): 267-76. Akin, J. S., C. Griffin, D. K. Guilkey et al. 1985. The Demandfor Primay Health Sewices in the Third World. Totowa, N.J.: Rowman and Allanheld. Aldarondo, E., and M. A. Straus. 1994. "Screening for physical violence in couple therapy: Methodological, practical and ethical considerations." FAMPROC 33 :425-39. American Medical Association, Council on Ethical and Judicial Affairs. 1996. Code ofMedical Ethics, Current Opinions with Annotations, p. 15 (Opinion z.o7(4)B.iii.). Chicago: American Medical Association. American Anthropological Association. 1990. Principles of Profssional Responsibility: Adopted by the Council of the American AnthropologicalAssociation. May 1971, as amended through October 1990. Angell, M. 1988. "Ethical imperialism?: Ethics in international collaborative clinical research." New EnglnndJournal ofMedicine 319 (October 20): 1081-83.

REFERENCES

259

. 1997. "The ethics of clinical research in the third world." New EnglandJournal of Medicine 337: 847-49. Annas, G. J. 1988.JudgingMedicine. Clifton, N.J.: Humana Press.

. 1989. "Faith (healing), hope and charity at the FDA: The politics of AIDS drug trials." Villanova Law Review 34: 771. . 1993. Standard of Care: The Law ofAmerican Bioethics. Oxford: Oxford University Press. . 1996. "Questing for grails: Duplicity, betrayal and self-deception in postmodern medical research." Journal of Contemporary Health Law and Poli~y12:297-324, Appell, G. N. 1978. Ethical Dilemmas in Anthropological Inquiry: A Case Book. Waltham, Mass.: Crossroads Press, African Studies Association, Brandeis University. Aracena, M., E. Balladares, and F. Roman. 1994. "Conceptualizaci6n de las pautas de crianza, maltrato infantil, y sus factores de riesgo en familias del estrato sociecon6mico bajo: una mirada cualitativa." Unpublished paper, Unidad de Epidemiologia Clinica, Deparramento de Psicologia. Temuco, Chile. Arnold, E. 1982. "The use of corporal punishment in child rearing in the West Indies." Child Abuse and Neglect 6 :141- 45. Arras, J. D., and B. Steinbock. 1995. "Moral reasoning in the medical context." In Ethical Issues in Modern Medicine, 4th ed., edited by J. D. Arras and B. Steinbock, pp. 1-39. Mountain View, Calif.: Mayfield Publishing. Bankowski, Z., and R. J. Levine, eds. 1992. Proceedings of the XXVth CIOMS Conference on Ethics and Research on Human Subjects: International Guidelines. Geneva, Switzerland: Council for International Organizations of Medical Sciences. Barnes, J. A. 1979. Who Should Know What? Social Science, Privacy and Ethics. New York: Penguin Books. Barry, M. 1988. "Sounding board-ethical considerations of human investigation in developing countries: The AIDS dilemma." New England Journal ofMedicine 319 :1083-85. Baumrind, D. 1978. "The nature and definition of informed consent in research involving deception." In National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, The Belmont Report, Vol. z (Appendix), pp. 23-52, Washington, D.C.: Department of Health, Education, and Welfare. Beauchamp, T. L., and J. F. Childress. 1994. Principles ofBiomedica1Ethics. Oxford: Oxford University Press. Beauchamp, T. L., R. R. Faden, R. J. Wallace, Jr., et al., eds. 1982. Ethical Issues in Social Science Research. Baltimore: Johns Hopkins University Press. Becker, H. S. 1967. "Whose side are we on?" SocialP~oblems14:239-47. Belshe, R. B., M. L. Clements, and M. C. Keefer. 1994. "Interpreting HIV serodiagnostic test results in the 1990's: social risks of HIV vaccine studies in uninfected volunteers." Annals of Internal Medicine 121: 584 - 89. Benjamin, A. F. 1990. "Ethnicity, experience, and sacred space: A view of the Jewish conununity of Curaqao through the 'lens' of a synagogue." Unpublished paper, Department of Anthropology, University of North Carolina. . 1996. "Ethnic Identities and Classifying Practices among Jews of Curaqao." Ph.D. dissertation, University of North Carolina. Berman, G. 1982. "Justifying social science research in terms of social benefit." In Ethical Zssues in Social Science Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 125-43. Baltimore: Johns Hopkins University Press.

260

REFERENCES

Bernard, C. 1957. An Introduction to the Study ofExperimentalMedicine. New York: Dover. Biller, H. B., and R. S. Solomon. 1986. "Child maltreatment: The definitional dilemma." In Child Maltreatment and Paternal Deprivation. A Manifesto for Research, Prevention, and Treatment, edited by H . B. Biller and R. S. Solomon, pp. 9-20. Lexington, Mass.: Heath and Company. Biros, M. H., R. J. Lewis, and C. M. Olson. 199s "Informed consent in emergency research: Consensus statement from the coalition conference of acute resuscitation and critical care researchers."Journal of the American Medical Association 273 :1283- 87. Blumenthal, D. G., M. Gluck, K. S. Louis, M. A. Stoto, and D. Wise. 1986. "Universityindustry research relationships in biotechnology: Implications for the university." Science 232: 1361- 66. Bok, S. 1995. "Shading the truth in seeking informed consent for research purposes." Kennedy Institute of Ethics Journal 5: 1-17. Bowman, P. J. 1991. "Race, class and ethics in research: Belmont principles to functional relevance." In Black Psychology,edited by R. L. Jones, pp. 747- 66. Berkeley, Calif.: Cobb and Henry. Brandt, A. M. 1978. "Racism and research: The case of the Tuskegee syphilis study." Hastings Center Report 8 (6): 21-29. Brazziel, W. F. 197%"White researchers in Black communities: When solutions become a part of the problem." Journal of Social Issues 29 (I): 41- 44. Brettell, C. B., ed. 1993. When They Read What We Write: The Politics ofEthnography. Westport, Conn.: Bergin and Garvey. Bronfenbrenner, U. 1952. "Principles of professional ethics: Cornell studies in social growth." The American Psychologist 7 : 452-55. Brown, D., and B. Weiser. 1994. "HIV infections cast pall over expanding vaccine experiments." Washington Post: A7, May 30. Bussell, D. 1994. "Ethical issues in observational family research." Fam Pro 33: 361-76. Callahan, D. 1984. "Autonomy: A moral good, not a moral obsession." Hastings Center Report 14 (5): 40-42. Capron, A. M. 1982. "IS consent always necessary in social science research?" In Ethical Issues in Social Science Research, edited by T. L. Beaucharnp, R. R. Faden, J. Wallace, Jr., er al., pp. 215-31. Baltimore: Johns Hopkins University Press. Cassell, J. 1982. "Does risk-benefit analysis apply to moral evaluation of social research?" In Ethical Issues in Social Science Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 144- 62. Baltimore: Johns Hopkins University Press. . 1987. "Cases and comments." In Handbook on Ethical Issues in Anthropology, edited by J. Cassell and S.-E. Jacobs, pp. 37-75. Washington, D.C.: American Anthropological Association. Cassell, J., and S.-E. Jacobs, eds. 1987. Handbook on EthicalIssues in Anthropology. Special no. 23, Washington, D.C.: American Anrhr~polo~ical Association. Cebu Study Team. 1991. "Underlying and proximate determinants of child health: The Cebu longitudinal health and nutrition study." American Journal of Epidemiology 133: 185-201. Cebu Study Team. 1992. "A child health production function estimated from longitudinal data." Journal of Development Economics 38 :323 -51. Central Bureau of Statistics. 1993 The Netherlands Antilles Statistical Orientation. Curaqao: Central Bureau of Statistics. Chatters, L. M. 1993. "HIVIAIDS within African American communities: Diversity and interdependence." Health Education Quarterly 20 (3): 321-26.

REFERENCES 261

Chren, M. M., C. S. Landefeld, and T. H. Murray. 1989. "Doctors, drug companies, and gifts." Journal of the American MedicrllAssociation 262 (q): 3448- 51. Christakis, N. A. 1992. "Ethics are local: Engaging cross-cultural variation in the ethics for clinical research." Social Science and Medicine 35 (9): 1079 -91. Churchill, L. R. 1997. "Bioethics in social context." In The Philosophy ofMedicine and Bioethics, edited by R. Carson and C. Burns, pp. 137- 51. Dordrecht, The Netherlands: Kluwer Academic Publishers. Clark, C.-X. 1973. "The role of the White researcher in Black society: A futuristic look." Journal of Social Issues 29 (I): 109-18. Clifford, J. 1988. The Predicdment of Culture: Twentieth-Centmy Ethnograp& Literature, and Art. Cambridge, Mass.: Harvard University Press. Clouser, K. D., and A. H. Hawkins. 1996. "Literature and medical ethics." Journal of Medicine and Philosophy 21: 237- 41. Cohen, J. 1991. "AIDS vaccine trials: Bumpy roads ahead." Science 25: 1312-13. Conquergood, D. 1991. "Rethinking ethnography: Towards a critical cultural politics." Cumrnunicatiun Monographs 58:179-94. Cornwall, A., and R. Jewkes. 1995. "What is participatory research?" Social Science and Medi1667-76. cine 41: Couchman, I. S. B. 1973. "Notes from a White researcher in Black society" Journal ofsocial Issues 29 (I): 4 j -52. Cournand, A. 1977. "The code of the scientist and its relation to ethics." Science 198: 699-705. Curtis, P. A., J. D. Boyd, M. Liepold et al. 1995. ChildAbwe and Neglect: A Look at the States. Washington D.C.: Child Welfare League of America. Dalton, H. L. 1989. "AIDS in blackface." Daedalus: Journal of the American Acadtmy. o.f A m and Sciences (Summer): 205-28. . 1990. "AIDS in blackface." In Living with AIDS, edited by S. R. Graubard, pp. 23758. Cambridge, Mass.: MIT Press. Dancy, B. L. 1991. "The development of an ethnically sensitive and gender-specific AIDS questionnaire for African-American women." Health Values 15 (6): 49-54. De La Cancela, V. 1989. "Minority AIDS prevention: Moving beyond cultural perspectives towards sociopolitical empowerment." AIDS Education and Prevention r (2): 141-53. De la Pefia, P., and L. Gonzbla. 1989. "Una universidad regional en el context0 rural. Plan integrado de educaci6n rural." Unpublished paper, Universidad de la Frontera, Temuco, Chile. Department of Health and Human Services. 1981. "Final Regulations Amending Basic HHS Policy for the Protection of Human Research Subjects." (Codified in the Code ofFederdE Regulatzons, vol. 45 Part 46.) Federal Register 46:8386-91. . 1991a. "Common Rule §-.roz(d)." Federal Register 56 (June 18): 28013. . 1991b. "Con~monRule %-.116(d)." Federal Register 56: 28017. . 1995a. "Action related to emergency research activity." Federal Regiiter 60 (143): 38353. . 199jb. "Protection of human subjects; informed consent." Federal Register 60 (183): 49086-103. . 1995~."Protection of human subjects; informed consent; proposed rule." Federal Register 60 (183): 49086-103. . 1996a. Federal Register 61: 51505. . 1996b. "Protection of human subjects; informed consent." Federal Registtpr 61 (October 2): 51497- 531.

262

REFERENCES

. r996c. "Waiver of informed consent requirements in certain emergency research." Federal Register 61: 51531-33. De Silva, W. 1981. "Some cultural and economic factors leading to neglect, abuse and violence in respect of children within the family in Sri Lanka." ChildAbuse and Neglect 5 :391-405. Des Jarlais, D. C., C. Casriel, B. Stepherson et al. 1991. "Expectations of racial prejudice in AIDS research and prevention programs in the United States." In AIDS andAlcoholl DrugAbuse: Psychosocial Research, edited by D. G. Fisher, pp. 1-7. New York: Haworrh Press. De Ville, K. A. 1995. "Commercialism in scientific research." In Encyclopedia ofBioethics, 2nd ed., edited by W. T. Reich, Vol. I., pp. 412-17. New York: Macmillan. Direcci6n de Maternidad e Infancia, Argentina. 1992. Mortalidad materna en areas seleccionah. Serie 8, No. 13. Dobbing, J., eds. 1988. Infdnt Feeding: Anatomy ofa Controversy 1973 -84. New York: SpringerVerlag. Dubanoski, R. A., and K. Snyder. 1980. "Patterns of child abuse and neglect in Japaneseand Samoan-Americans." ChildAbuse and Neglect 4: 217-25. Durbin, P. T. 1988. Dictionary of Concepts in the Philosophy of Science. Westport, Conn.: Greenwood Press. Dworkin, G. 1982. "Must subjects be objects?" In Ethical Issues in SocialScience Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 246-54. Baltimore: Johns Hopkins University Press. Edgar, H., and D. J. Rothman. 1990. "New rules for new drugs: The challenge of AIDS to the regulatory process." Milbank Mem. Fund Q 68 (Supp. I): 111-42. Elazar, D. J. 1989. The OtherJews: The Sephardim Today. New York: Basic Books. Elms, A. C. 1982. "Keeping deception honest: justifying conditions for social scientific research stratagems." In EthicalIssues in Social Science Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 232-45. Baltimore: Johns Hopkins University Press. El-Sadr, W., and L. Capps. 1992. "The challenge of minority recruitment in clinical trials for AIDS." Journal of the American MedicalAssociation 267 (7): 954-57. Jewry 16~6-1917. New Emmanuel, I. S. 1957. Precious Stones of the Jews of Curacao: Cura~aoan York: Bloch Publishing Company. Published by authority of the Wardens of the Portuguese Jewish Congregation M i k d Israel, Cura~ao. Emmanuel, I. S., and S. A. Emmanuel. 1970. History ofthe Jews ofthe Netherlands Antilles. Cincinnati: American Jewish Archives. Estroff, S. E. 1995. "Whose story is it anyway?Authority, voice, and responsibility in narratives of chronic illness." In Chronic Illness and Disability: From Experience to Policy, edited by R. Carson, K. Toombs, and D. Barnard, pp. 78-104. Bloomington: Indiana University Press. Faden, R. R., and T. L. Beauchamp, with N. M. P. King. 1986. A History and Theory ofInformed Consent. New York: Oxford University Press. Farinatti, F. A., N. M. Fonseca, M. Dondonis et al. 1990. "Child abuse and neglect in a developing country." Child Abuse and Neglect 14: 133-34. Farrer, C. 1976. "Fieldwork ethics." In Folklore Forum, Bibliographic and Special Series 15 (9): 59-63. Fernandez, H. 1995. "El maltrato infantil en el context0 de la violencia intrafamiliar. Una

REFERENCES

263

perspectiva de andisis legal." Unpublished paper, Ministerio de Justicia, Servicio Nacional de Menores. Fernandez, M. E., and B. M. Popkin. 1988. "Prelacteal feeding patterns in the Philippines." Ecohgy of Food and Nutrition 21 :303 -14. Fine, G. A. "Ten lies of ethnography." 1993.Journalof ContemporaryEthnography zz:267-94. Finkelhor, D., and J. Korbin. 1988. "Child abuse as an international issue." ChildAbwe and Neglect 12 :3-23. Flaskerud, J. H., and R. C. E. Flaskerud. 1989. "AIDS and traditional health beliefs and practices of Black women." Nursing Research j8 (4): 210-15. Fluehr-Lobban, C. 1991a. "Ethics and professionalism: A review of issues and principles within anthropology." In Ethics and The Profession ofAnthropology: DDiahgueforA New Era, edited by C. Fluehr-Lobban, pp. 15-35. Philadelphia: University of Pennsylvania Press. , ed. 1991b. Ethics and the Profession ofAnthropology:Dialogue for a N m Era. Philadelphia: University of Pennsylvania Press. Fontes, L. A. 1993 "Disclosures of sexual abuse by Puerto Rican children: Oppression and cultural barriers." Journalof Child SexualAbuse z (I): 21-35. Foster, P M., F. Phillips, F. 2.Belgrave et al. 1993. "AnAfricentric model for AIDS education, prcvention, and psychologicai scrvices within the hfrican Amcrican community." Journal of Black Ps~~hology 19 (2): 123- 41. Fox, R. C., and J. P. Swazey. 1978. The Courage to Fail: A Social View of Organ Transplants and Diulysis. Chicago: University of Chicago Press. Fraser, G., and P. L. Kilbride. 1980. "Child abuse and neglecr-Rare, but perhaps increasing, phenomena among the Samia of Kenya." ChildAbuse and Neglect 4: 227-32. Friedman, S. R., J. L. Sotheran, A. Abdul-Quader et al. 1987. "The AIDS epidemic among Blacks and Hispanics." The Milbank Quarterly 65 (Supp. 2): 455-99. Friedrich, W N., P Grambsch, D. Broughton et al. 1991. "Normative sexual behavior in children." Pediatrics 88 (3): 456- 64. Fry, G.-M. 1975. Night Riders in Black Fo'olk History. Knoxville: University of Tennessee Press. Gallup Organization. 1995. Disciplining children in America. Princeton, N.J.: The Gallup Organization. Geertz, C. 1973. The Interpretation of Cultures: Selected Essays by CEifford Geertz. New York: Basic Books. . 1988. "Being here: Whose life is it anyway?" In Works and Lives: The Anthropologist As Author, pp. 129-49. Stanford: Stanford University Press. Gelles, R. J. 1982. "Problems in defining and labeling child abuse." In ChildAbwe Prediction. Policy Implicutions, edited by R. H. Starr, pp. 1-30. Cambridge, Mass.: Ballinger Publishing. . 1987. "What to learn from cross-cultural and historical research on child abuse and neglect: An overview." In ChifdAbuse and Neglct. Biosocial Dimensions, edited by R. J. Gelles and J. B. Lancaster, pp. 15-30. New York: Aldine de Gruyter. Gelles, R J., and C. P. Cornell. 1983. "International perspectives on child abuse." ChildAhuse and Neglect 7 :375 - 86. Glantz, L. H. 1992. "The influence of the Nuremberg Code on U.S. statutes and regulations." In The Nazi Doctors and The Nuremberg Code: Human Rights in Human Experimentation, edited by G. J. A n n a and M. A. Grodin, pp. 183-200. New York: Oxford University Press.

264 REFERENCES Gluckman, M. 1962. "Les rites de passage" (Rites of passage). In Essays on the Ritual of Social Relntions, edited by M . Gluckman, pp. 1-52. Manchester: Manchester University Press. Goffman, E. 1959. The Presentation of Selfin Everyday Life. Garden City, N.Y.:Doubleday. Goldberger, P. 1994. "Philip Morris Calling In 1.O.U.s in the Arts." New York Times:AI, August 5. Gomes Casseres, C. 1990. Istoria kortiku di hudiunan di Korsou (Short history of the Jews of Curaqao). 2nd ed. Willemstad, Cura~ao:Ch. Gomes Casseres. Costin, L. 1991. "Ethical principles for the conduct of human subject research: Populationbased research and ethics." Law, Medicine andHealth Care 19 (3-4): 171-201. Gould, S. J. 1981. TheMismeasure ofMan. New York: W. W. Norton. Graves 111, W., and M. A. Shields. 1991. "Rethinking moral responsibility in fieldwork: The situated negotiation of research ethics in anthropology and sociology." In Ethics and the Profission ofAnthropology: Dialoguefor a New Era, edited by C. Fluehr-Lobban, pp. 132-51. Philadelphia: University of Pennsylvania Press. Guilkey, D. K., and J. F. Stewart. 1995. "Infant feeding patterns and the marketing of infant foods in the Philippines." Economic evel lo pent and Cultural Change qj:j69-99. Guinan, M. 199%%lack communities' belief in 'AIDS as genocide': A harrier to overcome for HIV prevention." Annals ofEpidemiology 3: 193-95. Haas, L. J., J. L. Malouf, and N. H. Mayerson. 1988. "Personal and professional characteristics as factors in psychologists' ethical decision making." Profissional Psychology Research and Practice 19 (I): 35-42. Haffejee, I. E. 1991. "Sexual abuse of Indian (Asian) children in South Africa: First report in a community undergoing cultural change." Child Abuse and Neglect 15:147-51. Hamilton, S., B. M. Popkin, and D. Spicer. 1984. Women andNutrition in Third World Countries. New York: Praeger. South Hadley Mass.: Bergin and Garvey. Hampton, R. L., and E. H. Newberger. 1985. "Child abuse incidence and reporting by hospitals: Significance of severity, class, and race." American Journal ofPublic Health 75: 56 - 60. Harrell-Bond, B. 1976 "Studying elites: Some special problems." In Ethics andAnthropoloP Dilemmas in Fieldwork, edited by M . A. Rynkiewich and J. P. Spradley, pp. 110-22 (1981 edition). Malabar, Fla.: Robert E. Krieger. Hatch, J., N. Moss, A. Saran et al. 1993. "Community research: Partnership in black communities." American Journal ofpreventive Medicine 9 (Supp.): 27-31. Henriquez, M. 1988. Ta asina! o ta asana?abla, uzo i kwtumbm Sefardi (This way or that way? Sephardi speech, use, and customs). Willemstad, Curapo: May Henriquez. Herner and Co. 1978. ChikiAbuse and Negkt: State Reporting Laws. Narional Center on Child Abuse and Neglect. Washington, D.C., Department of Health, Education, and Welfare. May. Higman, H. 1978. "The Effect on Social Science Research of Restricted Use of Human Subjects." In Safpguardsfor Human Subjects in Research: A Seminar, pp. 22-31, Washington, D.C.: American Association for the Advancement of Science. Hilts, P. J. 1996. Smokescreen: The Truth Behindthe Tobacco Industry Cover-Up. Reading, Mass.: Addison-Wesley. Homan R. 1992. "The ethics of open methods." British Journal ofSociology 43: 321-32. Hunter, K. M. I ' ) ~ IDuct~n'Stories: . T j ~ Narrative e Structure oJM~dicalKnowledge. Prin~eton: Princeton University Press. . 1995. "Narrative." In Encydopedia ofBioethirs, revised ed., edited by W. T. Reich, pp. 1789-94. New York: Simon and Schuster Macmillan.

REFERENCES

265

Hutchinson, J. 1992. "AIDS and racism in America." Journal of the National Medical Association 84 (2): 19-124. Ijsselmuiden, C., and R Faden. 1992. "Research and informed consent in Africa-another look." New England Journal of Medicine 326 :830 -34. Ikeda, Y. 1982. "A short introduction to child abuse in Japan." ChildAbwe andNeglect 6: 487-90. Israel, B. A., A. J. Schultz, E. A. Parker, and A. B. Becker. 1998. "Review of CommunityBased Research: Assessing Partnership Approaches to Improve Public Health." Annual Review of Public Health 19:173-202. Jacobs, S.-E. 1987. "Cases and solutions." In Handbook on Ethical Issues in Anthropolo~, edited by J. Cassell and S.-E. Jacobs, pp. 20 -36. Washington, D.C.: American Anthropological Association. Jinadu, M. K. 1986. "Combating child abuse and neglect in developing and newly industrializing countries: A unique primary health care approach." ChildAbuseand Neglect Jonas, H. 1969. "Philosophical reflections on human experimentation." Daedalus 98: 219. Jones, J. H. 1992. "The Tuskegee legacy: AIDS and the Black community." Hastings Center Report 22 (6): 38-40. . 1993. Bad Blood The Tuskegee Syphilis Experiment. New York: Free Press. Jonsen, A., and S. Toulmin. 1988. TheAbwe ofCasuistry. Berkeley, Calif.: Universiry of California Press. Kalichman, S. 1993. Mandated Reporting ofsuspected ChildAbuse. Washington, D.C.: American Psychological Association. Kalichman, S. C., and M. E. Craig. 1991. "Professional psychologists' decisions to report suspected child abuse: Clinician and situation influences." Professional Pychology: Research and Practice 22 ( I ) : 84 -89. Kalichman, S. C., M. E. Craig, and D. R. Follingsrad. 1988. "Mental health professionals and suspected cases of child abuse: An investigation of factors influencing reporting." Community Mental Health Journal 24 ( I ) : 43 -51. .1989. "Factors influencing the reporting of father-child sexual abuse: Study of licensed practicing psychologists." Profersional Pychology:Research and Practice 20 (2): 84 - 89. . 1990. "Professionals' adherence to mandatory child abuse reporting laws: Effects of responsibility attribution, confidence ratings, and situational factors." ChildAbuseand Neglect 14:69 -77. Karner, F. P. 1969. The Sephardics of Curapo: A S t d y of Socio-CulturalPatterns in F h . Assen, The Netherlands: Van Gorcum. Kau,J. 1972. Eqerimentation with Human Beings: The Authority of the Investigator, Subject, Profssions, and State in the Human Experimentation Process. New York: Russell Sage. . 1993a. "Ethics and clinical research revisited: A tribute to Henry K. Beecher." Hastings Center Repovt y ( 5 ) : 31-39. 1993b. "Human experimentation and human rights." Saint Louis Univmiy Law Journal 38 :7-54. Kelman, H. C. 1968. A Time to Speak. San Francisco: Jossey Bass. . 1982. "Ethical Issues in Different Social Science Methods." In Ethical Issues in Social Science Research, editcd by T. L. Beauchamp, R R. Faden, and R. J. Wallace, Jr., et al., pp. 40 -97. Baltimore: Johns Hopkins University Press. Kempe, C . H. 1978. "Recent developments in the field of child abuse." Child Abuse andNeglect 2:261- 67.

266

REFERENCES

. 1982. "Cross-cultural perspectives in child abuse." Pediatrics 69 :497-98. Kempe, C. H., F. N. Silverman, B. F. Steele et al. 1962. "The battered-child syndrome."Journal of the American MedicalAssociation 181:17-24. King, N. M. P. 1995. "Experimental treatment: oxymoron or aspiration?" Hastings Center Report 25 (4): 6-15. King, N. M. P., and G. E. Henderson. 1991. "Treatments of last resort: Informed consent and the diffusion of new technology." Mercer Law Reuiew 42:1007-50. Kittay, E. F., and D. T. Meyers. 1987. Women and Moral Theory. Totowa, N.J.: Rowan and Littlefield. Kline, A., E. Kline, and E. Oken. 1992. "Minority women and sexual choice in the age of AIDS." Social Science and Medicine 34 (4): 447-57. Koff, W C., and D. F. Hoth. 1988. "Development and testing of AIDS vaccines." Science Kopelman, L. M. 1994. "Case method and casuistry: The problem of bias." TheoreticalMedicine 15 (I): 21-37, 1994. . 1995. "Philosophy and medical education." Academic Medicine 70: 795-805. Korbin, J. E. 1980. "The cultural context of child abuse and neglect." ChildAbuse and Neglect 4:3-13. Korbin, J. E., ed. 1981. Child abuse and neglect: Cross-culturalperpectivps. Berkeley: University of California Press. Kuhn, T. S. 1970. The stmcture ofscientific revolutions. znd ed. Chicago: University of Chicago Press. La Vertu, D., and A. Linares. 1990. "Ethical principles of biomedical research on human subjects: Their application and limitations in Latin America and the Caribbean." Bulletin of PAHO 24 (4): 469-79. Lawless, E. J. 1992. "'I was afraid someone like you . . . an outsider . . . would misunderstand': Negotiating interpretive differences between ethnographers and subjects." Journal ofAmerican Folklore 105 :302-14. Levine, R. J. 1988. Ethics and the Regulation of Clinical Research. 2nd ed. New Haven: Yale University Press. . 1994. "The impact of HIV infection on society's perception of clinical trials." Kennedy Institute ofEthics Journal 4: 93 -98. . 1995. "Research in emergency situations: The role of deferred consent." Journalof the American MedicalAssociation 273 (16): 1300-1302. Li, X. 1996. "'Asian values' and the universality of human rights." China Rights Forum (Fall) :jz-35. Liakopoulou, M., and D. Xypolyta. 1989. "Sexual abuse in Greece: Management in two different settings." ChildAbuse and Neglect 13 (2): 149-15. Liebow, E. 1993. Tell Them Who I A m . New York: Free Press. Luce, J. 1990. "Ethical principles in critical care." Journal of the American Medical Association 263 :696-700. Luebke, P. 1991. Tarheel Politics: Myths and Realities. Chapel Hill: University of North Carolina Press. Lurie, P., and S. M. Wolff. 1997. "Unethical trials of interventions to reduce perinatal transmission of the Human Immunodeficiency Virus in developing countries." New England Journal ofMedicine 337: 853 -56. MacIntyre, A. 1982. ''Risk, harm, and benefit assessments as instruments of moral evaluation."

REFERENCES 267

In Ethical Issues in Social Science Research, edited by T. L. Beauchamp, R. R Faden, J. Wallace, Jr., et al., pp. 175- 89. Baltimore: Johns Hopkins University Press. . 1984.Afier Virtue. Notre Dame, Ind.: University of Notre Dame Press. . 1991. "Ethical dilemmas: Notes from outside the field." Anthropology Newsletter 34: 5-6. Macklin, R 1977. "Moral progress." Ethics 87 :370 - 82. . 1982. uThe problem of adequate disclosure in social science research." In Ethical& sues in Social Science Research, edited by T. L. Bcauchamp, R. R. Faden, J. Wallacc, Jr., et al., pp. 193-~14.Baltimore: Johns Hopkins University Press. . 1992. "Universality of the Nuremberg Code." In The Nazi Doctom and the Nuremberg Code: Human Rights in Humdn Experimentation, edited by G. J. Annas and M. A. Grodin, pp. 240-57 New York: Oxford University Press. MacQueen, K. M., S. Buchbinder, J. M. Douglas et al. 1994. "The decision to enroll in HIV vaccine efficacy trials: Concerns elicited from gay men at increased risk for HIV infection." AIDS Research and Human Retroviruses 10 (Supp. 2): sz61-s64. Madhubuti, H. R 1990. Black Men: Obsolete, Single, Dangerous?Essays in Discovery, Solution, and Hope. Chicago: Third World Press. Mahowald, M. B. 1993. Women and Children in Health Care: An UnequalMujorig. Oxford: Oxford University Press. Mann, B. J. 1976. "The ethics of fieldwork in an urban bar." In Ethics undAnthropology Dilemmas in Fieldwork, edited by M . A. Rynkiewich and J. P. Spradley, pp. 95-109. Malabar, Fla.: Robert E. Krieger. Marcus, G. E., and M. M. J. Fischer. 1986. Anthropology as Cultural Critique:An Experimental Moment in the Human Sciences. Chicago: University of Chicago Press. Marcus, J. R. 1953. "The West India and South America expedition of the American Jewish Archives." American Jewish Archives 5 :5-21. Mariner, W. K. 1995. "Research in emergency care without consent: New proposed FDA rules." Lancet 346 (8989): 1505-6. Marshall, l? 1992. "Research ethics in applied anthropology." IRB: A Review ofHumnn Subjects Research 14 (6): 1-5. Marwick, C. 1995. "Informed consent waiver for emergency research." Journal qftbe American MedicalAssociation 274 (15): 1184-85. Mastroianni, A. C., R Faden, and D. Federman, eds. 1994. Women and Health Research: Ethical and Legal Issues oj'lncluciing Women in Clinical Studies. Vol. I. Washington, D.C.: National Academy Press for the Institute of Medicine. Mays, V. M., and S. D. Cochran. 1987. "Acquired immunodcfici~ncysyndrome and Black Americans: Special psychosocial issues." Public Healtb Reports 102 (2): 224-31. Mays, V. M., and J. S. Jackson. 1991. "AIDS survey methodology with Black Americans." Social Science and Medicine 33 ( I ) : 47-54. McBeth, S. 1993. "Mythsof objectivity and the collaborative process in life history research." In When They Read What We Write: The Politics ofEthnography,edited by C. B. Brettell, pp. 145- 62. Westport, Conn.: Bergin and G m e y . McCarrick, P. M., and M. Darragh. 1996. "Scope note 30: Feminist perspectives on bioethics." Kennedy Institute ofEthicsJournal 6: 85-103. M a d , M. 1977. Lettersfiorn the Field: 19~5-1971.New York: Harper and Row. Melton, G. B. 1990. "Certificates of confidentiality under the Public Health Service Act: Strong protection but not enough." Vioknre and Kctirns 5 (I): 67-71.

268 REFERENCES Miller, E. 1981. "International trends in ethical review of medical research." IRB: A Review of Human Subjects Research 3 (October): 9-10. Miller, J. 1992. "Ethical standards for human subject research in developing countries." I R B A Review ofHuman Subjects Research 14 (May-June): 7-8. Muehleman, T., and C. Kimmons. 1981. "Psychologists' views on child abuse reporting, confidentiality, life, and the law: An exploratory study." Professional Psychology 12 (5): 631-38. Murray, T. H. 1980. "Learning to deceive." Hastings Center Report 10:11-14. . 1990. "The poisoned gift: AIDS and blood." Milbank Quarterly 68 (Supp. 2): 205-25. Nathan, L., and W. T. Hwang. 1981. "Child abuse in an urban centre in Malaysia." Child Abuse and Neglect 5 :241- 48. National Clearinghouse on Child Abuse and Neglect. 1995. ChildAbuse and Neglect CDROM 1961-July 1995. Baltimore: National Information Services Corporation. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. 1978. The Belrnont Report: Ethical Principles and Guidelinesfor the Protection of Human Subjects ofResearch. Government Printing Office, Washington, D.C., Department of Health, Education, and Welfare, Publication no. (OS) 78-0014. National Research Council. 1993. Understanding ChildMaltreatrnent. Washington, D.C.: National Academy of Sciences. Nelkin, D. 1982. "Forbidden research: Limits to inquiry in the social sciences." In Ethical Issues in Social Science Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 163-74. Baltimore: Johns Hopkins University Press. "Nestle Accused of False AIDS Claim." 1989. The Lancet z (8674): 1289. November 25. Newberger, E. H. 1985 "The helping hand strikes again: Unintended consequences of child abuse reporting" In Unhappy Families, edited by Newberger, E. H., and R. Bourne, pp. 171-78. Littleton, Mass.: PSG Publishing Company. Newton, L. H. 1990. "Ethical imperialism and informed consent." IRB: A Review of Human Subjects Research 12 (3): 10-11. Nightingale, N. N., and E. F. Walker. 1986. "Identification and reporting of child maltreatment by head start personnel: Attitudes and experiences." ChildAbuse and Neglect 10:191-99. North Carolina Department of Environment, Health and Natural Resources. 1994. Personal communication. AIDS Care Branch. Okeahialam, T. C. 1984. "Child abuse in Nigeria." ChildAbuse and Neglect 8 :69-73 Olson, C. M. 1994. "The letter or the spirit: Consent for research in CPR."joumal of the American MedicalAssociation 271 (18): 1445-7. OI'RR Reports. 1993. Number 93-3, August 12. Organizacion Mundial de la SaludlOrganizaci6n Panamericana de la Salud. 1990. Plan de Accidn Regionalpara la Reduccidn de la Mortalidad Materna en las Americas. Vol. CPS ~3110.Washington, D.C.: OMSIOPS. Osborne, N. G., and M. D. Feit. 1992. "The use of race in medical research." Journal of the American Medical Association 267 (2): 275 -79. Osborne, W. T. 1993. "The ethical symbiosis between anthropologists and the peoples they study." Anthropology Newsletter 34: I, 6. Patai, D. 1991. "U.S. academics and third world women: Is ethical research possible?" In Women5 Words, edited by S. B. Gluck and D. Patai, pp. 137-53. New York: Routledge. Paul, R., and J. Rudinow. 1988. "Bias relativism and critical thinking" Journal of Thought 23 : 125-38.

REFERENCES

269

Paync, M. A. 1989. "Use and abuse of corporal punishment: A Caribbean view." ChildAbuse and Neglect 13: 389-401. Peterson, J. L., and G. Marin. 1988. "Issues in the prevention ofAIDS among Black and Hispanic men." American Psychologist 43 (11): 871-77. Phillips, K. A. 1993. "Factors associated with voluntary HIV testing for African-Americans and Hispanics." AIDS Education and Prevention 5 ( 2 ) : 95-103. Phillips, S. R. 1994. "Asking the sensitive question: The ethics of survey research and teen sex." IRB: A Review of Human Subjects Research 16 (6): 1-7, Popkin, B. M., R. E. Bilsborrow, and J. S. Akin. 1982. "Breast-feeding patterns in low incomc countries." Science 218 :1088-93. Popkin, B. M., R. E. Bilsborrow, J. S. Akin et al. 1983. "Breast-feeding determinants in low income countries." Medical Anthropology 7: 1-31. Popkin, B. M., D. K. Guilkey, J. S. Akin et al. 1993. "Nutrition, lactation, and birth spacing in Filipino women." Demography 30 :333 -52. Popkin, B. M., T. Lasky, J. Litvin et al. 1986. TheInfant Feeding Triad: Infant, Mother, and Howehold. New York: Gordon and Breach. Porter, J. E, M. J. Glass, and W. C. Koff. 1989. "Ethical considerations in AIDS vaccine testing." IRB: A Review ofHurnan Subjects Research 11 (3): 1-4. . 1993. "International AIDS vaccine trials." In Emerging Issues in Biomedical Policy: An Annual Review, edited by R. H. Blank and A. L. Bonnicksen, 2: 189-99. N& York: Columbia University Press. Public Health Service. 1971. The Institutional Guide to DHEWPolicy on Protection of Humnn Subjects. Department of Health, Education, and Welfare, Publication no. (NIH) 72102. Washington, D.C.: U.S. Government Printing Office, December. Quinn, S. C. 1993. "AIDS and the African American women: The triple burden of race, class, and gender." Health Education Quarterly 20 (3): 305-20. Repliblica de Chile. 1994. Ley 19.32~.Congreso National, Valparafso, Chile. Rindfleisch, N., and G. J. Bean. 1988. "Willingness to report abuse and neglect in residential facilities." ChiMAbuse andNeglect 12:509-20. Ringheim, K. 1995. "Ethical issues in social science research with special reference to sexual behaviour research." Social Science and Medicine 40:1691-97. Robertson, B. A., and M. A. Hayward. 1976. "Transcultural factors in child abuse." South African Medical Journal 50 :1765- 67. Rothenberg, K. 1996. "Feminism, law, and bioechics." Kennedy Institute ofEthics Journal 6 : 69-83.

Rothman, D. J. 1991.Strangers at the Bedside: A History of How Law and Bioethics TransfomedMedical Decision Making. New York: Basic Books. Runyan, D. K., P. A. Curtis, W. M. Hunter, et al. 1998. "LONGSCAN: A consortium for longitudinal studies of maltreatment and the life course of children." Aggvession and Violent Behavior 3: 245-55 Rpkiewich, M. A., and J. P. Spradley, eds. 1976. Ethics andAnthropology:Dilemmas in Fieldwork. Reprint, 1981.Malabar, Fla.: Robert E. Krieger. Samuda, G. M. 1988. "Child discipline and abuse in Hong Kong." ChildAbwe and Neglect 12:283-87.

Savitt, T. L. 1982. "The use of Blacks for medical experimentation and demonstration in the old South." Journal ofSouthern History 48 (3): 331-48. Scheper-Hughes, N. 1982. Saints, Schokrs, and Scbizophrenics:Mental Illness in Rural Irehnd. Berkeley University of California Press.

270

REFERENCES

Scofield, G. R. 1995. "Letter to FDA Dockets Management Branch commenting on proposed rule." Unpublished. November 3. Seltzer, R., and R. C. Smith. 1988. "Racial differences and intraracial differences among Blacks in attitudes towards AIDS." AIDS and Public Policy Journal 3 (3): 31-35. Sethi, S. P. r994. Multinational Corporations and the Impact ofPublic Advocacy on Corporate Strategy: NestlJund the Infant Formula Controversy. Issues in Business Ethics, Vol. 6., Boston: Kluwer Academic. Sheehan, E. A. 1993. "The student of culture and the ethnography of Irish intellectuals." In When They Read What We Write: The Politics ofEthnography, edited by C. B. Brettell, pp. 75-89. Westport, Conn.: Bergin and Garvey. Skomal, S. N. 1993. "The ethics of fieldwork." Anthropology Newsletter 34:1, 26. Star, R H., J. Dubowitz, and B. A. Bush. 1990. "The epidemiology of child maltreatment." In Children at Risk: An Evaluation ofFactors Contributing to ChildAbuse and Neglect, edited by R. T. Ammerman and M. Hersen, pp. 23 -50. New York: Plenum Press. Stein, J. 1997. Empowerment and Women; Health: Theory, Methods, and Practice. London: Zed Books. Stewart, J. F., B. M. Popkin, D. K. Guilkey ec al. 1991. "Influences on the extent of breastfeeding A prospective study in the Philippines." Demography 28 :181-99. Stoller, R 1988. "Patients' responses to their own case reports."journal ofthe American Prychoana&ic Association 36 (2): 371-91. Straus, M. A. 1979. "Measuring intrafarnily conflict and violence: The conflict tactics scales." Journal of Marriage and the Family 4:75 -88. Streiner, D. L., D. Bender, and M. Barry. 1995. "Informed Consent in Collaborative Research in Third World Countries: Whose Responsibility?" Unpublished paper, McMasters University and the University of North Carolina. Swoboda, J. S., A. Elwork, B. D. Sales, and D. Levine. 1978. "Knowledge of and compliance with privileged communication and child-abuse-reporting laws." Professional Psychology. August: 448 -57. Tacket, C. O., and R. Edelman. 1990. "Ethical issues involving volunteers in AIDS vaccine trials." Journal of1nfictiou.i Diseases 161: 356. Thomas, J. E. 1978. ~ a b e r ofLife s andDeath. Toronto: Samuel-Stevens Ltd. Thomas, S., and S. Quinn. 1991. "The Tuskegee syphilis study, 1932-1972: Implications for HIV education and AIDS prevention programs in the black community." American Journal ofpublic Health 81: 1498-505. Titmus, R. 1971. The Gzfi Relationship: From Human Blood to Social Policy. New York: Pantheon. Tong, R. 1993. Feminine and Feminist Ethics. Belmont, Calif.: Wadsworth. UNICEF. 1994. "Maltrato infantil en la familia. La voz desde 10s niiios y nifias." Unpublished paper, Oficina para Argentina Chile y Uruguay, Santiago, Chile. Valentine, D. P., D. Stewart, M. L. Freeman et al. 1984. "Defining child maltreatment: A multidisciplinary overview." Child Welfare 43 :497-509. VanDerslice, J., and J. Briscoe. 1993. "All coliforms are not created equal: A comparison of the effects of water source and in-house contamination on infantile diarrheal disease." Water Resources Research 29 :1983-95. . 1995, "Environmental interventions in developing countries: Interactions and their implications." Americun]ournal ofEpidemiology 141:135-44. VanDerslice, J., B. M. Popkin, and J. Briscoe. 1994. "Drinking-water quality, sanitation, and

REFERENCES 271

breast-feeding: Their interactive effects on infant health." Bulletin of the WorU Health Organization 72: 589- 601. Van Maanen, J. 1988. Tales of the Field: On Writing Ethnography. Chicago: University of Chicago Press. Varmus, H., and D. Satcher. 1997. "Ethical complexities of conducting research in developing countries." New England Journal ofMedicine 337: 1003-5. Wanvick, D. P. 1982. "Types of harm in social research." In Ethical Issues in Social Science Research, edited by T. L. Beauchamp, R. R. Faden, J. Wallace, Jr., et al., pp. 101-24. Baltimore: Johns Hopkins University Press. Wax, M. L., and J. Cassell. 1981. "From regulation to reflection: Ethics in social research." American Sociologist 16 :224 -29. Wolf, S. M., ed. 1996. Feminism and Bioethics: Beyond Reproduction. Oxford: Oxford University Press. Wyatt, G. E. 1991. "Examining ethnicity versus race in AIDS related sex research." SocialScience andMedicine 33 (I): 7 - 4 5 . Zohoori, N., B. M. Popkin, and M. E. Fernandez. 1993. "Breast-feeding patterns in the Philippines: A prospective analysis."Journal of Biosocial Science 25 :127-38.

Contributors

Marcela Aracena Alvarez, M.Sc., is a licensed psychologist and a faculty member in the School of Psychology of the Pontificia Universidad Cat6lica de Chile in Santiago, Chile. She is the author of several articles on child abuse in Chile. Jorge BalAn, Ph.D., is a sociologist and a Program Officer at the Ford Foundation. Formerly, he was Senior Researcher and Director of Centro de Estudios de Estado y Sociedad (Center for the Study of State and Society), a private, nonprofit research center in Argentina. With Silvina Ramos, he has coordinated research on reproductive health and sexuality.

B. Susan Bauer, B.A., is currently teaching English at a Lutheran seminary in Namibia. She was Director of the Office of Human Research Studies at the University of North Carolina School of Medicine in Chapel Hill. She has led workshops on writing clear consent forms and is very interested in language and comprehension issues. Alan F. Benjamin, Ph.D., is a visiting lecturer in the Department of Anthropology University of North Carolina at Chapel Hill. Lynn Blanchard, M.P.H., Ph.D., is a health educator with the Department ofcommunity Health and Health Studies of the Lehigh Community Hospital, Lehigh, Pennsylvania. Allan M. Brandt, Ph.D., is Kass Professor of the History of Medicine at Harvard University, where he teaches in the Department of Social Medicine and the Department of the History of Science, and is Director of the Ethics Program at the Harvard School of Medicine. Among his publications is N o Magic Bullet: A Social Histoly of Venereal Disease Since 1880.

J. Pat Browder, M.D., Ph.D., is a physician and Regional Ethics Coordinator for Kaiser Permanente in Durham, North Carolina. He has written about a variety of ethical issues, including the ethics of managed care. Barbara Entwisle, Ph.D., is Professor of Sociology and Fellow, Carolina Population Center, University of North Carolina at Chapel Hill. She is the author of numerous articles on social and demographic change in China, Thailand, and Russia. Sue E. Estroff, Ph.D., is Professor of Social Medicine and Adjunct Professor of Anthropology and Psychiatry, University of North Carolina at Chapel Hill. She is the author of

CONTRIBUTORS

273

Making It Crazy:An Ethnography ofPsychiatric Clients in an American Community, as well as articles on the experiences of persons with severe, persistent psychiatric illness. RenCe C . Fox, Ph.D., is Annenberg Professor of the Social Sciences at the University of Pennsylvania. She is the author of a number of books on the social and ethical implications of new medical technology including (with Judith Swazey) Experiment Perilous, The Courage to Fail, and Spare Parts, as well as articles on a wide variety of topics in medical sociology. Lara Freidenfelds is a doctoral candidate in the Department of the History of Science at Harvard University. Gail E. Henderson, Ph.D., is Associate Professor of Social Medicine and Adjunct Associate Professor of Sociology, University of North Carolina at Chapel Hill. She is the author of The ChineseHospital:A Socialist Work Unit, and articles on health and health care in China. Nancy M. P. King, J.D., is Associate Professor of Social Medicine, University of North Carolina at Chapel Hill. She is the author of Making Sense ofAdvance Directives and articles on decisionmaking in health care and medical research. Loretta M. Kopelman, Ph.D., is a philosopher and Professor and Chair ofthe Department of Medical Humanities, East Carolina University. She is editor of the research section of the Encyclopedia ofBioethics (revised edition), and articles on literature and medicine, classical philosophy, and research ethics. Ernest N. Kraybill, M.D., is a neonatologist and Professor of Pediatrics, University of North Carolina at Chapel Hill. He is Chair of the School of Medicine Committee for the Protection of the Rights of Human Subjects. His research focuses on developmental outcomes of premature and low birth weight babies. Ruth Macklin, Ph.D., is Professor of Epidemiology and Social Medicine at the Einstein College of Medicine, Yeshiva University. She is the author of a number of books, including Enemies ofPatients, Mortal Choices, and Surrogates and Other Mothers, as well as articles on philosophy and medicine. She was a member of the President's Advisory Committee on Human Radiation Experiments. Barry M. Popkin, Ph.D., is Professor of Nutrition and Fellow, Carolina Population Center, University of North Carolina at Chapel Hill. H e has written extensively on nutrition and economic change in the Philippines, China, Russia, and the United States. Silvina Ramos, Ph.D., is a sociologist and Co-Director of the research program on women and AIDS at Centro de Estudios de Estado y Sociedad (Center for the Study of State and Society), a private, nonprofit research center in Argentina. With Jorge Ballin, she has coordinated research on reproductive health and sexuality. Desmond K. Runyan, M.D., Dr.P.H., is Professor of Social Medicine and Pediatrics, and Clinical Associate Professor of Epidemiology, University of North Carolina at Chapel Hill. H e has written extensively on child maltreatment in the United States.

274

CONTRIBUTORS

Jane Stein, Dr.P.H., is a lecturer in the University of North Carolina School of Public Health's Department of Maternal and Child Health and a consultant with Meliora Associates. She is the author of Empowerment and Women?Health: Theory, Methods, and Practice. Her research interests include women's health, the social determinants of health, and participatory research and evaluation. Ronald P. Strauss, D.M.D., Ph.D., is Professor and Chair of the Department of Dental Ecology and Professor of Social Medicine, University of North Carolina at Chapel Hill. H e has written about social and ethical issues in the care of chronic illness, particularly of children, and his research interests include HIVIAIDS and community involvement in research. Keith A. Wailoo, Ph.D., is Associate Professor of History and Social Medicine, University of North Carolina at Chapel Hill. H e is the author of Drawing Blood: Technologyand Dis-

ease Identity in Twentieth-Century America. Cynthia Waszak, Ph.D., is Senior Research Associate, Family Health International. She has conducted research on reproductive health in a number of countries in Asia and Africa.

Index

Abortion research, 32-35, 39, 137-57 Absolutism, moral, 25, 32, 217 Academic freedom, 54,60, 69,78, 79,215; and free speech, 64 (n. 4), 219 Adolescents: as research subjects, 39-40 Advisory Committee on Human Radiation Experiments, 5, 42 Advocacy: as role of researchers, 138, 140, 151-52, 216; values framework, 119, 124-25 African Americans: as research subjects, 6, 83-107, 210 American AnthropologicalAssociation, 75 American Medical Association Code of Medical Ethics, 199 American Psychological Association, 174 Anonymity in research, 181 Anthropological research. See Ethnography Assent, 9,168 Autonomy, 8, 25, 31, 43,178,19~-96,199, 205; of children, 162,168,184; and culture, 21,176 AZT research, 223-24 Battered child syndrome, 163 Belmont Report, 8,9,10,12, 17 (n. 41, zy-26, 37,195, 197, 208 Beneficence, 9,1o, 195. 206 Benefits: to society, zoo, 205, 206; to subjects, 150-52, 206, 210 Bernard, Claude, 5,17 (n. 2) Bias, 61, 62, 111, 118-19, 155 Bill of Rights, 64 (n. 4) Boas, Franz, 72 Breast-feeding, infant formula, and health, I Z O - ~ I , I Z125-26 ~,

Business: as research sponsor, 113-22, 128-34; values framework, 125-26 Callahan, Daniel, 200 Canterbury u. Spence, 185 (n. 4) Casuistry, 220 Cebu Longitudinal Health and Nutrition Survey, 115-16, 120-21 Censorship, 52, 54,69 Center for the Study of State and Society, 140 Child abuse, 162-64,169-70; defined, 172-79; interventions, 165,169,182-84; reporting requirements for, 163,167,168, 170,171,174,180-84; research, 161-62, 164-69 Children: as research subjects, 7,9,161-70, 216 Child Sexual Behavior Inventory, 167 Clifford, James, 104 Clinical trials. See Medical research Code of Federal Regulations, 42 Coercion, 146, 149 Common Rule, 4,10,1z, 23-26,191, 204-5, 208, 227-38 Community, 2,48,60,63, 83-107, 220-21, 224; defined, 3,198, 201; and informed consent, 12-13,190,192,196,197, 210-12 Community advisory board, 84, 88-106 passim, 215, 216 Confidentiality, 3, 8, 31, 32, 33, 34, 35, 40. 98-99,145-46,148,161; certificate, 99, 168; versus reporting requirements, 173-75.180 Conflict of interest, 16,119, 123-24,127,129, 133. 1557 215

276

INDEX

Conflict Tactics Scale, 166,167 Congregation MikvC Israel-Emanuel, 51-56 Conley, John, 64 (n. 4) Consent. See Informed consent Consent form, 40,41,43,145,155, 171-72. See also Informed consent Context: of ethics, I, 13, 62-63, 138, 176, 222; of research, 61~62,156-57, 207-8 Contract: pre-publication approval requirement in, 50, 52-53, 56, 65 (n. 12). 68, 75; brrwren researcher and subjects, 2, 47-80,252-54 Council for International Organizations of Medical Sciences, 12-13, 26 Covenant, 50, 55, 57, 63, 70 Cross-cultural research, I, 10-13, 28,32-33, 197, 211-12, 222, 223-24 Deception, 6, 7, 36, 41-42.142-46,155-56, 180, 181 Decisional capacity: of children, 167, 177; subjects without, 6-7, 9,192,196, 197 Declaration of Helsinki, 26,42, 197, 7-05 Department of Health and Human Services, U.S. (DHHS), 191-94, 201 Design of research: ethical implications of, 34,156, 216; role of community in, 93, 149, 151 Disclosure: of research funding, "9, 124,127, 129,133, 216; as research risk, 147,148,151, 154,164,167-69.180, r8r Discrimination: as research risk, 85,106 Drug Amendments of 1 9 6 2 , ~ Durham, N.C., 89-90 Elites: as research subjects, 48, 53, 54, 64-65 (n- 6) Emergency medicine research, 3, 189-212 Emergency treatment, 199 Emmanuel, Isaac, 52-53, 78 Enlightenment, 8,13 Equipoise, 196-97 Ethics: cross-cultural, 12,14, 21, 24,26-28, 177-78, 217; descriptive versus normative, 22, 25; history of, 5-8; regulations and guidelines, 10-13,37-43; universal, 21-37, 42-43, 139, 178, 213, 222. See aho Principles; Relationships

Ethnography, 13-14,47-~0, 55-56, 67,69, 70,255-57; ethics in, 5-63,73-75; roles in, 76-79 Exploratory research, 137, 138,141-41,156, 172-73 Faden, Ruth, 26, 211 Families: as research subjects, 163-85 Farrer, Claire, 58 Fear of prosecution: as harm to subjects, 145. 146,151; as hedth risk, 143-44, 155 Feminist ethics, 32, 220, 221 Flauhert, Gustave, 17 (n. 2) Food and Drug Administration, U.S., 192, 193, 201, 210 Food, Drug, and Cosmetic Act, 5 Funding of research, 3, 60, 97, 111-34, 206, 209, 215; and review requirements, 145, I47 Geertz, Clifford, 59, 73-74 Genetic fallacy, 25 Gifts versus grants in research, I I ~ , I I119, ~, 122,126, 129,130 Goals: of medical treatment, 208, 209; of research, 99-100, 140-42,208,210, 219-21 Gordon, Malcolm, 168 Government: as research sponsor, 4,60, 116-17, 219 Griaule, Marcel, 104 Harms, 6, 7, 11,76, 86,147, 200, 205; dignitary, 37; to groups, 74; minimizing, 14-50; psychological, 7,149,154; versus wrongs, 26, 35, 206 HIV, 2,207; research, 3,83-107,181, 223-24; vaccine, 85,87, 99-100 Holaday, Duncan, 42 Holocaust, 48, 54 Humphreys, Laud, 7 Ijsselmuiden, Carel, 26, 211 Illegal behavior: research on, 137-57.161-85 Infant feeding research, 113-22 Infant mortality, 128,129 Informed consent, 21, 25, 34, 35, 36,143, 148, 155,167,176-77,178,181,199, 200, 205,

INDEX

216, 217; and culture, 26, 27, 211; deferred, 192; and illiteracy, 30; process versus form, 29-30,33, 145; in research regulations, 23-24; research without, 7-8,189, 200, 204-6; and understanding, 38; and women, 30-31 Institutional review board, 10, 24, 43, 76, 98, 169,192,198, 201, 202-3, 220 Integrity: scholarly, yo, 59-61; scientific, 123, 124, 127,129, 133-34 International Ethical Guidelines for Biomedical Research Involving Human Subjects, 26, 247-P International Guidelines for Ethical Review of Epidemiological Studies, 239-46

Jehovah's Witnesses: and refusal of medical treatment, 209 Jenner, Edward, 17 (n. 2) Jewish Chronic Disease Hospital case, 6-7 Jones, James, 105-6 Justice, 9, 195

Longitudinal research, 161-62,164-65 LONGSCAN, 161,164-70,178,184, 216 Luce, john M., 195,196 Maclntyre, Alasdair, 63 Madame B o u q , 17 (n. 2) Marcus, Jacob R., yz Mead, Margaret, 72 Media: effect on research, 87 Medical personnel: as research subjects, 143, 145,153 Medical research, 67,189-212; versus medical treatment, 155,169, 182; versus social science research, 11,37, 206 Mentoring, 72-73, 78-79 Methods of research. See Ethnography; Exploratory research; Longitudinal research; Participatory research; Psychology research; Social science research; Survey research Milgram, Stanley, 6 Minorities: as research subjects, 195, ZIO

277

Moral progress, 32 Morton, Samuel George, 62, 66 (n. 14) Narrative ethics, n o , 222 National Acute Brain Injury Study, 191-93 National Center o n Child Abuse and Neglect, 168 National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 8,17 (n. 4) National Human Investigation Board, ZOO National lnstitute of Mental Health, 168 National Institutes of Health (NIH), 2, 36, 85, 112,115-17, 119, 129. 191 National Study of Child and Adolescent Well-Being, 170 (n. I) Native Americans: as research subjects, 58 N e ~ t k111, , 113. 114, 115, 124-27, 128-32 Newton, Lisa, 26 Nonmaleficence, 9 Nuremberg Code, 6, 26, 42, 176, 197, 205, 225-26 Nuremberg trials, 6, 200 Objectivity in research, 59, 61,124 Office for Protection from Research Risks, 36,192 Participatory research, 9,14, 83, 219-20, 221 Pasteur, Louis, 17 (n. 2) Paternalism, 1~6,184-85,206, 217 Patients: as research subjects, 6-7, 11, 137-57. 189-212 Payment to subjects, 36,98 Peer review. See Review of research Phillip Morris, Inc., 133 Placebos, 37,195, 223 201-2, 221-22; effect of fundPolicy, ing on, 130-31,132-33; effect of research on, 150,151,182-84,216 Population Council, 145,1y7 Poverty of research subjects, 195,210 Power, 9, 13, 38, 48, 5y, 56, 61, 69, 83, 97.149, 176, 216, 218-20 President's Commission for the Study or Ethical Problems in Medicine and Biomedical and Behavioral Research, 17 (n. 5)

278

INDEX

Principles, 4-5, 8-1o,13-16, 21-29, 42-43, 195; adjudicatory role of, 33-34; as ethics paradigm, 15,137, 213-14; versus relationships paradigm, 217-23 Privacy, 8,25,175,176 Professional dominance, 47, 69 Project LinCS, 85,88,93 (n. I) Psychology research, 6, 7-8 Public Health Service, U.S., 6, 41-42, 84, 85,92,102, 105-6,223 Publishing, scholarly: effect on research subjects, 5% 59.74 Pure Food and Drugs Act, 5 Radiation research, 5, zoo Refusal of treatment, 209 Refusal to participate: as source of research bias, 166,16g,r70,180-81 Regulations, 21, 23-24; early federal, 8,1o; international, 12-13, 26, 223. See also Common Rule Relationships, 11, 63,101, 102, ray, 162; as ethics paradigm, 2,14-16,72-79, 137, 207-12, 213-17, 224; versus principiist paradigm, 4-5.21, 22, 217-23 Reproductive health research, 28-36,137-57 Reputation: harm to, 50, 52, 53-54, 55.70. 75 Rcscarchers: as service providers, 206, 207 Respect for persons, 8-9,1o, 25-26,jz, 41. 170,178,195,196,199, zoo Review of research: external, 138,145,150. 155,157. zoo, 203; peer, 116-17, 130-31; pre-publication, 50, 52-53, 56, 58, 65 (n. I$, 68, 75, 145, 147. See also Institutional review board Risk-benefit assessment, j7,40.145,154-$5, I75 Risks. See Harms St. Martin, Alexis, 17 (n. 2) Scheper-Hughes, Nancy, 59 Science: values framework, 124 Scientific merit, 138,151,156, 218 Screening: consent for, 41 Self-determination, 26, 43 Self-incrimination: as research risk. See Disclosure

Sensitive topics, 39-.+0,97, 99,149,169 Sephardi Jews, 51, 64 (n. 5), 255-56 Sexual abuse, 167 Slater v. Baker and S t a p h n , 17 (n. 2) Slippery slope, 197 Social medicine: defined, ix Social risk, 86, 207 Social science research, 10, 32-34> 39-40; benefits of, 35-j6,38, 206;compared to medical research, 11,17 (n. 6), 34; harms of, 3439; lack of regulation for, 38 Stigma, 84-85, 91, 106, 149 Strauss, Ronald I?,65 (n. 9) Students: as researchers, 72-73> 78-79 Subjects. See African Americans; Children; Decisional capacity: subjects without; Elites; Families; Medical personnel; Native Americans; Patients; Vulnerable populations Support services: expected in research, 39-40 Survey research, 116, 181 Ten Commandments, 25 Thalidomide, 6,17 (n. 3) Tobacco industry: effects of funding by, 132-33 Tobacco Industry Research Council, 130, 132 Tokyo Amendment, 26 Tti-Council Code of Ethical Conduct for Research Involving Humans, 17 (n. 6) Trust, 48, 61,86-87, 94,133,146,161-62, 203-10, 218, 220 Tuskegee syphilis study, 6, 84, 85, 86,92, 95, 97, IOZ,IOj, 105-6,183, ZOO, 223 UNICEF, 117,124,125,127,130,178 Use of research findings, 100-IOI,I++, 150-52, 182 Utilitarianism, 31 Values: conflicts in research, 200-201; of research sponsors, 124-27 Violence: in families, 163,171, 174; perceptions about, 177; in schizophrenia, 74 Virtue ethics, 218, 220

INDEX

Vulnerable populations: as research subjects, 6-7, 9,11, 94, 99,101,145,148-49,155, 196,199, 210 Waiver of consent, 191, 193,195-98, 209, 210-12 Washington, Booker T., 105 Wellstart, 117-18, 124, 125,1z7,1zg

205.

279

Whitehead, Alfred North, ix Wichita jury study, 7-8 Willowbrook study, 6-7 World Health Organization (WHO), 26, 31, 36,145 Zenner, Walter, 64 (n. 5) Zimbardo prison experiment, 6