Research Misconduct Policy in Biomedicine: Beyond the Bad-Apple Approach

© 2013 Massachusetts Institute of Technology

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher.

Library of Congress Cataloging-in-Publication Data

Redman, Barbara Klug.

Research misconduct policy in biomedicine : beyond the bad-apple approach / Barbara K. Redman.

   p.  cm.

Includes bibliographical references and index.

ISBN 978-0-262-01981-1 (hardcover : alk. paper)

ISBN 978-0-262-31776-4 (retail e-book)

1. Medicine— Research—Corrupt practices. 2. Medical scientists—Professional ethics. 3. Medical research personnel—Professional ethics. 4. Biotechnology—Research. I. Title.

R852.R435 2013

610.72′4—dc23

2013009438

10 9 8 7 6 5 4 3 2 1

d_r1

Contents

Series Foreword

Introduction

1    Has Something Important Happened to Scientific Ethics?

2    Will Approaches from the Psychosocial and Organizational Sciences Improve Research Misconduct Regulation?

3    What Combination of Regulatory Tools Will Minimize Research Misconduct?

4    Are Current Publication Standards Protecting the Integrity of the Scientific Record?

5    Lessons from Clinical Case Studies in Research Misconduct

6    Research Misconduct Policy, Its Evolution and Culture of Morality

7    The Unity of Moral Science

References

Index

Series Foreword

I am pleased to present the thirty-eighth book in the Basic Bioethics series. The series makes innovative works in bioethics available to a broad audience and introduces seminal scholarly manuscripts, state-of-the-art reference works, and textbooks. Topics engaged include the philosophy of medicine, advancing genetics and biotechnology, end-of-life care, health and social policy, and the empirical study of biomedical life. Interdisciplinary work is encouraged.

Arthur Caplan

Basic Bioethics Series Editorial Board

Joseph J. Fins

Rosamond Rhodes

Nadia N. Sawicki

Jan Helge Solbakk

Introduction

Overview

Public policy on research misconduct, which has developed contentiously in the United States and a few other countries over the past thirty years, remains largely untested as to whether it yields clearly specific outcomes; alternative policies that might reach those outcomes remain unexamined. Each widely publicized case of research misconduct creates a new scandal, leading to questions about whether current regulation is effective or just, and whether it supports the progress of science.

It is important to remember the broader context of research misconduct, the system of incentives and disincentives embedded in the production and dissemination of research. Yet, in current public policy, all attention is on those individuals who have been caught fabricating, falsifying, or plagiarizing. Research Misconduct Policy in Biomedicine reviews a variety of frameworks, each of which provides options for preventing and controlling this phenomenon. The management of research misconduct policy and other reforms, such as the control of conflict of interest, can improve the quality of scientific research; they can help it to build trust in its conduct and the knowledge it produces and thus to better fulfill its compact with the public, which provides much of its funding.

We will start with the current regulatory framework of the Office of Research Integrity (ORI), whose purview is largely confined to the life sciences and to research funded by the Public Health Service (PHS). This framework has developed from an informal system of peer review to incorporate legal standards and processes. Similar systems are being formed across the globe, but as this subsystem of research ethics evolves (other subsystems include human and animal subject protection, and containment of conflicts of interest), a full range of policy options should be considered.

Current research misconduct policy is built on the belief that individuals are able to control both the risk factors they encounter and their own destinies. The premise of this book is that a fair and effective policy must reflect both the context of research misconduct and the various conditions that affect it. In my view, the goals of a system regulating research misconduct should be (1) to protect scientific capital (knowledge, scientists, institutions, resources, norms of science); (2) to support fair competition in scientific research; and (3) to contain harms to end users (Redman 2009, 216). These specific goals make explicit what is implied in the broader goals set forth in the Code of Federal Regulations (42 CFR 93.101 (e), as in effect October 1, 2011): To protect the health and safety of the public, promote the integrity of PHS supported research and the research process, and conserve public funds.

Current Regulation on Research Misconduct

According to the Code of Federal Regulations (42 CFR 93.102–104, as in effect October 1, 2011):

(a) Each institution that applies for or receives PHS support for biomedical or behavioral research, research training or activities related to that research or research training must comply with this part.

(b)(1) This part applies to allegations of research misconduct and research misconduct involving:

(i) Applications or proposals for PHS support for biomedical or behavioral extramural or intramural research, research training or activities related to that research or research training, such as the operation of tissue and data banks and the dissemination of research information;

(v) Plagiarism of research records produced in the course of PHS supported research, research training or activities related to that research or research training.

Research misconduct means fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.

(a) Fabrication is making up data or results and recording or reporting them.

(b) Falsification is manipulating research materials, equipment, or processes, or changing or omitting data or results such that the research is not accurately represented in the research record.

(c) Plagiarism is the appropriation of another person's ideas, processes, results, or words without giving appropriate credit.

(d) Research misconduct does not include honest error or differences of opinion.

A finding of research misconduct made under this part requires that—

(a) There be a significant departure from accepted practices of the relevant research community; and

(b) The misconduct be committed intentionally, knowingly, or recklessly; and

(c) The allegation be proven by a preponderance of the evidence.

As can be seen, the text of this regulation admits of no ambiguity in the standards of scientific research—it assumes that there is clear agreement as to what constitutes fabrication or falsification, accuracy of the research record, and accepted practices in the relevant research communities. Yet where is the evidence for this assumption? More important, the text addresses neither the context in which misconduct may occur nor the seriousness of its consequences, even though sanctions for the misconduct do vary, presumably based on some underlying metric of seriousness.

Current regulations and the policies that gave rise to them acknowledge certain aspects of the complex behavior called research misconduct and place other aspects in the background, where they remain as unanswered questions. As the dual system of federal regulation and self-regulation by the scientific community evolves, however, these questions about ethics and quality control in scientific research come to the fore, demanding our attention in anticipation of the next phase of policy evolution.

Unanswered Questions

Current research misconduct policy leaves pressing questions to ponder:

• How could the framework of scientific ethics, which seemed to be so firmly established, allow repeated instances of serious and often protracted research misconduct to occur? Even though there is a history of widely accepted documents addressing the protection of human research subjects, such as the Nuremberg Code (1947), Declaration of Helsinki (1964; later revised several times to reflect the evolution of scientific ethical standards), and the Belmont Report (1978), there are no guiding documents of this stature that address fabrication, falsification, and plagiarism (FFP) in scientific research. Rather, it is assumed that all scientists and their support staffs are taught scientific ethics, which clearly defines and precludes FFP. Wouldn’t a guiding ethical document that formally addressed fabrication, falsification, and plagiarism be useful?

• How might scientists, employers, funders, journal editors, and the scientific community itself each improve oversight of research misconduct?

• Why do scientists and their staff fabricate, falsify, and plagiarize? Are there predictive risk factors in individual researchers or the research environment that could alert us to the need for preventive interventions? Whose responsibility is it to determine those factors?

• Why is FFP wrong? These behaviors violate basic ethical tenets not to lie (F/F) or steal (plagiarize). But they also create harm to other scientists, to end users such as patients or other research subjects, and to the public trust. Why is there so little documentation about the harms produced to all stakeholders (patients, subjects, other researchers, providers)?

• Are current regulations and the structure for their implementation effective? Each employing institution is required to receive approved assurance for policies and procedures employed to deal with allegations of research misconduct. Unlike human and animal subjects protection, the regulations don’t require other forms of oversight such as regular audits.

• Are the current sanctions an effective deterrent against research misconduct? Do they help to protect scientific capital, support fair competition in scientific research, and contain harms to end users (Redman 2009, 216)? And do they help scientific research meet its societal obligations? Common sanctions against those found to have committed research misconduct include exclusion from receiving federal research funds or from serving on federal advisory panels for a period of three years and retraction of publications corrupted by fabricated or falsified data or correction of them.. Supervision of future research or ethics education may also be imposed.

• Why is the present system for discovering research misconduct so dependent on whistleblowers, who must put themselves at great professional and personal risk?

• Would a systematic, cumulative analysis of misconduct cases yield important insights into the present system for preventing and managing research misconduct? Although some research (based on the ORI’s confidential files) is currently providing this kind of information, much more is needed. One such study of trainee research misconduct found mentors not reviewing source data or setting standards (Wright, Titus, and Cornelison 2008).

• How could a more logical, coherent, and comprehensive system of oversight be constructed? Oversight of research misconduct in any given instance is commonly the responsibility of a particular funding entity in a particular country, even though the research may be funded from a variety of sources and may involve scientists and staff from around the globe.

Voices Framing the Issue

Since the early 1980s, independent voices have raised concerns about the ability of the scientific community to control research misconduct. Writing in the early 1980s when news about research misconduct was emerging in public, William Broad and Nicholas Wade, two journalists who had covered many of the scandals, noted that the ideology of science (error of all sorts will be cast out quickly) could not account for FFP. Instead, the discovery of repeated fraud offered a glimpse at the pathology embedded in a scientific knowledge system regarded as the ultimate arbiter of truth: The roots of fraud lie in the barrel, not in the bad apples that occasionally roll into public view. The elite don’t get detected—they have firm control over the social institutions of science to perpetuate their own ideas and promote their own supporters and students (Broad and Wade 1983, 256). The two journalists concluded that fraud is most probably a small, albeit significant and endemic, feature of the scientific enterprise and noted that government agencies, not the institutions of science, had taken the initiative in trying to control it.

In the 1993 Shattuck Lecture, Congressman John Dingell (D-MI), chair of the U.S. House Subcommittee on Oversight and Investigations, described research misconduct as the same as any other fraud against the American taxpayer (Dingell 1993). The cases his subcommittee examined went on for years without attention, whistleblowers were treated badly and often worse than the offender, internal procedures in place failed, and acts of research misconduct were trivialized as minor mistakes … (Dingell 1993, 1613). Congressional hearings were a forum of last resort, to shine light on severe problems not being addressed.

Science author John Ziman (2000, 31) notes that, unlike the norms of other professions, scientific norms are usually presented as traditions and are not codified or enforced by specific sanctions within a formal system of governance. This has serious consequences, Ziman explains. By assuming general virtue, the norms of science open opportunities for the unscrupulous (p. 31). Indeed, because academic science is not systematically policed against fraud, it is relatively easy to get fraudulent research claims into the literature, and when found out, miscreants are condemned as social deviants but with relatively easy sanctions. No regular mechanisms are in place to distinguish between obsessive self-delusion and systematic deceit (p. 269). Contrary to the popular perception that the science knowledge system quickly detects and eliminates errors, Ziman found the formal research record was full of them. It is a normal social institution inhabited by normal, morally frail people… these revelations do not prove that science is morally corrupt. On the contrary, they show that its formal communication system is very far from an empty ritual (Ziman 2000, 160).

These and other such voices have, over the last three decades, come to the same conclusions—the structures and norms of the scientific community don’t reflect the reality of doing science; the community has repeatedly failed to regulate itself and, indeed, hasn’t even grasped the problem of research misconduct.

Plan for the Book

Potential answers to the questions and concerns raised above are available from several sources. The literature in related fields—moral psychology, the policy and organizational sciences, and the law—offers suggestions for oversight policy, as does analysis of actual cases of misconduct. Empirical studies, most funded by the Office of Research Integrity and the National Institutes of Health, are few but instructive. In addition, stakeholders from government, scientific associations, journalism, and the law have played a role in the move toward regulation of research misconduct and offer valuable perspectives on the effectiveness of current efforts and alternative approaches.

Research Misconduct Policy in Biomedicine systematically examines each of these sources and summarizes policy options. Its predominant focus is on clinical research involving human subjects, with far less attention paid to research with animal subjects and to basic science. The clinical emphasis reflects both my own background and the fact that many cases that have risen to the level of scandal involve clinical research with humans. Research Misconduct Policy analyzes unexamined assumptions embedded in current research misconduct policy. It presents options for the necessary next phase in the evolution of that policy. These policy options must, of course, remain hypotheses until they can be tested against a set of goals that don’t currently exist.

Chapter 1 considers the normative beliefs that have served science well for so long. Although seriously challenged by Broad and Wade (1983), Robert K. Merton’s norms for the ethos of science (1942)—universalism, communism, disinterestedness, and organized skepticism—are still widely subscribed to in the scientific community. More recent thinking on the theory of the commons suggests a framework for successful self-regulation by scientific research communities to manage and protect their common-pool resources, which I interpret as the quality and integrity of scientific data and as trust both between fellow scientists and between them and the public. Such self-regulation requires users of the common resources to agree on rules, to monitor compliance, to report violations, and to enforce sanctions. Moreover, it also depends on shared norms and on political institutions that generate and sustain trust.

Chapter 2 examines several bodies of research on matters that current misconduct regulation still fails to acknowledge—in particular, the effect of organizational contexts on the behavior of individual scientists and staff. Significant bodies of research in the psychosocial and organizational sciences tell us that bad apple behavior can be precipitated by social pressure to achieve goals and