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

5 Lessons from Clinical Case Studies in Research Misconduct

5 Lessons from Clinical Case Studies in Research Misconduct Although countries are building infrastructures to deter and manage research misconduct and to anticipate its consequences, there is still a general sense that such infrastructures may not be sufficient. Each reported scandal appears as a shock that frequently precipitates a reactive move. Can preventive measures such as education in the ethically responsible conduct of research be more proactive, describing conditions that lead to research misconduct (Dion 1998) and acting on risk factors for it? Because there are few empirical studies, most of the information on research misconduct is to be found in case reports published in general science journals such as Nature and Science and in general medical journals, although these reports, of course, cover only a subset of such misconduct. In the United States, publicly available information likely overreports those cases falling under the authority of the Office of Research Integrity (with Public Health Service funding), but it also covers international cases, frequently those in countries that are just establishing an infrastructure to deal with misconduct. Although it is unclear how representative these reports are in terms of either the extent or the nature of the misconduct, they can often provide a frame of reference for how the scientific community views these cases. 80 Chapter 5 Since records of employers of those alleged to have committed research misconduct and of government regulatory agencies aren’t public, we can expect several biases in the published case reports. They likely underreport allegations that were eventually not substantiated and routine fabrication, falsification, or plagiarism, especially by research staff, and they overreport egregious research misconduct. Because the United States has the largest research portfolio and its research misconduct infrastructure is developed, the reports may also overreport misconduct on the part of U.S. researchers. And, finally, reflecting the current regulatory framework, case studies seldom contain information about organizational or psychosocial causative factors. Although instructive, tracking cases over time is of little help in answering the following questions: 1. What workplace or personal issues are involved in cases where research misconduct is found? 2. What consequences does the misconduct finding have on the offender? 3. What effects does the finding have on others? 4. How important is a country’s research misconduct infrastructure in dealing with cases of misconduct? To each of these questions, analysis of a case or cases can suggest tentative answers. Question 1 involves (a) organizational factors: pressure to obtain funding and degree of oversight including inspection of data and replication before publication; (b) personal factors: evidence of moral disengagement or self-deception or confusion between clinical ethics and research ethics, leading to infractions other than the principal misconduct found, including those involving human subjects; (c) system factors: availability of grants and moral economy in the discipline; and (d) Lessons from Clinical Case Studies in Research Misconduct 81 time factors: duration of the research misconduct before it was discovered, reported, and dealt with. Question 2 includes consequences for the offender’s (a) research productivity in the short (when sanctions are in effect) and long term (does the offender suspend research or cease research altogether?); and (b) professional career (does the offender change careers or continue in a position requiring research?). Note that, although reoffending is relevant, there is almost no reliable information on it. Question 3 includes effects (a) on the offender’s colleagues and other scientists studying the same issue (does the finding slow them down, or does it have no especial effect?); (b) on end users (how do human subjects and clinicians in trials using the falsified or fabricated results respond to the finding, and how do doctors and patients affected by those results?); and (c) on the scientific community (Does it call for greater disciplinary self-regulation, or does it blame current research misconduct regulations?). Question 4 involves the country’s (a) science aspirations; (b) history of corruption or cronyism in science; and (c) development of appropriate regulatory infrastructure. Answers to these specific aspects of the four questions could help policy makers or administrators identify and reduce the risk factors for research misconduct. Clinical Cases to be Analyzed I have analyzed four prominent cases of misconduct in clinical research that continued undetected for a protracted period of time; I have chosen clinical cases because there is greater potential for fabrication or falsification of data to harm subjects and eventually patients in clinical than in basic science research...