Hastings Center Report

In early June, Brian C. Martinson, Melissa S. Anderson, and Raymond de Vries published an illuminating survey in Nature of 3,247 investigators who receive funding from the National Institutes of Health. One third of the respondents reported having engaged recently in one or more of what focus groups of investigators and compliance officers considered "serious misbehaviors." Most common were "inadequate record keeping," "changing the design or methodology of a study in response to pressure from a funding source," and "dropping observations or data points from analyses based on a gut feeling that they were inaccurate."

The findings lend support to longstanding calls to establish comprehensive educational programs encouraging the responsible conduct of research—RCR, as it is often known. They also suggest, however, that the educational programs need to be even more comprehensive than most of their proponents recognize.

Congressional inquiry into scientific misconduct in the late 1980s prompted new regulation and professional statements on integrity. Since 1991, when the NIH's National Research Service Award (NSRA) institutional training grants first required recipient institutions to provide trainees formal instruction in RCR, the Institute of Medicine, National Academy of Science, American Association of Medical Colleges, American Society for Microbiology, and Department of Health and Human Services Commission on Research Integrity have recommended educational initiatives to improve researchers' understanding of standards and commitment to integrity in research.

In 2000, the DHHS Office of Research Integrity (ORI) issued a policy that required institutions to provide RCR education to everyone involved in proposing, performing, reviewing, or reporting federally funded research. The policy met opposition in Congress, however, and university administrators and science organizations like the Federation of American Societies for Experimental Biology argued that its requirements were too broad, its costs prohibitive, or the needed curricular resources unavailable. ORI withdrew the universal education policy in February 2001, and for now the only federal policy on research integrity education in the biomedical sciences is the NSRA training grant mandate.

Martinson and colleagues' view of the wider research environment coincidentally provides evidence that universities cannot rely only on the traditional apprenticeship system and role modeling to transmit standards of scientific integrity to trainees. Our own research (ongoing and as yet unpublished) suggests that new trainees may readily interpret as acceptable practice what Martinson et al. call "mundane 'regular' misbehaviours." Preliminary data from entering graduate students at three universities show that, before receiving any RCR instruction, new graduate students with previous research experience scored statistically significantly lower on a test of basic RCR concepts than did new students with no such experience. In particular, new students with prior research experience appeared less likely to identify standards of data management and the demands of required institutional oversight.

The Martinson study also supports some educators' concern that the NSRA mandate has a limited effect on students' actual practice. Formal RCR education can improve trainees' knowledge of standards, but establishing its effects on their behavior remains the holy grail. In the Martinson study, the limited effect of required RCR education is reflected in the data on early-career investigators, 58 percent of whom were postdoctoral fellows with grants that would have required an RCR education component.

Given findings like these, trainees cannot be the only focus of research integrity education. To have an effect on the broader research environment, education to promote research integrity must reach researchers at all levels.

These comments represent the authors' opinions and do not necessarily reflect the opinion of the Uniformed Services University of the Health Sciences or the Department of Defense. [End Page c2]