Research Misconduct

A significant number of cases of research misconduct in academic labs go unreported each year, according to a recent survey (Nature 2008, 453, 980). Conducted by Sandra L. Titus and Lawrence J. Rhoades at the Department of Health & Human Services' Office of Research Integrity (ORI) and James A. Wells, director of the Office of Research Policy at the University of Wisconsin, Madison, the survey evaluated responses from 2,212 researchers and found 201 instances of research misconduct over a three-year period. Research misconduct was defined as "fabrication, falsification, or plagiarism in proposing, performing, or reviewing research or in reporting research results."

The survey found that 60% of the cases cited involved fabrication and falsification and 36% involved plagiarism. Respondents observed misconduct across all scientific ranks—from graduate students through senior faculty. They also indicated that of the 201 instances of misconduct only 58% were reported to institution officials.

Titus and coauthors acknowledge that their study is limited. Because the survey set included only NIH grantees, it may not be representative of all sectors of science, and surveying a broader research pool could reveal a different frequency of research misconduct. However, using the resulting data, the authors extrapolate that about three cases of research misconduct are committed for every 200 researchers working each year. This number translates to significantly more cases of research misconduct than the average of 24 institutional investigations reported to ORI each year, the authors note.

To improve research integrity, the paper recommends six strategies for institutions, including adopting a zero-tolerance policy, protecting whistleblowers, and clarifying how to report misconduct incidents.