Correctable Myths About Research Misconduct in the Biomedical Sciences

Abstract

A recent National Academy report on research integrity noted that policies are not evidence-based, with no formal entity responsible to attend to this deficit. Here we describe four areas of research misconduct (RM) regulations governing Public Health Service funded research that are empirically and/or ethically questionable. Policies for human subject protection, RM and conflict of interest are not harmonized, making it extremely difficult to deal with complex cases which often contain allegations in all of these areas. Second, detection of RM has depended entirely on whistleblowers in spite of evidence of significant under-reporting. Third, the scientific record is far from cleansed of the effects of falsified/fabricated work through current mechanisms of retraction. Finally, lack of fairness in the regulations may reflect lack of a Belmont Report-like document to guide ethics of RM policy. These issues are likely common in other countries. RM regulations should be harmonized with related regulations and their effectiveness tracked, open access to data for independent replication and improved statistical tests are an essential supplement to whistleblowers, correction of the scientific record will require a major effort, and further ethical analysis and guidance are as important as is empirical study for the improvement of RM regulations. Further consideration should be given to assigning current regulations for human subjects protection, RM and conflict of interest to a single authority and to the further development of a Belmont-like report of essential principles, for RM.