ORI findings of scientific misconduct in clinical trials and publicly funded research, 1992-2002

Abstract

Background: Since 1992 the Office of Research Integrity (ORI) had reviewed investigations of scientific misconduct in research funded by the US Public Health Service (PHS). ORI defined scientific misconduct as "fabrication, falsification, plagiarism, or other practices that seriously deviate from those that are commonly accepted within the scientific community for proposing, conducting, or reporting research".

Purpose: The purpose of this study was to summarize the findings and administrative actions reported in ORI notices of scientific misconduct in clinical trials occurring between May 1992 and 2002.

Methods: Findings of misconduct were gathered from publicly available sources: the ORI annual reports and the NIH Guide to Grants and Contracts.

Results: Clinical trials accounted for 17 (13%) of the 136 investigations that resulted in findings of scientific misconduct, and they were noted in 12 (11%) of the 113 brief reports of investigations closed with findings of no scientific misconduct. In clinical trials, the most severe sanction, debarment from US Government funding, was applied in six (35%) of the cases of misconduct compared to 79 (66%) of 119 cases from all other types of research combined. Of individuals cited for misconduct in clinical trials, three (18%) held doctorates in contrast to 81 (68%) in other types of research.

Conclusions: In clinical trials, junior employees may bear the burden of sanction for scientific misconduct. The most frequently applied sanction was the requirement that a plan of supervision of the sanctioned employees accompany any future application for funding which would include them. This imposition of sanction on an individual employee does not address possible causes of misconduct which may be inherent in the overall pattern of leadership, training and supervision in the trial. Furthermore, the definition of misconduct, as interpreted by the Departmental Appeals Board, excludes carelessness and other poor research practices that may lead to dissemination of more incorrect data than misconduct.