Data audit by a regulatory agency: Its effect and implication for others

There has been much discussion about the prevalence of scientific misconduct. Investigational drug trials and cancer clinical trials have been audited systematically in recent years. Review of the results of these audits indicates that misconduct is not uncommon: serious deficiencies have been identified in 11% of audits of drug trials by the United States Food and Drug Administration. The trend has been toward fewer of the most serious problems in the most recent years of the audit program, suggesting that it is having a deterrent effect. Still, many problems remain. A series of possible strategies for addressing them are critically assessed. The experience of this program of fers a model approach to evaluating potential misconduct in science against which programs in other agencies and other areas of science should be measured.