In a Yale School of Medicine analysis of 96 research trial results published in top journals, almost all had at least one discrepancy between what was reported on the public clinical trial registry clinicaltrials.gov and what was posted in the journal article.
"This study raises serious questions about the accuracy of results reporting in both clinical trial registries and publications, and the importance of consistent presentation of accurate results," said Joseph Ross, M.D., assistant professor of medicine and public health at Yale School of Medicine and senior author of the research letter published in the March 12 Journal of the American Medical Association.
In 2007, the Food and Drug Administration mandated that all clinical trial results must be registered within one year of completion through clinicaltrials.gov, a National Institutes of Health online registry and database of clinical trials.
Ross and his colleagues compared the results of 96 trials reported through clinicaltrials.gov and those published in high-profile, peer-reviewed journals between 2010 and 2011. They compared the reported and published results regarding the patients, trial interventions, efficacy, and other results of the studies.
Of the 96 trials examined in this study, 93 of them had at least one discrepancy between the two reported results. "We focused only on reports in top journals such as the New England Journal of Medicine and Lancet, which tend to go through the most scrutiny before being published," explained Ross. "We believe that the results from this study represent the best-case scenario with respect to reporting accuracy in published clinical trials."
According to the analysis, discrepancies found between the clinicaltrials.gov database and published results could result from clerical or typographical errors, failure to report results online, or intentional changes to share more favorable results.
"Fortunately, of the 96 trials studied, only six of the discordant results altered the interpretation of the trial," said Ross. "But there still needs to be greater efforts to ensure accurate reporting in the future."
Other authors on this study include first author Jessica E. Becker, Harlan M. Krumholz, M.D., and Gal Ben-Josef.
The study was funded in part by the National Heart, Lung, and Blood Institute Cardiovascular Outcomes Center Award (1U01HL105270-02), the National Institute on Aging (K08 AG032886), and by the American Federation for Aging Research through the Paul B. Beeson Career Development Award Program.
Citation: JAMA doi:10.1001/jama.2013.285634; Available pre-embargo to the media at http://media.