Public Release: 

Agreements for industry-sponsored clinical trials often fail to protect researchers' independence

Duke University Medical Center

Durham, N.C. -- Academic medical centers frequently engage in industry-sponsored research that fails to adhere to international guidelines established to protect the integrity of research and the rights and interests of academic investigators, according to a study published in the Oct. 24, 2002 issue of the New England Journal of Medicine. This failure to adhere to guidelines regarding trial design, access to data, and publication rights could pose a variety of risks to investigators and research participants.

Researchers at the Duke University Medical Center and the Duke University School of Law found that agreements between academic medical centers and industry often do not uphold the revised guidelines issued in October 2001 by the International Committee of Medical Journal Editors (ICMJE). The ICMJE guidelines are used by more than 500 medical journals as a standard for researchers submitting papers for publication.

"We didn't expect to find full compliance with the guidelines, but we were surprised by the extent to which the agreements entered into by medical schools did not protect the independence of investigators in clinical studies and the integrity of their research," said Kevin A. Schulman, M.D., professor of medicine at the Duke University Medical Center and lead author of the study. "This is the first study to look at compliance with the new ICMJE guidelines, so we hope it will offer medical schools a road map for how they might improve their agreements with industry sponsors."

The study was funded by the Duke Clinical Research Institute.

The Duke team surveyed 108 member institutions of the Association of American Medical Colleges (AAMC) from November 2001 through January 2002 to assess the extent to which their contracts with clinical trial sponsors complied with three main provisions of the ICMJE guidelines, specifically:
· whether authors of reports of multicenter trials had access to all trial data;
· whether the authors or an independent committee controlled editorial and publication decisions; and
· whether investigators were fully involved in the design and conduct of the trials.

The Duke team found that academic medical centers' agreements with industry sponsors rarely included such provisions: only 1 percent (median score) of agreements required that authors of reports from multicenter studies have independent access to all data from the trial, though institutions serving as coordinating centers for multicenter trials stipulated this level of access at a higher rate (50 percent of contracts). While agreements at both local sites and coordinating centers secured access to data from their own site in most trials, the Duke researchers say this is not adequate to fulfill the ICMJE standards.

"Many institutions and investigators assume that having access to site-level data is enough to guarantee the integrity of the study," said Robert M. Califf, M.D., director of the Duke Clinical Research Institute and senior author of the study. "But publishing data, positive or negative, that are collected at a single site as part of a multicenter trial is not sound scientific practice. The number of patients enrolled at a single site is almost never large enough to yield statistically significant results."

Industry sponsors also typically control decisions regarding the publication of data from clinical trials, according to the researchers. Provisions addressing editorial control of multicenter manuscripts appeared in 40 percent (median score) of contracts, while an independent publications committee and a guarantee of publication were specified in less than 1 percent (median score) of contracts. The agreements also did not ensure dissemination of findings from research on stored biologic materials collected during clinical trials. These materials are often collected during trials for potential future research.

"Institutional agreements often fail to ensure researchers' access to the final data from clinical trials, and there is also no guarantee that the collected data will ever be published or made available for professional use," said study co-author Jeremy Sugarman, M.D., director of the Center for the Study of Medical Ethics and Humanities at Duke University Medical Center. "Patients often participate in clinical trials not only for personal benefit, but also because they believe they are contributing to scientific knowledge as a whole. If trial data are not made available to others, this may break an implicit promise to research participants that their contributions will be used in such a way."

The Duke study also found scant adherence to the ICMJE recommendation for the participation of investigators in the design of trials. Only 2 percent required the presence of an independent executive committee as a condition of participation in a trial, and only 1 percent required a data and safety monitoring board.

The authors note that while current compliance is low, the ICMJE recommendations provide a framework to improve academic freedom and integrity. "To ensure that research is being conducted in the best interests of the public -- that positive and negative trial results are being published and made available to the scientific community -- academic medical centers as a group will need to require that their contracts with industry comply with the ICMJE guidelines," said Schulman.


Joining Schulman, Califf and Sugarman on this study were Damon M. Seils, Justin W. Timbie and Kevin P. Weinfurt, Ph.D., of the Center for Clinical and Genetic Economics; and Daniel B. Mark, M.D., of the Outcomes Research and Assessment Group, all of the of the Duke Clinical Research Institute at Duke University Medical Center; as well as Lauren A. Dame, J.D., of the Center for the Study of Medical Ethics and Humanities at Duke University Medical Center and the Duke University School of Law.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.