AMA wants all clinical trial results in database

An initiative started by the American Medical Association (AMA) to create a national clinical trials registry appears to be gaining momentum around Washington.

It started in late June when the AMA House of Delegates approved a proposal calling for the Department of Health and Human Services to establish a new registry that would ensure all trials, with negative or positive results, are made publicly available.

The AMA also called for institutional review boards to require companies to register in the database as a condition of approval for all trials.

Proposal gaining support 

Since the release of the proposal, a few lawmakers have reportedly jumped on the bandwagon by asking the National Institutes of Health and the FDA to provide guidance as to what can be done to improve such a database, which currently is available under the FDA Modernization Act of 1997.

The law requires companies and other organizations to make public studies of experimental medications for serious or life-threatening diseases. The FDA reportedly does not enforce the regulation.

However, trial data that are available can be found on ClinicalTrials.gov, an Internet databank of study information launched in February 2000 by the NIH’s National Library of Medicine.

Indeed, the Chicago-based AMA took up the issue out of concern that negative or null trial results were not being published. Joseph Heyman, an AMA trustee, said the association believes that pattern of publication might distort the medical literature, affecting the validity and findings of systematic reviews, the decisions of funding agencies and, ultimately, the best practice of medicine.

Also, the AMA contends that investigators and authors are reluctant to submit poor trial results because they don’t believe the medical journals will publish them.

Journals are more interested in publishing studies that are likely to affect clinical practice, and, as a result, confirmatory trials with negative results and trials that show no significant result are less likely to be published, the AMA said.