Questions for FDA, NIH on Failure to Publish Clinical Trial Data
Hat tip to FDAWebview:
Three leading Democratic Congressmen are wondering why companies and researchers who run clinical trials are systematically failing to file final outcomes data on ClinicalTrials.gov as required by the 2007 FDA reform law. A series of articles in January in the British Medical Journal found that only 22 percent of 738 trials reported in 2009 had outcomes data.
Hat tip to FDAWebview:
Three leading Democratic Congressmen are wondering why companies and researchers who run clinical trials are systematically failing to file final outcomes data on ClinicalTrials.gov as required by the 2007 FDA reform law. A series of articles in January in the British Medical Journal found that only 22 percent of 738 trials reported in 2009 had outcomes data.
In a letter to FDA Commissioner Margaret Hamburg and NIH director Francis Collins, Reps. Henry Waxman, Edward Markey and Diana DeGette, all Democrats, asked the government officials if the journal’s findings were accurate, and whether the agency had fined any companies or researchers for failing to publish their outcomes data. Research sponsors can be fined up to $10,000 a day for failing to report results.
Why is this important? Clinical trial sponsors must register a trial on ClinicalTrials.gov prior to its start if they want to publish its final results in any of the leading medical journals. Journal editors demanded that so the initial protocols could be checked in advance of publication as part of peer review. That ensures sponsors won’t cherry pick data, lower the bar on outcomes measures mid-stream or engage in other research tricks to resuscitate a failing trial. Publication of results also provides outside researchers with the raw data needed to conduct meta-analyses of multiple trials. If fewer than one in four trial sponsors are following through, it defeats the whole purpose of the registry.
Yesterday’s letter from Waxman et al to the FDA is an appropriate coda to former FDA Commissioner Andrew Van Eschenbach’s suggestion in yesterday’s Wall Street Journal that new drugs or devices be approved after Phase I safety trials. Evidence of efficacy can come later, he suggested, based on outcomes data drawn from registries. Here we have a real world example of how it works in practice: a government-run registry — with penalties attached — failing miserably at its job of collecting the required outcomes data. I await another op-ed calling for more rigorous enforcement.
