FDA wants more transparency for meta-analyses

FDA held a workshop on Monday to discuss what requirements the agency should include in draft guidance on the design of meta-analyses of randomized controlled trials. Workshop attendees -- which included FDA officials, academics and industry representatives -- agreed that any meta-analysis submitted to the agency by a third party should be held to the same design requirements as randomized controlled trials, including a pre-planned protocol and predefined hypothesis as well as details about the rationale for the selection of trials that were included and excluded.

"This is the criteria that you are going to have to deal with so that we can review your meta-analysis," said Robert O'Neill, a statistical advisor at FDA's Center for Drug Evaluation and Research (CDER) and former director of the center's Office of Biostatistics. O'Neill added that the guidance is intended to address meta-analyses from third parties hoping to get "some kudos for unexpectedly throwing a safety study over the fence and saying 'FDA will have to deal with it.'" ...