Monday, April 04, 2011

Recording medical systematic reviews

Well-conducted systematic reviews are generally considered higher-calibre evidence than individual trials in decision-making for clinical practice and health policy. But there is increasing evidence that publication bias exists for such reviews, and that non-publication of completed studies is as much of a problem as it is for trials. Increased clarity surrounding systematic review conduct and reporting would be possible if the protocols for systematic reviews, just like those for trials, were registered.

Until now there has not been an overarching registry for recording the existence and development of systematic reviews from inception through to completion. This month, the Centre for Reviews and Dissemination (University of York, UK) announces PROSPERO, its international Prospective Register of Ongoing Systematic Reviews (http://tinyurl.com/6g9xfkx). Registration is free, is available to anyone around the world, and generates a unique identifying number for each registered systematic review, which can (and should) be reported in any publications that arise from the study. Investigators should use the registry to record the existence of the protocol for a planned or ongoing systematic review of health care interventions even before screening studies for inclusion in the systematic review.

No comments: