A crib sheet for Cochrane Review authors’ use of trial registries

Tags: Poster
Abrams A1, Pienaar E1, Kredo T1
1South African Cochrane Centre, MRC, South Africa

Background: Clinical trial registries offer free, useful tools for identifying ongoing trials to be included in updates of reviews, assessing topics where reviews should be undertaken, and as a networking tool for review authors seeking further information or unpublished research. Prospective trial registration is an important means of ensuring transparency and completeness in trial reporting. Cochrane authors face challenges assessing selective outcome reporting when adjudicating bias in trials they include in reviews. A trial registry is a useful cross-referent for judging bias. Increasingly, trial registries allow retrospective registration of trials to ensure that databases make accessible as much information as possible. This shift in practice means reporting identification numbers provided by registries no longer provides authors with proof that a trial’s outcomes have not changed or can be tracked from the trials inception.

Objectives: To describe and demonstrate the important differences across trial registries that relate specifically to Cochrane Review authors ability to judge bias.

Methods: First we describe the utility of a registry in adjudicating bias. We then describe each of the member registries in the WHO Network of Primary Registers, their timing policies for the trials they accept, and the related impact this may have on Cochrane authors. Of the 14 registries, the majority are now accepting retrospectively registered trials. Such a change could affect the utility of the registry as a tool, and we provide warnings to ensure that review authors are aware of the implications.

Conclusions: A trial registry is a free, valuable tool if used correctly. In order to do so, users must be kept abreast of changes made to the registration processes of such registries. This research will provide a clear outline of those changes and the potential ramifications that such changes have for adjudicating bias and inclusion of trials for systematic reviews.