Background: All member journals of the International Committee of Medical Journal Editors (ICMJE) now require that clinical trials must be registered in a public trials register before they are considered for publication. Such registration ensures that trial details are available to all researchers, including systematic reviewers, irrespective of the results, which may encourage improvements in trial methods and minimise publication bias. The introduction of the ICMJE requirement in July 2005 is likely to have led to an increased number of trials being registered and there is interest in the extent to which such registers can now be relied upon as a source of trials for Cochrane systematic reviews.

Objectives: To explore the current state of trial registration and to assess whether it is adequate to search for trials, for inclusion in systematic reviews, using only public trials registers. To assess the number and type of trials which would be excluded if unregistered trials were not eligible for inclusion in Cochrane Reviews.

Methods: We identified three recently updated Cochrane Injuries Group reviews with a large number of included trials published since 2005. We examined the proportion of the included trials which had been registered and the trials register in which they appeared. We explored the yield of searches in individual databases compared to searching via the ICTRP search portal.

Results: Results will be presented at the Colloquium.

Conclusions: Using this sample we have examined whether extensive database searches were necessary, or whether we might have found the included trials by searching trials registers alone. In addition we will be able to report on search approaches for trials registers and the number and type of trials which are unregistered. It is recognised that results may not be extrapolated to all reviews.