Publication bias, the failure of small negative trials to be published, has been well documented in all fields of medicine, including rheumatic disease. Evaluating a treatment by summarizing data from only published trials creates an inaccurately rosy picture of a treatment's efficacy and safety especially when trials have been carried out that remain unpublished. Treatments shown to be ineffective have sometimes been painted as effective in published trial data. Web-based clinical trial registries such as clnicaltrials.gov were created to encourage trialists to publish trial protocols and to even report trial results in these registries. Editors of major medical journals have required prerecruitment registration of trial protocols, creating further pressure to register trials. Current evidence suggests that while trials are increasingly being registered, negative trial results are not necessarily reported there, so that publication bias persists. New strategies to enhance the likeihood of detecting the existence of negative trials include changes in requirements for trial registration and the possible availability of trial data through companies and regulatory agencies.
Disclosure of Interest None declared