Rasmussen N, Lee K, Bero L. Association of trial registration with the results and conclusions of published trials of new oncology drugs. Trials 2009,10:116.
Registration of clinical trials was intended to reduce bias toward statistically significant results in published studies. But registration alone may not be enough to reduce selective publication, selective outcome reporting, and biased design. Registration was widely adopted in oncology, but most registered oncology trials remain unpublished. This study compares prevalence of favorable findings in published reports of registered and unregistered trials of new oncology drugs. Trial registration alone does not seem to reduce bias towards results that favour new drugs. These findings support the inclusion of protocols and full results reporting in trial registers.