The Registration of Clinical Trials

Posted July 22, 2005
Addendum March 12, 2013
 
The effectiveness of medical interventions should be based on the results of all properly conducted clinical trials, whether or not the trials have been published. The results of unpublished trials are systematically different from those that are published; they tend to be "negative" (find no effect or harm) or fail to support the interests of the funding agency. Therefore, relying on published trials alone can provide a biased view of effectiveness and safety.
 
Registration of clinical trials at their inception, in widely available registries, makes it possible for all stakeholders to take unpublished trials into account when summarizing the evidence for an intervention's effects. Information in registries can also prompt efforts to discover the original objectives and results of unpublished trials and the reasons why they were not published.
 
For these reasons, the World Association of Medical Editors supports efforts to register all clinical trials at their inception. Because registration is useful only to the extent that it includes all trials, it should be required of the research community as a whole and not voluntary according to the source of funding or preferences of the investigators. The contents of registries should be freely accessible and searchable to anyone in the world who wants to examine them.
 
Many practical issues will need to be resolved before universal trial registration is a reality. A single, worldwide registry would be ideal but multiple registries are more feasible in the short term because trials are conducted in many regions of the world and in many clinical research areas and there is currently no single body that oversees the world's research. To ensure the integrity of registries, they should be managed by groups without potential conflict of interest in the results of the trials they include.
 
Journal editors should support the development of registries by participating in efforts to develop a consensus on requirements for registry contents, responsibility, access, search ability, and comprehensiveness and by promoting their implementation. When suitable registries are available, editors should require prior registration of all trials published in their journals. Editors of small journals with limited resources should be engaged in this process to ensure that their role in registries is feasible in their setting.
 
Addendum, March 12, 2013, in support of the Alltrials.net initiative and posted on their website:
 
World Association of Medical Editors (WAME) seeks to promote the highest standards in conducting, reporting, and disseminating medical research in the service of improving personal and public health worldwide. To this end, WAME lends its full support to the AllTrials initiative, extending WAME’s policy that all clinical trials should be registered. We echo the call to governments, regulators, research and medical associations, universities, ethics committees, as well as medical journals and publishers, to commit to publishing in full all treatments tested in clinical trials.