Endorsement of the ICMJE’s clinical trial registration policy by biomedical journals: A survey of instructions for authors

Article type
Authors
Hooft L1, Molenaar N2, van der Haar S2, Scholten R3
1Dutch Cochrane Centre/Netherlands Trial Register, Netherlands
2Netherlands Trial Register, Netherlands
3Dutch Cochrane Centre, Netherlands
Abstract
Background: Publication and selective outcome reporting bias are recognized concerns in the medical literature. Prospective registration of trials may provide transparency in clinical trial conduct and reporting. Journals play an important role in encouraging investigators to adopt trial registration. In September 2005, the members of the International Committee of Medical Journal Editors (ICMJE) issued a joined registration policy that included that submitted manuscripts would only be considered for publication if they had been registered in a public register before the enrolment of the first patient.

Objectives: To determine to what extent biomedical journals have adopted the ICMJE’s clinical trial registration policy and whether reported data are cross-checked with the prospectively registered data.

Methods: Two researchers searched for items addressing prospective trial registration in the Instructions for Authors on the websites of journals following the ICMJE’s Uniform Requirements for Manuscripts (URM) and journals that endorsed CONSORT. Only Instructions for Authors available in English were analyzed.

Results: We identified 866 biomedical journals. Of those 143 included non-English instructions, were discontinued journals, or journals without a website. Prospective trial registration was required by 319 of the remaining 723 journals (44%) and 145 of these (45%) referred to the WHO International Clinical Trials Registry Platform (ICTRP). No detailed information was given on whether or how submitted manuscripts will be compared with trial registry entries as a component of peer review. Results of an additional survey of journal editors about this topic will be available at the Colloquium.

Conclusions: Although the number of biomedical journals requiring registration is encouraging, there is still room for improvement. The registration policy suggests that submitted manuscripts are compared with trial registry entries as a component of peer review; clear instructions to authors were mostly not given. Reviewers' measure of registration quality should be more than verifying the registration number at manuscript submission.