Background: A new trial should always be justified by reference to earlier, similar trials, if possible, ideally in the form of a systematic review of such trials. Unfortunately, this is not always the case which may lead to superfluous trials exposing participants to known inferior treatment.
Objectives: To evaluate whether planned trials were ethically and scientifically justified by the existing literature. We obtained a cohort of protocols for randomised clinical trials with patient-relevant outcomes approved by an ethics committee in Denmark (October 2012 to March 2013). Trials with surrogate outcomes were excluded.
Methods: We searched for descriptions of systematic reviews, earlier trials, any search strategy used, and procedures for monitoring descriptions of harms emerging in the literature during the trial. We compared this information with trials and reviews in the Cochrane Database of Systematic Reviews, PubMed and Embase. We also screened ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform (ICTRP) to ensure that commercial sponsors did not withhold information of similar trials either planned or conducted simultaneously with the trial protocol we assessed.
RESULTS: We extracted data from 67 protocols with a planned enrolment of 74 998 participants (median 405, range 30 to 18 000). Most trials were either fully or partially industry funded (N = 43) and multinational (N = 38), and most were either Phase 3 trials (N = 26) or studying procedures (N = 23). 42 protocols cited a systematic review of the intervention investigated, and several of the remaining trials were so unique that a systematic review clearly was not needed. Only two protocols provided evidence of having performed a systematic literature search. Procedures for monitoring harms described in the literature during the trial were not described.
Conclusion: Our preliminary results are encouraging. Most contemporary clinical trials with patient-relevant outcomes were ethically justified, however, only 2 described their search strategy and none described monitoring publications of harms.