Are antidepressant and antipsychotic drug trials registered? A cross-sectional analysis of registration and reporting of methodologic characteristics.

INTRODUCTION: In 2005, the International Committee of Medical Journal Editors (ICMJE) proposed that all submitted trials in all 11 member journals must be prospectively registered in order to be considered for publication. Registering drug trials was meant to reduce the likelihood of selective reporting. The aim was to determine the proportion of antipsychotic and antidepressant trials that were registered.

METHODS: We searched in Pubmed for all randomized controlled trials of any antidepressant or antipsychotic published between July and December 2014. The primary objective was to determine the proportion of trials that were registered. Secondary objectives included comparing the reporting of methodological details and positive study findings between registered and unregistered trials.

RESULTS: Of the 67 studies identified, 58% were registered. 75% of the antipsychotic trials and 51% of the antidepressant trials were registered, respectively. Registered trials were more likely to report important methodological details associated with risk of bias in RCTs. There was no significant difference in trials reporting positive outcomes for the study intervention between registered and unregistered trials.

CONCLUSION: Approximately 60% of published antidepressant and antipsychotic drug trials during July to December 2014 were registered. Unregistered trials were less likely to report important methodological details.