OBJECTIVES: When conducting a systematic review it is common practice to search for peer-reviewed publications and conference proceedings to identify studies relevant to a research question. However, information about studies is increasingly available through other sources and can be of importance in systematic reviews.

Clinical trials registries (CTRs) are increasingly providing unpublished results of studies that can be used in systematic reviews. Clinicaltrials.gov is one of the most commonly used CTRs and provides search facilities that enable the identification of trials through common search terms. In addition, there is the potential to request information from study sponsors through clinicalstudydatarequest.com. This website is supported by several prominent study sponsors and allows reviewers to request access to unpublished data which may be of importance in a systematic review. METHODS: We searched two disease areas (melanoma and juvenile idiopathic arthritis (JIA)) for instances where there were discrepancies in reporting of endpoints between peer-reviewed publications and the clinicaltrials.gov webpage for corresponding trials. We submitted requests to clinicalstudyrequest.com for additional information on trials in both disease areas. RESULTS: We identified additional reporting of subgroups as well as efficacy endpoints in clinicaltrials.gov that were not available in peer-reviewed publications. Results included one trial in melanoma (METRIC) which reported on only mixed line patients in a peer-reviewed publication; results stratified by previous therapy were available from the CTR.  In addition, results from our search in JIA included additional reporting of efficacy outcomes such as change in component scores from baseline. We detail length of time for response and issues with submission of data requests to clinicalstudyrequests.com. CONCLUSIONS:  We conclude that sources other than peer-reviewed articles and conference abstracts should be considered when identifying study information that may be relevant to a particular review. Unpublished data may be available that can impact a systematic review and evidence synthesis.