BACKGROUND Publication bias within systematic reviews may result in incorrect conclusions leading to inappropriate clinical decisions and a decreased quality of patient care. Searching clinical trial registries for unpublished studies is one possible solution to minimise publication bias.
OBJECTIVES To examine rates of clinical trial registry searches in systematic reviews published in respected anaesthesiology journals and whether these searches found trials (or data) eligible for inclusion; to compare rates of registry searches between published reviews and similar reviews within the Cochrane Anaesthesia, Critical and Emergency Care Group; to conduct trial registry searches for a subset of reviews, determining whether eligible studies were overlooked; to investigate whether reporting of results in completed anaesthesia trials on ClinicalTrials.gov followed guidelines.
DESIGN A cross-sectional study of systematic reviews published in 10 anaesthesiology journals and the Cochrane Library.
SETTING AND PARTICIPANTS PubMed and the Cochrane Library were searched for systematic reviews or meta-Analyses.
MAIN OUTCOME MEASURES The primary outcome was the number of systematic reviews that searched clinical trial registries for unpublished trials. Secondary outcomes included the number of registered trials in the ClinicalTrial.gov registry and the number of trials reporting trial results which were available on the ClinicalTrials.gov database and which should have been considered in a systematic review.
RESULTS The PubMed search yielded 507 records, and 415 remained after exclusions. Of these, 49 (11.8%) included a search of clinical trial registries. In total, 12 systematic reviews reported finding unpublished data but only five incorporated the data into their analyses. Of the Cochrane reviews, 58.9% (43/73) reported registry searches. Among a sample of 30 systematic reviews that omitted registry searches, we found many studies within the registries that were probably eligible to be included in the systematic reviews. For completed trials within the ClinicalTrials.gov database, only 15.4% reported results.
CONCLUSION The majority of systematic reviews in anaesthesiology did not include data from clinical trial registries. Exclusion of statistically nonsignificant data may lead to a biased interpretation of the data and hence inappropriate clinical interventions.
TRIAL REGISTRATION Registered in University hospital Medical Information Network (UMIN) Clinical Trials Registry (UMIN000021932).