Purpose: Publication bias has a significant impact on the results of systematic reviews. Clinical trial registry searches, which include unpublished research, should be conducted when performing systematic reviews to reduce publication bias. We aimed to analyze the use of clinical trial registry searches in critical care systematic reviews.
Methods: Systematic reviews published between 01/01/2010-02/18/2020 from the top 5 critical care journals were extracted from PubMed and screened for trial registry use. Additionally, of the studies not performing registry searches, we assessed ClinicalTrials.gov for potentially relevant trials that were missed by not performing a registry search.
Results: Three hundred and twenty six systematic reviews were analyzed, of which 37 (11.3%) performed trial registry searches. Of the studies not performing clinical trial registry searches, 56% had at least 1 potentially relevant trial that was missed.
Conclusions: The omission of relevant, unpublished clinical trial results may be negatively impacting the accuracy of critical care systematic reviews. We recommend all systematic reviewers conduct clinical trial registry searches to reduce publication bias.
Keywords: Clinical trial registry; Critical care medicine; Publication bias; Systematic review.
Copyright © 2020 Elsevier Inc. All rights reserved.