Risk of Bias in Iranian Randomized Trials Included in Cochrane Reviews

Abstract

Background: Among interventional studies, randomized controlled trials (RCTs) provide the highest level of evidence. However, RCTs can be susceptible to the risk of bias (RoB). Systematic reviews can be performed to appraise RoB in the included articles using evaluative tools. This study aimed to describe the main characteristics and focus on the RoB of RCTs conducted in Iran and included in Cochrane Reviews (CRs). Methods: We searched "Iran" by selecting the "Search All Text" and "Review" fields in the Cochrane Database of Systematic Reviews within Ovid. CRs that included the RCTs conducted in Iran were retrieved. A trial was selected only if it was included in CRs, described as a controlled clinical trial, involved human subjects and CR authors assessed its RoB. The trials were characterized by investigating the relevant articles and the table "Characteristics of included studies" in each CR. The RoB was investigated by collecting the review authors’ judgments based on RoB assessment tables in the CRs. Results: Out of 1166 Iranian RCTs included by 571 CRs, low RoB was found in 44.9% for random sequence generation, 20.8% for allocation concealment, 32.3% for blinding of participants/personnel, 36.5% for blinding of outcome assessors, 56.3% for incomplete outcome data, 41.3% for selective outcome reporting and 53.8% for other sources of bias. Conclusion: The RoB in Iranian RCTs was found to be mostly high or unclear. While this is similar to the global situation, it is recommended that the methodological quality of conducting and reporting RCTs be addressed in Iran.