The risk of bias in Iranian randomized controlled trials included in Cochrane Reviews

Article type
Authors
Kabir A1, Karimi Behnagh A2, Eidkhani V3, Baradaran HR4, Kabiri P5, Haghdoost AA6, Mesgarpour B7
1Minimally Invasive Surgery Research Center, Iran University of Medical Sciences, Tehran
2Student Research Committee, Faculty of Medicine, Endocrine Research Center, Institute of Endocrinology and Metabolism, Iran University of Medical Sciences, Tehran
3Student Research Committee, School of Medicine, Shahid Beheshti University of Medical Sciences, Tehran
4Endocrine Research Center, Institute of Endocrinology and Metabolism, Iran University of Medical Sciences, Tehran
5Department of Epidemiology and Biostatistics, School of Public Health, Tehran University of Medical Sciences, Tehran
6Research Center for Modeling in Health, Institute for Futures Studies in Health, Kerman University of Medical Sciences, Kerman
7Cochrane Iran, National Institute for Medical Research Development, Tehran
Abstract
Background: Bias or systematic error can result in over or under reporting of treatment effects. Although randomized controlled trials (RCTs) are seen as the gold standard of clinical research designs, they are frequently at risk of bias (RoB). Cochrane Reviews appraise the RoB in all studies in a review by using a specific domain-based evaluation tool. This comprises a judgement for each item as ‘low risk’, ‘high risk’, or ‘unclear risk’ where there is lack of information or uncertainty over the potential for bias.

Objectives: The purpose of this study was to provide an overview of the risk of bias in RCTs conducted in Iran and included in Cochrane Reviews, as well as to identify those items where improvement is most required.

Methods: We searched Cochrane Database of Systematic Reviews (CDSR) on 5 July 2017 to identify Cochrane Reviews, that included at least one RCT from Iran. We searched for 'Iran' by selecting the 'Search All Text' field and 'Review' in CDSR within the Cochrane Library. We included trials only if they involved human subjects and were described as a controlled clinical trial. We used all RoB judgments for items as assessed by the review authors of the included RCTs, according to the table of the Cochrane tool for assessing RoB. The seven items in a standard RoB table of a Cochrane Review are defined within six domains of selection bias, performance bias, detection bias, attrition bias, reporting bias and other bias.

Results: Among the 552 retrieved Cochrane Reviews, we located 1453 Iranian RCTs; 838 of them have been included in relevant Cochrane Reviews and their RoBs have been assessed. The RCTs were mainly published between 2003 and 2013 and included in 518 Cochrane Reviews published by 50 Cochrane Review Groups. Overall, 43.6% of studies were low risk-of-bias for random sequence generation, 20.5% for concealment of randomisation, 31.4% for blinding of participants/personnel, 34.7% for blinding of outcome assessors, 57.5% for attrition, 40.9% for selective reporting. Only 4.9% of studies were low risk for all rated items. The risk for allocation concealment was assessed as unclear in 72.3% of Iranian RCTs.

Conclusions: Total RoB was mainly high or unclear in Iranian RCTs. Accordingly, special attention must be paid to the methodological quality of RCTs in Iran.

Patient or healthcare consumer involvement: NA