Assessing risk of bias judgments for blinding of outcome assessors in Cochrane reviews

Adequate judging of risk of bias (RoB) for blinding of outcome assessors (detection bias) is important for supporting highest level of evidence. Judgments and supporting comments for detection bias were retrieved from RoB tables reported in Cochrane reviews. We categorized comments, and then compared judgment and supporting comment with instructions from the Cochrane Handbook. We analyzed 8656 judgments for detection bias from 7626 trials included in 575 reviews. Overall, 1909 judgments (22%) were not in line with the Cochrane Handbook. In 9% of trials, the authors split the detection bias domain according to outcomes. Here, prevalence of inadequate judgments was 19%. Interventions to improve RoB assessments in systematic reviews should be explored.