Blinding outcome assessors in randomised controlled trials of psychological interventions: how is it done? A narrative review

BACKGROUND: Blinding of patients, clinicians and outcome assessors is a key measure for minimising bias in randomised controlled trials (RCTs). In trials of psychological interventions, where outcome measures often use subjective measures, the blinding of outcome assessors is particularly important. However, there are few guidelines on how blinding of outcome assessors can be achieved in practice. The aim of this study was to identify procedures used for blinding outcome assessors in RCTs of psychological interventions. METHODS: A narrative review was conducted through a PubMed database search covering the period from January 2023 to December 2024, restricted to studies published in the six highest impact factor journals in the field of psychiatry and mental health. Studies were included if they were (1) an RCT and (2) focused on psychological intervention(s). Descriptive statistics were used to summarise the study characteristics, and content analysis of the published reports was conducted to describe the methods used to maintain blinding of outcome assessors. RESULTS: Fifty-two eligible studies were identified which reported on RCTs of psychological interventions. Of those, the majority reported that outcome assessors were blinded to participants' treatment allocation (n = 44, 85%). Of the studies with blinded outcome assessors, 27 reported information on the procedures used to maintain blinding (61%). Methods for maintaining blinding were avoiding unblinding by participants (n = 16), tracking and assessing blinding efficacy (n = 12), replacing unblinded assessors (n = 11), managing communication within the team (n = 9), clear designation of blind and unblind staff roles (n = 8) and management of records (n = 6). The majority of studies reported a single method for maintaining blinding of outcome assessors. CONCLUSION: While the majority of RCTs of psychological interventions reported that outcome assessors were blind to participants' treatment allocation, there was often no or minimal information about procedures used to maintain blinding. A limitation of this study is that it was restricted to RCTs reported in high-impact journals so it does not provide a comprehensive review of all relevant studies. Maintaining blinding of outcome assessors requires substantial planning and efforts throughout the conduct of an RCT and this should be reported so that conclusions drawn from RCTs can be made with confidence.