Trustworthiness assessment of Gloster et al. (2020) ‘Treating treatment non-responders: A meta-analysis of randomized controlled psychotherapy trials’

Gloster et al. (2020) presents the first meta-analysis of the efficacy of psychotherapy in treatment-resistant clients. Here, I present an assessment of the trustworthiness of the results presented in Gloster et al. (2020), following previous work on assessing the trustworthiness of original research (e.g., Wilkinson et al., 2024, Wilkinson et al., 2025) and meta-analyses (e.g., Hussey, 2025; Maassen et al., 2020). Serious concerns are raised about the plausibility of the magnitude of the effect sizes included in the meta-analyses and the correct extraction of effect sizes from original studies. The findings arguably presented below represent clear evidence of major errors that compromise the reliability of the research findings presented in Gloster et al. (2020), therefore, following COPE guidelines, the work requires substantial correction or possibly retraction (COPE, 2019).