RESUMEN
Research consistently provides evidence for the relationship between the therapeutic alliance (TA) and outcome across various therapies and presenting problems. Depression is considered the leading cause of disability worldwide, and there is substantial evidence for the efficacy for Cognitive Behaviour Therapy (CBT) in its treatment. At present, there is lack of clarity specifically about the relationship between the TA and outcome in CBT for depression. The present review is the first meta-analytic review to explore this relationship and also considering moderators. Within a random-effects model, an overall mean effect size of r = 0.26 (95% CI [.19-.32]) was found, indicating that the TA was moderately related to outcome in CBT for depression. The mean TA-outcome correlation is consistent with existing meta-analysis that looked across a broad range of presenting problems and psychological therapies. A secondary exploratory analysis of moderators suggested the TA-outcome relationship varied according to the TA rater, where the relationship was weaker for therapist raters compared with clients and observer raters. Additionally, the results indicated that the TA-outcome relationship marginally increased over the course of CBT treatment. The results of the meta-analysis are discussed in reference to the wider body of research, methodological limitations, clinical implications, and future directions for research.