Does the working alliance mediate the therapist competence-outcome relationship in cognitive behavior therapy for depression?

Objective: This study examined whether the working alliance mediated the effect of therapist competence on subsequent depression symptomology during Cognitive Behavioral Therapy (CBT). We also tested the potential moderation effect of alliance on subsequent depressive symptomology, based on participants' cognitive aptitude. Method: A total of 86 sessions were coded as the prediction interval across 50 patient-therapist dyads (age M = 39.22, SD = 8.78; 76% female). While accounting for prior depression, competence, and alliance levels, predictors were assessed early treatment (session 1; n = 45 sessions), mid-treatment (session 12; n = 41 sessions), and depressive symptomology was assessed at the subsequent session to the predictor assessments to investigate within-session variability of process variables. Results: Mediation analysis revealed that the effect of early treatment therapist competence on symptom change was mediated by alliance (indirect effect: beta = -.17, 95% percentile bootstrap CI [-.32, -.01]). The positive association involving early treatment alliance and next session outcome was conditional upon low cognitive aptitude levels. Conclusions: Our result offers preliminary support for alliance as a mediator of the effect of competence, and that alliance-outcome relations vary as a function of client aptitude. These novel findings require replication and extension.