Alliance and Outcome in Cognitive-Behavioral Therapy for Adolescent Depression

Abstract

This study examined predictive relations between therapeutic alliance and treatment outcomes in manual-guided, cognitive-behavioral therapy for adolescent depression. Fifty-four adolescents met criteria for a depressive disorder and were treated in school-based clinics. Alliance was measured after the third session from both therapist and adolescent perspectives, and change in depressive symptoms was assessed by structured interview and self-report. Two models of alliance–outcome relations were assessed, one direct and one through treatment participation. Results showed significant associations between adolescent-reported alliance and change in depressive symptoms, even after controlling for number of sessions completed. Therapist-reported alliance was only marginally related to outcomes but was predictive of number of sessions completed. On average, alliance showed a modest relation with outcomes (r = .26). Results are discussed in the context of differential alliance–outcome relations in prior studies of cognitive-behavioral compared to nonbehavioral therapy with children and adolescents.

Notes

Note: Adol. All. = adolescent-reported alliance; Ther. All. = therapist-reported alliance; Sessions = number of sessions completed; BDI (pre) = pretreatment Beck Depression Inventory Score; BDI (post) = posttreatment Beck Depression Inventory Score; MDD (pre) = pretreatment major depressive disorder symptoms; MDD (post) = posttreatment major depressive disorder symptoms; BDI Δ = BDI (pre) – BDI (post); MDD Δ = MDD (pre) – MDD (post).

∗p < .05; ∗∗p < .01; ∗∗∗p < .001.

Note: BDI = Beck Depression Inventory; C-DISC = Depressive Symptoms on Computerized Diagnostic Interview for Children; Pre-Symptoms = pre-symptoms on BDI or C-DISC; Sessions = number of sessions completed; Adol. Alliance = adolescent-reported alliance; Ther. Alliance = therapist-reported alliance.