Predictors of short-term response and the role of heavy alcohol use in treatment of depression

Background: Depression and alcohol use disorders frequently co-occur. However, research on psychosocial interventions for treating this dual pathology is limited. The Ostrobothnian Depression Study (ODS) aimed to increase the systematic use of evidence-based methods, particularly among patients with comorbid depression and substance use in a naturalistic setting. This is a secondary analysis of the ODS study. The aim of the present study was to explore the predictors of a response to treatment during the first six months of the ODS intervention with a specific focus on the role of comorbid heavy alcohol use. Methods: The study sample (n = 242) comprised psychiatric specialist care patients with depression (Beck Depression Inventory score ≥ 17) at baseline. Patients with a baseline Alcohol Use Disorders Identification Test (AUDIT) score > 10 (n = 99) were assigned to the AUD (Alcohol Use Disorder) group in this study. The ODS intervention comprised behavioral activation (BA) for all and additional motivational interviewing (MI) for those in AUD group. The predictors of response to treatment (minimum of 50% reduction in depressive symptoms) during the first six months were analyzed with logistic regression models. Results: In the total sample at six months (n = 150), predictors of response to treatment were more severe depression (OR 1.10, CI 1.02–1.18), larger amounts of alcohol consumed (OR = 1.16, CI 1.03–1.31) and antipsychotic medication "not in use" (OR = 0.17, CI 0.07–0.44). In the non-AUD group (n = 100), more severe depression (OR 1.12, CI 1.01–1.25) and antipsychotics "not in use" (OR 0.20, CI 0.06–0.67) also predicted a positive response. Among AUD group patients (n = 50), larger amounts of alcohol consumed (OR 1.54, CI 1.04–2.27) and antipsychotic medication "not in use" (OR 0.12, CI 0.02–0.60) predicted a response to the treatment intervention. Conclusions: The severity of symptoms and comorbid disorders were found to predict better treatment response