Prediction of outcome by multimodal monitoring of treatment processes in alcoholism therapy

Objectives: This study investigates treatment processes of 64 chronic alcohol dependent patients at 3 time-points, t1 (week 3), t2 (month 6), and t3 (month 12) during the first year of the Outpatient Longterm Intensive Therapy for Alcoholics (OLITA). It focuses on two major objectives: (1) to analyze the change of therapeutic processes between t1, t2, and t3, (2) to predict cumulative abstinence probability during a follow-up of 4 years. Methods: 175 video recordings of therapy sessions were analyzed with the VAMP (Video-Assisted Monitoring of Psychotherapeutic processes in chronic psychiatric disease). Changes of therapeutic processes over time were tested with repeated measures ANOVA, prediction of cumulative abstinence probability by therapy processes was determined with Cox Regression Analysis. Results: Most of the processes changed only marginally between t1, t2 and t3. By integrating the process variables with the highest predictive validity the composite score TOPPS (Therapy Orientation by Process Prediction Score) was constructed. It strongly predicts cumulative abstinence probability at all of the 3 time-points (p<0.001). Patients who relapsed after 12 months showed continuously low TOPPS from t1 to t3. However, patients who maintained long-term abstinence had high TOPPS at t1 which increased slightly between t2 and t3 (p<0.015). Conclusions: The results suggest to employ the TOPPS in addiction therapy as a guideline of individual relapse prevention strategies.