Therapist effects in three treatments for alcohol problems

Prior research indicates that therapist effects can be sizeable in substanceabuse treatment. Therapist differences were examined within a multisite (N = 1726) randomized trial of three psychosocial treatments for alcohol problems: twelve-step facilitation (TSF), cognitive-behavioral skills training (CBT), and motivational enhancement therapy (MET). Therapists (N = 80) were nested within treatments, selected and trained for expertise in a specific approach. This report describes: (1) differences in therapist characteristics across the three treatments; (2) the magnitude of therapist effects within each treatment; and (3) exploratory analyses of therapist attributes associated with successful outcomes. Therapist characteristics differed between TSF and the other two conditions. Significant therapist effects were found in client satisfaction and outcomes, even after covarying for effects of treatment sites and client baseline characteristics. Specific therapist attributes were predictive of client outcomes. Outlier therapists whose caseloads showed unusually poor outcomes accounted for most of the observed effects.