Effect of early working alliance on retention in outpatient substance abuse treatment

Abstract

This study examined the client–therapist early working alliance effect on retention in outpatient substance abuse treatment. The study was implemented on naturalistic principles adopting a prospective design. Data were collected in southern and western Finland in a multisite clinical trial including outpatient treatment units (N = 7). The sample consisted of clients (N = 327, 111 women, 216 men) starting a new treatment period. Clients were assigned randomly to therapists (N = 33). Both clients and therapists’ ratings of the working alliance were measured. Generalised linear mixed model was used in the analyses. The covariate was client's per cent days abstinent before treatment. Therapist's rating of the alliance of the first therapy session and the client's percentage of days abstinent predicted treatment retention. The therapists’ ratings were better for clients whose treatment continued than for those whose treatment was discontinued. There was considerable between-therapist variation in retention and this became wider as treatment progressed. Lower per cent days abstinent at baseline negatively predicted retention. Since the therapist's evaluation of the working alliance has proved to be a predictor of retention in treatment and differences between the therapists exist in this particular and also in earlier studies, future research should be directed more towards the therapists and their characteristics associated with good treatment results.

Keywords:

• substance abuse treatment
• treatment retention
• early working alliance
• therapist effect
• per cent days abstinent