Background

This paper examines the role of therapeutic alliance in predicting outcomes in a Randomised Controlled Trial of Motivational Interviewing and Cognitive Behavioural Therapy (MICBT) for problematic cannabis use in recent onset psychosis.

Methods

All clients were participating in a three arm pragmatic rater-blind randomised controlled trial of brief MICBT plus standard care compared with longer term MICBT plus standard care and standard care alone. Participants completed measures to assess clinical symptoms, global functioning and substance misuse at baseline, 4.5 months, 9 months and 18 months. Clients and therapists completed the Working Alliance Inventory approximately one month into therapy. Client alliance data was available for 35 participants randomised to therapy and therapist alliance data was available for 52 participants randomised to therapy.

Results

At baseline, poorer client-rated alliance was associated with more negative symptoms, poorer insight and greater cannabis use, whereas poorer therapist-rated alliance was only associated with amount of cannabis used per cannabis using day. Alliance ratings were also positively associated with amount of therapy: client-rated alliance was higher in the longer compared to the briefer therapy; therapist-rated alliance was associated with greater number of sessions attended (controlling for type of therapy) and therapy completion. In predicting outcome, client-rated alliance predicted total symptom scores and global functioning scores at follow-up. Neither client nor therapist alliance predicted changes in substance misuse at any time point.

Conclusions

Findings demonstrate that individuals with psychosis and substance misuse who form better alliances with their therapists gain greater benefits from therapy, at least in terms of improvements in global functioning.