Are co-occurring behavioural health risk factors related to reach and efficacy of in-person versus computer-generated brief alcohol interventions? Findings from a randomized controlled trial.

OBJECTIVES: The common co-occurrence of behavioural health risk factors (HRFs; e.g. at-risk alcohol use, tobacco smoking, physical inactivity, overweight) greatly impacts the global burden of disease. Little is known about how well patients with multiple co-occurring HRFs are reached by and benefit from recommended behavioural interventions. We investigated whether the number of co-occurring HRFs among at-risk alcohol using hospital patients was associated with reach for brief alcohol intervention and whether it moderated the (comparative) efficacy of in-person intervention delivery versus computer-generated feedback. METHODS: Of 1188 proactively approached at-risk alcohol using general hospital patients (18-64 years), 961 were allocated to three groups: motivational interviewing based in-person counselling, computer-generated individualized written feedback tailored to motivational stage of change, and assessment only. Both interventions included contacts at baseline and months 1 and 3. Outcome was change in alcohol use at months 6, 12, 18, 24. Moderator was a sum score of all co-occurring baseline HRFs next to at-risk alcohol use (0-3). Analyses included multivariate logistic regression and latent growth curve modelling. RESULTS: Reach (p = .677) and intervention efficacy (p = .179) were independent of HRF score. Participants with all 3 co-occurring HRFs decreased alcohol use more strongly after computer-generated feedback (up to month 24, p = .004) and in-person counselling (up to month 18, p = .044) versus after assessment only. Both interventions were equally efficacious across different HRF scores. CONCLUSIONS: Intervention reach and -efficacy were independent from the overall number of behavioural HRFs. PRACTICE IMPLICATIONS: Our findings suggest that proactive, theory-driven approaches of in-person- as well as computer-generated brief alcohol interventions may be beneficial, also to patients with multiple co-occurring HRFs.