Substance use outcomes of patients served by a large US implementation of Screening, Brief Intervention and Referral to Treatment (SBIRT).

Aims: To estimate changes in the substance use behaviors of patients who received services as part of the US Substance Abuse and Mental Health Services Administration's (SAMHSA) Screening, Brief Intervention and Referral to Treatment (SBIRT) grant program.

Methods: We use a pre-post design and performance monitoring data collected by SBIRT organizations. For a sample of 17 575 patients, we compare pre-SBIRT substance use with substance use 6 months after receipt of SBIRT services. SBIRT's correlation with changes in substance use was estimated using generalized linear mixed models to account for the clustering of patients within health-care facility and US state.

Results: From pre- to post-SBIRT we found large and statistically significant decreases for almost every measure of substance use. Model-adjusted means indicate that the prevalence of alcohol use was lower 6 months later by 35.6%, heavy drinking by 43.4% and illicit drug use by 75.8%. Greater intervention intensity was associated with larger decreases in substance use. The study design does not support causal conclusions and estimated decreases in reported substance use are due, at least in part, to a well-known set of confounders and natural substance use patterns that may be unrelated to any particular SBIRT intervention.

Conclusions: Compared with previously published findings on the Screening, Brief Intervention and Referral to Treatment grant program, our estimates of substance use reduction were smaller, but still consistently large in absolute magnitude and within ranges of estimates from past trials of Screening, Brief Intervention and Referral to Treatment.