Association Between Alcohol Brief Intervention and Emergency Department and Inpatient Utilization Over 2 Years

05/08/2026

AI Summary

Brief intervention in primary care associated with lower Emergency Department utilisation over 2 years (adjusted OR 0.96, 95% CI 0.94-0.98)
Brief intervention associated with lower inpatient utilisation (adjusted OR 0.96, 95% CI 0.93-1.00), significant only in first 6 months
Population-based EHR study of 254,190 adults screening positive; IPTW GEE models analysed ED and inpatient utilisation across 24 months

AI Summary

This study aims to examine the association between receiving a BI in the outpatient setting and Emergency Department (ED) and inpatient utilization over 2 years among adults screening positive for unhealthy alcohol use.

Separate generalized estimating equation (GEE) logistic models with inverse probability treatment weighting (IPTW) were fit to examine the associations between BI receipt and ED and inpatient utilization over the 2-year follow-up period.

However, the association between BI and ED utilization attenuated over time, and receiving a BI was only significantly associated with lower inpatient utilization in the first 6 months following index.

Basic summary

Alcohol Clin Exp Res (Hoboken). 2026 Aug;50(8):e70395. doi: 10.1111/acer.70395.

ABSTRACT

BACKGROUND: Brief Interventions (BIs) in primary care are an effective, evidence-based tool for reducing the cost and harm associated with unhealthy alcohol use. However, little is known about the impact of alcohol BIs on the utilization of acute healthcare services over time. This study aims to examine the association between receiving a BI in the outpatient setting and Emergency Department (ED) and inpatient utilization over 2 years among adults screening positive for unhealthy alcohol use.

METHODS: This population-based, retrospective observational study utilized electronic health record (EHR) data collected between January 1, 2013 and June 30, 2019 of 254,190 adult primary care patients who screened positive for unhealthy alcohol use. Patients were grouped by whether or not they received a BI for unhealthy alcohol use from their primary physician at the index date. ED and inpatient utilization was assessed for each 6-month interval following the index screening up to 24 months. Separate generalized estimating equation (GEE) logistic models with inverse probability treatment weighting (IPTW) were fit to examine the associations between BI receipt and ED and inpatient utilization over the 2-year follow-up period.

RESULTS: IPTW GEE models indicated that receiving BI was associated with significantly lower odds of ED (aOR [95% CI] = 0.96 [0.94-0.98]) and inpatient utilization (aOR [95% CI] = 0.96 [0.93-1.00]) over the 2-year follow-up period. However, the association between BI and ED utilization attenuated over time, and receiving a BI was only significantly associated with lower inpatient utilization in the first 6 months following index.

CONCLUSION: The observed reductions in ED and inpatient utilization contribute novel insight into the duration of BI’s harm-reduction effects. Ultimately, our findings support widespread implementation of BI programs in primary care settings as a means of reducing acute healthcare utilization.

PMID:42554234 | DOI:10.1111/acer.70395

Document this CPD