Variance in the Efficacy of Brief Interventions to Reduce Hazardous and Harmful Alcohol Consumption Between Injury and Noninjury Patients in Emergency Departments : A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Zerbi, Catherine El and Donoghue, Kim and Boniface, Sadie and Drummond, Colin (2017) Variance in the Efficacy of Brief Interventions to Reduce Hazardous and Harmful Alcohol Consumption Between Injury and Noninjury Patients in Emergency Departments : A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Annals of Emergency Medicine, 70 (5). 714-723e13.

Abstract

Study objective: We adopt a comparative framework to measure the extent to which variance in the efficacy of alcohol brief interventions to reduce hazardous and harmful drinking at less than or equal to 5-, 6-, and 12-month follow-up in emergency department settings can be determined by differences between study populations (targeted injury and noninjury specific). Methods: A systematic review and meta-analysis of randomized controlled trials published before September 2016 was undertaken. Twenty-three high-quality and methodologically similar randomized controlled trials were eligible, with a total number of 15,173 participants included. Primary outcome measure was efficacy of brief intervention compared with a control group in reducing quantity of alcohol consumed. An inverse variance model was applied to measure the effect of treatment in standard mean differences for brief intervention and control groups. Results: At 6-month follow-up, an effect in favor of brief intervention over control was identified for targeted injury studies (standardized mean difference=–0.10; 95% confidence interval [CI] –0.17 to –0.02; I2=0%). For pooled noninjury-specific studies, small benefits of brief intervention were evident at less than or equal to 5-month follow-up (standardized mean difference=–0.15; 95% CI –0.24 to –0.07; I2=0%), at 6-month follow–up (standardized mean difference=–0.08; 95% CI –0.14 to –0.01; I2=1%), and at 12-month follow-up (standardized mean difference=–0.08; 95% CI –0.15 to –0.01; I2=0%). Conclusion: Meta-analysis identified noninjury-specific studies as associated with better response to brief intervention than targeted injury studies. However, the inclusion of injured patients with noninjured ones in the experimental and control groups of noninjury-specific studies limited the interpretation of this finding.