Underage drinking, brief interventions, and trauma patients: Are they really special?

Abstract

Background: While the relationship between underage drinking and injury has been well established, few studies have examined whether presenting for an acute injury moderates the efficacy of a brief intervention (BI) on alcohol misuse.

Methods: Patients (aged 14-20 years) in the emergency department screening positive for risky drinking (Alcohol Use Disorders Identification Test-Consumption score) completed a baseline assessment, were randomized to conditions (a standalone computer-delivered BI [n = 277], a therapist-delivered BI [n = 278], or a control condition [n = 281]), and completed a 3-month follow-up. This secondary analysis of Project U-Connect examined regression models (controlling for baseline values) to examine the main effects of injury and the interaction effects of injury by BI condition on alcohol consumption and consequences.

Results: Among 836 youth enrolled in the randomized controlled trial (mean age, 18.6 years; 51.6% were male; 79.4% were white), 303 (36.2%) had a primary complaint of intentional or unintentional injury. At baseline, injured patients were more likely to be male (p < 0.001) and have higher alcohol consumption (p < 0.01), but were less likely to misuse prescription drugs (p = 0.02) than those presenting for medical reasons. Regression models (controlling for baseline values) demonstrated that injury presentation predicted greater alcohol consumption prior to a BI. The computer BI was more effective at reducing alcohol consequences among those presenting with injury than those presenting for other reasons. Injury did not affect the efficacy of the computer BI on alcohol consumption, and injury did not affect the efficacy of the therapist BI on alcohol outcomes.

Conclusions: A therapist or computer BI reduced alcohol consumption and consequences among risky drinkers regardless of reason for emergency department presentation highlighting the opportunity to reach a broad array of youth. Although the therapist BI was not moderated by injury presentation, the computer BI was particularly effective at reducing alcohol consequences among those presenting with injury at 3-month follow-up.

Level of evidence: Therapeutic/care management study, level III.

Trial registration: ClinicalTrials.gov NCT01051141.

Figure 1

Study Design Flowchart.

Figure 2. Percent Change in Mean Alcohol Consequences (from Baseline to 3-Month Follow-up) by Condition and ED Presentation (Injury, Non-Injured)