Efficiency of brief interventions on alcohol-related risks in occupational medicine

Aim. To evaluate the effectiveness of brief interventions (BIs) in reducing alcohol use among hazardous drinkers consulting their occupational doctors. Materials and methods. Design: Randomized controlled trial with 12-month follow-up, conducted between November 2004 and September 2006. Setting: Fifteen French regional occupational medicine centers with 147 doctors and 157 assistants who were trained in BI and screening with the AUDIT questionnaire. Participants: We invited 33,488 individuals to fill out AUDIT. These patients were eligible if their scores were 6-12 for women and 7-12 for men, indicating hazardous drinking without dependence (found in 7.1% of respondents). Intervention: After randomization, BIs (informative advice using motivational approach, with 10-minute average duration) were performed by the occupational doctors. The control group received information booklets from the doctors' assistants. Measurements: Situations were evaluated 12 months after inclusion. Outcome measures: AUDIT scores, self-reported alcohol consumption (SRAC) and biological assays. The main criteria were the differences observed between groups for SRAC, the AUDIT score at follow-up, and any reduction in score between inclusion and final assessment. Success of intervention, which was defined as an AUDIT score below the hazardous drinking threshold at follow-up, was considered to analyze the variables associated with the efficiency of intervention. Results. The analyzed sample included 787 persons, among whom 435 were met again 12 months later. In the BI group, we found a lower AUDIT score (p = 0.01), a higher reduction in reported consumptions (-60 g/week versus -44g/week, p = 0.04) and in AUDIT scores (p = 0.009). In the control group, 44.8% reduced their AUDIT scores below hazardous drinking thresholds, as compared to 51.6% in the BI group (p = 0.15). Success was associated with a significant reduction in biological and clinical indices. Conclusions. Though the high attrition rate led to careful conclusions, BIs seem to be efficient in occupational medicine, in comparison with written information, which also seems to influence drinking behavior.