BACKGROUND: Brief alcohol interventions aim at motivating individuals to reduce their drinking and encompass a multitude of activities with limited duration or frequency such as counseling or individualized feedback. Brief interventions are effective in alcohol users who exceed the low-risk drinking limits. Considering the health risks associated with alcohol consumption below these limits, brief interventions should be targeted at all alcohol users to maximize their public health impact.

OBJECTIVE: The study aimed (1) to test the long-term effects of a brief alcohol intervention consisting of computer-generated individualized feedback letters among individuals who consume alcohol, irrespective of how much; and (2) to explore how intervention effects may be moderated by alcohol use severity and school education.

METHODS: In the waiting area of the municipal registry office in Greifswald, Mecklenburg-Western Pomerania, Germany, a general population sample of 1646 adults (n=920, 55.89% women; mean age 31.0, SD 10.8 y) who reported alcohol use at least once in the past year were randomized to an intervention (n=815, 49.51%) or a control group (n=831, 50.49%). The intervention comprised up to 3 computer-generated individualized feedback letters based on the transtheoretical model of behavior change at baseline and after 3 and 6 months. The control group received assessment only at the same time points and no feedback. The outcome was a change in self-reported number of drinks per week from baseline to follow-up after 36 and 48 months. Moderators of intervention efficacy were self-reported alcohol use severity (low risk vs at-risk drinking) and school education (<12 y vs ≥12 y) at baseline. Data were analyzed using latent growth modeling with full-information maximum likelihood estimation, ensuring an intention-to-treat analysis. Bayes factors (BFs) were calculated to estimate the sensitivity of evidence.

RESULTS: Unadjusted and adjusted models revealed no group difference after 36 months (incidence rate ratio 1.05, 95% CI 0.87-1.27; BF=0.37). After 48 months, a decrease in weekly alcohol consumption was observed in the control group and no change in the intervention group (incidence rate ratio 1.29, 95% CI 1.05-1.57; BF=0.16), indicating strong evidence against the hypothesized intervention effect. Intervention efficacy was not moderated by alcohol use severity or school education at baseline.

CONCLUSIONS: In a randomized controlled trial, no evidence for the efficacy of individualized feedback letters was found after 3 and 4 years. Unexpectedly, drinking reductions in the control group and no change in the intervention group were observed 4 years after the study start. Intervention strategies effective in at-risk drinkers may necessitate adaptation for applicability to alcohol users as a whole.