Peer overestimation, social identification, and Personalized Normative Feedback in brief alcohol interventions.

Personalized Normative Feedback (PNF) is a widely used component of brief interventions (BI). However, little is known about the moderators that may enhance its effectiveness. To examine whether the overestimation of peers' drinking and social identification with them moderate the efficacy of PNF as a component of BI, we conducted a secondary analysis of a randomized clinical trial. We included 806 students who reported high-risk drinking and randomized them into: (1) control group (CG), (2) BI, or (3) BI with PNF (BI+PNF). We assessed quantity and frequency of drinking, frequency of heavy episodic drinking, and alcohol-related problems in the last 12 months at baseline and at a 3-month follow-up. Moderators were the overestimation of consumption and social identification with same-sex and age peers. We performed Generalized Estimating Equations, adjusted for sex and age, and complemented these analyses with clinical significance analyses using an intention-to-treat approach. Among students who overestimate peers' consumption, BI and BI+PNF were effective in reducing alcohol consumption outcomes. In contrast, among students who did not overestimate, BI was more effective than BI+PNF in reducing the frequency of heavy episodic drinking. Contrary to our expectations, we found no differences between receiving BI and BI+PNF for students with high identification. To advance in implementation, we recommend BI without PNF, as it is effective across all student groups. To ensure suitability, future studies should examine the reference group used in PNF, ranging from more proximal to more general groups.