PurposeThe Brazilian version of the prevention program Unplugged, #Tamojunto, has had a positive effect on bullying prevention. However, the curriculum has recently been revised, owing to its negative effects on alcohol outcomes. This study evaluated the effect of the new version, #Tamojunto2.0, on bullying. For adolescents exposed to the school-based program #Tamojunto2.0, we investigated (1) whether the prevalence of bullying victimization and perpetration was reduced, (2) whether this reduction was moderated by gender, and (3) whether the program's effect on bullying was mediated by adolescents' alcohol use. MethodsA cluster randomized controlled trial was conducted using 5,208 eighth-grade students from 73 Brazilian public schools. Baseline data were collected before program implementation, and follow-up data were collected nine months later. We used a multilevel mixed-effects model to examine the effect of #Tamojunto2.0 on bullying, and a moderation model to test the moderating effect of gender on program outcomes. A mediation analysis was performed to determine lifetime alcohol use as a mediator of the intervention effect on bullying. ResultsWe found that the positive effect of #Tamojunto2.0 on bullying victimization (β = −0.019, 95% confidence interval = −0.035; −0.002) and perpetration (β = −0.027, 95% confidence interval = −0.051; −0.004) was mediated by a decrease in alcohol use, but not moderated by gender. Discussion#Tamojunto2.0 program can be indirectly effective in the prevention of bullying by decreasing adolescents' alcohol use. Moreover, alcohol and drug use prevention programs might also affect bullying outcomes through mediation, and we suggest that future studies consider this.