Changes in alcohol beliefs mediate the effects of a school-based prevention program on alcohol use among Brazilian adolescents

Objective: To investigate the mechanisms of the #Tamojunto2.0 program that mediated the prevention of lifetime alcohol and drug use, including drug knowledge, behavioral beliefs, attitudes, decision-making skills, and refusal skills.Methods: A cluster-randomized controlled trial was conducted in 73 public middle schools in three Brazilian cities. The sample included 5208 students (49.4 % girls; Mage = 13.2 years). The intervention group attended twelve #Tamojunto2.0 lessons conducted by their previously trained teachers. The control group did not receive any intervention. Data were collected pre-intervention and at the 9-months follow-up. We performed multiple mediation models (for the whole sample, users, and non-users) with a post-estimation adjustment to standard errors to account for nesting. We analyzed all available mediators simultaneously according to each drug: alcohol, binge drinking, tobacco, marijuana, and inhalant lifetime use. To handle missing data, we used the full -information maximum-likelihood paradigm.Results: Outcomes in the whole sample and among non-users showed that #Tamojunto2.0 indirectly prevented lifetime alcohol use and binge drinking by increasing negative and non-positive alcohol beliefs. Only the direct effect on decreasing lifetime alcohol consumption was statistically significant. However, an indirect increase in binge drinking was observed through knowledge about alcohol, but the direct effect was not statistically significant. No effects were reported for marijuana, tobacco, or inhalants. Among users, no statistically significant effects were found for alcohol or drug use.Conclusions: The results suggest that the #Tamojunto2.0 program was only effective in delaying alcohol consumption via increasing negative and non-positive alcohol beliefs. It seems that mediating mechanisms vary depending on contextual characteristics, differences in socializing among adolescents, features of the educational systems, psychosocial conditions, or, fidelity issues of program implementation.