Aims The goal of this group-randomized trial was to test the effectiveness of an adapted alcohol use preventive intervention for urban, low-income and multi-ethnic settings.
Design and setting Sixty-one public schools in Chicago were recruited to participate, were grouped into neighborhood study units and assigned randomly to intervention or ‘delayed program’ control condition.
Participants The study sample (n = 5812 students) was primarily African American, Hispanic and low-income.
Intervention Students, beginning in sixth grade (age 12 years), received 3 years of intervention strategies (curricula, family interventions, youth-led community service projects, community organizing).
Measurements Students participated in yearly classroom-based surveys to measure their alcohol use and related risk and protective factors. Additional evaluation components included a parent survey, a community leader survey and alcohol purchase attempts.
Findings Overall, the intervention, compared with a control condition receiving ‘prevention as usual’, was not effective in reducing alcohol use, drug use or any hypothesized mediating variables (i.e. related risk and protective factors). There was a non-significant trend (P = 0.066) that suggested the ability to purchase alcohol by young-appearing buyers was reduced in the intervention communities compared to the control communities, but this could be due to chance. Secondary outcome analyses to assess the effects of each intervention component indicated that the home-based programs were associated with reduced alcohol, marijuana and tobacco use combined (P = 0.01), with alcohol use alone approaching statistical significance (P = 0.06).
Conclusions Study results indicate the importance of conducting evaluations of previously validated programs in contexts that differ from the original study sample. Also, the findings highlight the need for further research with urban, low-income adolescents from different ethnic backgrounds to identify effective methods to prevent and reduce alcohol use.