A school-based, multicomponent drug prevention model operating in a middle school in Charleston, South Carolina was evaluated. The project included schoolwide curricular changes, instructional changes, changes to school nonns aimed at all staff and students in the school, and changes targeting the 10% of the students in the school judged to be at highest risk for drug involvement. The evaluation results to date indicate a positive effect of the program on grade point average and a decrease in peer drug influence for high-risk program students as compared with high-risk comparison students. Although some negative effects also were observed during the 1992-1993 school year, those were likely a result of the mix of services provided that year and the influence of variables outside theframework of the program. As the services were modified, the negative effects disappeared. Morefine-grained analyses are under way.