Middle school outcomes for a population-based randomized preventive intervention trial for conduct problems are reported. Students (n = 361), and their parents from six elementary schools participated in the trial. Randomization into control and preventive intervention conditions occurred at the school level. Schools were located in neighborhoods characterized by high rates of juvenile delinquency. The Linking the Interest of Families and Teachers (LIFT) multi modal preventive intervention, which included behavioral parent management training, child social and problem solving skills training, a recess behavior management program, and a classroom dedicated phone line and answering machine, was offered to students and parents from all fifth and combined fourth-fifth classrooms within preventive intervention schools. Survival analyses were conducted using logistic regression for substance use outcomes and Cox regression for police arrest. The onset of police arrest and patterned alcohol use during the middle school years was less likely for participants in the preventive intervention condition than for participants in the control condition, but no differences were found in terms of onset of tobacco use or marijuana use.