Objective: The present study evaluated the efficacy of the Middle School Success intervention (MSS) for reducing substance use and delinquency among girls in foster care, using a randomized controlled trial design. The program was designed to fill a service gap during the summer prior to the middle school transition and to prevent delinquency, substance use, and related problems. Method: One hundred girls in foster care and their caregivers were randomly assigned either to the intervention (n = 48) or to a regular foster care control (n = 52) condition. The girls completed a baseline (T1) assessment and follow-up assessments at 6 months (T2), 12 months (T3), 24 months (T4), and 36 months (T5) postbaseline. Caregivers participated in assessments from T1 through T4. This study is a follow-up to Smith, Leve, and Chamberlain's (2011) study, which examined immediate outcomes at T2. Results: Girls in the intervention condition showed significantly lower levels of substance use than did girls in the control condition at 36 months postbaseline. The group difference was only marginally significant for delinquency. Further analyses indicated significant indirect effects of the intervention through increased prosocial behaviors that led to decreased internalizing and externalizing symptoms and then to lower levels of substance use. The direct effect of the intervention on substance use remained significant in the presence of the indirect effects. For delinquency, the intervention had positive effects mainly through increased prosocial skills. Conclusions: Findings highlight the importance of providing preventive intervention services for early adolescent girls in foster care.