Objective: This study examined the sustained efficacy of a media violence intervention in reducing media violence use, normative acceptance of aggression, and aggressive behavior in adolescents. It used an experimental design to evaluate the effects of the intervention over a period of 30 months. Method: N = 627 German 7th and 8th graders were assigned to a 5-week school-based intervention to reduce media violence use or to a no-intervention control group. Media violence use, normative acceptance of aggression, and aggressive behavior were measured 3 months before the intervention (T1), 7 months post intervention (T2), and at 2 follow-ups 18 (T3) and 30 (T4) months after the intervention. This article focuses on the findings from the 2 follow-ups. Results: Controlling for baseline levels and various demographic covariates, media violence use at T2, T3, and T4 and self-reported physical aggression at T3 were significantly lower in the intervention group, and the indirect path from the intervention to T3 aggression via T2 media violence use was significant. Lower T2 media violence use predicted lower T3 normative acceptance of aggression among participants with lower initial aggression. No effects on nonviolent media use and relational aggression were observed. Conclusion: The findings show that a short class-based intervention can produce lasting changes in media violence use that are linked to a decrease in aggression. (PsycINFO Database Record (c) 2015 APA, all rights reserved)