The current study was designed to evaluate the efficacy of a mailed feedback and tips intervention as a universal prevention strategy for college drinking. Participants (N = 1,488) were randomly assigned to feedback or assessment-only control conditions. Results indicated that the mailed feedback intervention had a preventive effect on drinking rates overall, with participants in the feedback condition consuming less alcohol at follow-up in comparison with controls. In addition, abstainers in the feedback condition were twice as likely to remain abstinent from alcohol at follow-up in comparison with control participants (odds ratio = 2.02), and feedback participants were significantly more likely to refrain from heavy episodic drinking (odds ratio = 1.43). Neither gender nor severity of baseline drinking moderated the efficacy of the intervention in these analyses, but more conservative analyses utilizing last-observation carryforward suggested women and abstainers benefited more from this prevention approach. Protective behaviors mediated intervention efficacy, with participants who received the intervention being more likely to use strategies such as setting limits and alternating alcohol with nonalcoholic beverages. Implications of these findings for universal prevention of college drinking are discussed.