RESEARCH REPORT © 2005 Society for the Study of Addiction doi:10.1111/j.1360-0443.2005.00973.x Addiction, 100, 335–345 Blackwell Science, LtdOxford, UKADDAddiction0965-2140© 2005 Society for the Study of Addiction 100 Original Article Preventing youth access to alcohol Alexander C. Wagenaar et al. Correspondence to: Alexander C. Wagenaar University of Florida College of Medicine Department of Epidemiology and Health Policy Research 1329 SW 16th Street Room 5287 PO Box 100177 Gainesville FL 32608 USA Tel. (352) 265 7220 Fax: (352) 265 8047 E-mail: Wagenaar@ufl.edu Submitted 12 January 2004; initial review completed 20 March 2004; final version accepted 4 September 2004 *Initial findings of this study were pre- sented at the 27th Annual Scientific Meeting of the Research Society on Alcoholism, June 30, 2004 in Vancou- ver, British Columbia. RESEARCH REPORT Preventing youth access to alcohol: outcomes from a multi-community time-series trial* Alexander C. Wagenaar 1 , Traci L. Toomey 2 & Darin J. Erickson 2 University of Florida, College of Medicine, Department of Epidemiology and Health Policy Research 1 and University of Minnesota, School of Public Health, Division of Epidemiology, University of Minnesota, Minneapolis, MN, USA 2 ABSTRACT Aims/intervention The Complying with the Minimum Drinking Age project (CMDA) is a community trial designed to test effects of two interventions designed to reduce alcohol sales to minors: (1) training for management of retail alcohol establishments and (2) enforcement checks of alcohol establishments. Design CMDA is a multi-community time-series quasi-experimental trial with a nested cohort design. Setting/participants CMDA was implemented in 20 cities in four geographic areas in the US Midwest. Measurements The core outcome, propensity for alcohol sales to minors, was directly tested with research staff who attempted to purchase alcohol without showing age identification using a standardized protocol in 602 on-premise and 340 off-premise alcohol establishments. Data were collected every other week in all communities over 4 years. Mixed-model regression and Box–Jenkins time- series analyses were used to assess short- and long-term establishment-specific and general community-level effects of the two interventions. Findings Effects of the training intervention were mixed. Specific deterrent effects were observed for enforcement checks, with an immediate 17% reduc- tion in likelihood of sales to minors. These effects decayed entirely within 3 months in off-premise establishments and to an 8.2% reduction in on-premise establishments. Conclusions Enforcement checks prevent alcohol sales to minors. At the intensity levels tested, enforcement primarily affected specific establishments checked, with limited diffusion to the whole community. Finally, most of the enforcement effect decayed within 3 months, suggesting that a regular sched- ule of enforcement is necessary to maintain deterrence. KEYWORDS Alcohol, deterrence, drinking age, enforcement, RBS, youth. INTRODUCTION Despite reductions in youth alcohol consumption over the past two decades, drinking among teens in the United States is still high (Johnston, O’Malley & Bachman 2003). Alcohol consumption by youth is linked to a range of health and social problems, including traffic crashes, risky sex, assault, suicide, drowning and recre- ational injuries (Spain et al. 1997; Borowsky, Ireland & Resnick 2001; Hingson et al. 2002; Dunn, Bartee & Perko 2003). One reason many teens drink is that they have easy access to alcohol from various sources, including directly purchasing alcohol from commercial establish- ments, despite laws prohibiting such sales to youth under