This project will develop and test the effects of two theory-based interventions to reduce illegal sales of alcohol to underage persons, using a multi-community time-series design. The first intervention is based on a compliance model, where incentives are structured to encourage compliance with laws prohibiting sales to youth, and alcohol outlets are provided with information, skills, training, procedure manuals, and other support needed to optimally comply with the law. The second intervention is based on a deterrence model, where active police and liquor licence enforcement of laws against sales to youth is implemented, significantly increasing the probability of detection for violators, and increasing the probability nd speed of experiencing a penalty. The interventions will be implemented in nine diverse communities that previousluy have considerable experience enforcing laws prohibiting sales of tobacco to youth. In each community majors, police chiefs, and city administrators, have made a commitment to active participation and collaboration in the project. Intervention effects will be assessed using a time-series design with several key enhancements, including multiple treatments, periods of removed or low-intensity intervention alternating with periods of high-intensity intervention and a set of similar communities not experiencing the interventions as a comparison group. The primary outcome measure is based on direct testing of outlets' propensity to sell to youth, using previously developed and tested protocols. Intermediate outcomes are measured via surveys of alcohol outlet owners and managers, direct observation of outlets, and process evaluation indicators of intervention activities. A continuous data collection system will be established to provide weekly measures over multiple years, maximizing power of the time-series design. Main outcomes will be analyzed across the entire set of eight communities to maximize statistical power, while stratified analyses (e.g., urban vs. Suburan) will provide additional information regarding generalizability of observed intervention effects. Statistical analyses will include assessment of simple intercept and slope changes, as well as testing more complex gradual or decaying intervention effects and combinations of effects using Box-Jenkins ARIMA and transfer function analyses and mixed-model regression modeling. Results of the project will be both theoretically and practically significant. Testing compliance and deterrence approaches with a set of targets for change not often previously studied-alcohol outlets-will contribute to the refinement of these approaches. Results regarding the relative effectiveness of the two approaches will be of direct use to community decision-makers nationwide, as they work to prevent youth drinking and alcohol problems in a period of expanded responsibilities and shrinking resources.
| Wagenaar, Alexander C; Toomey, Traci L; Erickson, Darin J (2005) Preventing youth access to alcohol: outcomes from a multi-community time-series trial*. Addiction 100:335-45 |
| Wagenaar, Alexander C; Toomey, Traci L; Erickson, Darin J (2005) Complying with the minimum drinking age: effects of enforcement and training interventions. Alcohol Clin Exp Res 29:255-62 |
