Aim: To examine the association between alcohol in school environments and adolescent alcohol use over the previous 6 months. Design: A multi-level logistic regression analysis was performed of cross-sectional surveys conducted in 2004, 2005 and 2006. Participants and setting: A total of 52214 students aged 11-19 years from 387 middle or high schools were selected from a nationally representative, multi-stage, stratified probability sampling across Taiwan. Measurements: Information on socio-demographic features and substance use experiences was collected using self-administered questionnaires. The alcohol in the environment was measured using the availability of convenience stores surrounding the schools. Using geographical information systems, the weighted numbers of convenience stores within 1km, a 12-15-minute walk, of a school were calculated. The schools were later categorized into three subgroups via the tertile of nearby convenience stores. Findings: Considering the compositional characteristics, the availability of convenience stores was found to account for 1.5% of the school-level variance of youthful drinking. The odds ratios (95% confidence interval) of alcohol use over the previous 6 months among youth attending schools with medium and high availability were 1.04 (0.96-1.13) and 1.08 (1.00-1.17), respectively, with a P-value of 0.04 in the trend test. Conclusion: The greater availability of convenience stores near a school is associated with an increased risk of alcohol use among adolescents over the previous 6 months.
