Effects of the local alcohol environment on adolescents' drinking behaviors and beliefs.

Aims To examine relationships between characteristics of the local alcohol environment and adolescent alcohol use and beliefs in 50 California cities. Design The study used longitudinal survey data collected from adolescents; city-level measures of local alcohol policy comprehensiveness, policy enforcement, adult drinking and bar density; and multi-level modeling with three levels (city, individual, time), allowing for random effects. Models included interaction terms (time × alcohol environment characteristics) and main effects, controlling for city and youth demographic characteristics. Analyses also examined possible mediating effects of alcohol-related beliefs. Setting Fifty California cities (50 000-500 000 population). Participants Random samples of 1478 adolescents and 8553 adults. Measurements Past-year alcohol use and heavy drinking, and alcohol-related beliefs (e.g. perceived alcohol availability) among adolescents; past 28-day alcohol use among adults; ratings of local alcohol control policies; funding for enforcement activities; bars per roadway mile. Findings Local alcohol policy comprehensiveness and enforcement were associated with lower levels of past-year alcohol use (betas = −0.003 and −0.085, P < 0.05). Bar density was associated with a higher level of past-year alcohol use (beta = 1.086, P < 0.01). A greater increase in past-year alcohol use and heavy drinking over time was observed among adolescents living in cities with higher levels of adult drinking (betas = 0.224 and 0.108, P < 0.01). Effects of bar density appeared to be mediated through perceived alcohol availability and perceived approval of alcohol use. Conclusions Adolescent alcohol use and heavy drinking are related to characteristics of the local alcohol environment, including alcohol control policies, enforcement, adult drinking and bar density. Change in adolescents' drinking appears to be influenced by community-level adult drinking. Bar density effects appear to be mediated through perceived alcohol availability and approval of alcohol use. [ABSTRACT FROM AUTHOR]