Aim: This study examines the relationship between physical, socio-economic and social environments and alcohol consumption and drunkenness among a general population sample of drinkers aged 12-17 years. DESIGN, SETTING, PARTICIPANTS AND MEASURES: The study was conducted in Auckland, New Zealand. The design comprised two components: (i) environmental measures including alcohol outlet density, locality-based measure of willingness to sell alcohol (derived from purchase surveys of outlets) and a locality-based neighbourhood deprivation measure calculated routinely in New Zealand (known as NZDEP); and (ii) the second component was a random telephone survey to collect individual-level information from respondents aged 12-17 years including ethnicity, frequency of alcohol supplied socially (by parents, friends and others), young person's income; frequency of exposure to alcohol advertising; recall of brands of alcohol and self-reported purchase from alcohol outlets. A multi-level model was fitted to predict typical-occasion quantity, frequency of drinking and drunkenness in drinkers aged 12-17 years.
Findings: Typical-occasion quantity was predicted by: frequency of social supply (by parents, friends and others); ethnicity and outlet density; and self-reported purchasing approached significance. NZDEP was correlated highly with outlet density so could not be analysed in the same model. In a separate model, NZDEP was associated with quantity consumed on a typical drinking occasion. Annual frequency was predicted by: frequency of social supply of alcohol, self-reported purchasing from alcohol outlets and ethnicity. Feeling drunk was predicted by frequency of social supply of alcohol, self-reported purchasing from alcohol outlets and ethnicity; outlet density approached significance. Age and gender also had effects in the models, but retailers' willingness to sell to underage patrons had no effects on consumption, nor did the advertising measures. The young person's income was influential on typical-occasion quantity once deprivation was taken into account.
Conclusion: Alcohol outlet density was associated with quantities consumed among teenage drinkers in this study, as was neighbourhood deprivation. Supply by family, friends and others also predicted quantities consumed among underage drinkers and both social supply and self-reported purchase were associated with frequency of drinking and drunkenness. The ethnic status of young people also had an effect on consumption.