Access to alcohol outlets and harmful alcohol consumption: a multi-level study in Melbourne, Australia

Aims To assess the association between access to off‐premises alcohol outlets and harmful alcohol consumption. Design, setting and participants Multi‐level study of 2334 adults aged 18–75 years from 49 census collector districts (the smallest spatial unit in Australia at the time of survey) in metropolitan Melbourne. Measurements Alcohol outlet density was defined as the number of outlets within a 1‐km road network of respondents' homes and proximity was the shortest road network distance to the closest outlet from their home. Using multi‐level logistic regression we estimated the association between outlet density and proximity and four measures of harmful alcohol consumption: drinking at levels associated with short‐term harm at least weekly and monthly; drinking at levels associated with long‐term harm and frequency of consumption. Findings Density of alcohol outlets was associated with increased risk of drinking alcohol at levels associated with harm. The strongest association was for short‐term harm at least weekly odds ratio (OR) 1.10, 95% confidence interval (CI) 1.04–1.16]. When density was fitted as a categorical variable, the highest risk of drinking at levels associated with short‐term harm was when there were eight or more outlets (short‐term harm weekly: OR 2.36, 95% CI 1.22–4.54 and short‐term harm monthly: OR 1.80, 95% CI 1.07–3.04). We found no evidence to support an association between proximity and harmful alcohol consumption. Conclusions The number of off‐premises alcohol outlets in a locality is associated with the level of harmful alcohol consumption in that area. Reducing the number of off‐premises alcohol outlets could reduce levels of harmful alcohol consumption.