The impact of outlet densities on alcohol-related crashes: a spatial panel approach

A number of studies using cross-sectional data have demonstrated that the availability of alcohol, measured by the number and types of alcohol outlets, is directly related to numerous measures associated with drinking and driving. The current study contributes the first observation of relationships over time between alcohol outlet densities on one hand and both automobile crashes and related injuries on the other hand.

The study examined longitudinal data from 581 consistently defined zip code areas represented in the California Index Locations Database, a geographic information system that coordinates population and ecological data with spatial attributes for areas across the state. Six years of data were collected on features of local populations (e.g., demographics, household size) and places (e.g., retail markets) thought to be related to two measures of automobile crashes (hospital discharges related to car crash injuries geocoded to the zip code of patient residence, and police reports associated with car crashes geocoded to the zip code of crash location). Both crash measures were positively associated with two outlet types: bars, and off-premise outlets. Additionally, restaurants appear to provide a protective effect relative to the residence-based measure. Crash rates were also related to changes in population and place characteristics using random effects models with controls for spatial autocorrelation (nxt=3486 observations). Changes in population and place characteristics of adjacent (spatially lagged) areas were also considered.

Over time, both local and lagged population and place characteristics were related to automobile crash-related measures.

Controlling for cross-sectional differences between zip code areas, changes in numbers of licensed alcohol retail establishments, especially bars and off-premise outlets, affect rates of car crashes and related injuries.