The location and concentration of alcohol outlets influences individual patterns of alcohol use. In Millar and Gruenewald's study on spatial models, a community prevention program that sought to alter the density of alcohol outlets was evaluated (1997).

They chose to use to analyze the issue because it permitted them to place outlets in a map layer above drunk driving events, which made viewing relationships between problem behavior and alcohol outlets more simple (Millar and Gruenewald 1997).

Indicators of alcohol problems such as police calls to alchol outlets, public intoxication arrest data, DUI data, and alcohol-related crashes were mapped as point data (Millar and Gruenewald 1997). Geographic boundaries of areas of average incomes and average drinking patterns were also looked at using GIS (Millar and Gruenewald 1997).

In one specific part of Miller and Gruenewald's study, data from 4 communities in California was tesselated into 102 units to be used for spatial analysis of self-reports of drinking and driving as it relates to alcohol outlet density (retail outlets/ square mile). The data was mapped using GIS, and analysis showed that units with a common border were similar in terms of self-reports of drinking and driving (Millar and Gruenewald 1997).