Related Links

Driving Under the Influence, the Summer 2003 TSC Online Newsletter.

Zeroing in on Drinking-Driving

Mapping, Data Analysis, and Surveys to Help Highlight Problems and Reduce Risks

As part of a California Highway Patrol program targeting alcohol-involved collisions on the state’s highways, the UC Berkeley Traffic Safety Center is providing technical assistance in identifying clusters of collisions using GIS (geographic information systems)-based maps. 

The 15-month project, begun in early 2004, included assessing the current literature on alcohol-involved collisions and collecting and analyzing present collision and demographic data in order to determine how the problem is expected to change during the next five to 10 years.

This work was conducted in the context of a generally observed phenomenon: that serious alcohol-related collisions were on the rise in California. The GIS mapping was intended to bring out more details and provide guidance in setting enforcement priorities. It was also known that rural counties were bearing a large share of the burden.

Urban Counties and Rural Counties Differ in Surprising Ways

Not surprisingly, because they have larger populations, urban counties had the highest number of alcohol-involved fatal collisions. But the highest rates for such collisions (the amount of crashes per a given amount of population) were found in rural counties.  

Alcohol-involved collisions are different from other types of crashes in a number of ways. Most commonly, vehicles involved in drinking-driving collisions crash into an object, not another vehicle, and drivers most commonly were "proceeding straight" just before impact. The second most common behavior was: “driver drove off the highway.” These crashes tend to occur more often at night than in the day, on weekends as opposed to work days, and more occur in the summer months.  

The TSC developed an approximate prediction of future rates of alcohol-related crashes, using arrest numbers and case outcomes, and matching them to key factors. Those included an anticipated increase in vehicle miles traveled and a growth in the driving population of groups that are typically over-represented in DUI-related crashes—younger drivers, male drivers, and Latino drivers. TSC researchers predicted that by 2008, annual alcohol-involved crash fatalities would rise approximately 10 percent, or by an additional 500 deaths.

Hotspot Maps  

To refine the analysis further, TSC researchers used GIS software and other tools to match data about DUI crashes from the California Highway Patrol's (CHP) Statewide Integrated Traffic Records System (SWITRS) with demographic data from the California Department of Finance to create detailed maps that included information about DUI victims and violators to help the CHP shape possible local, regional, and statewide interventions. 

The resulting analysis identified 162 high-density alcohol collision zones varying in size from seven-tenths of a square mile to 71 square miles, ranked by collision count, collision density per square mile, and the percentage of crashes investigated by the CHP. (Not all the crashes happened on roadways that fall under the CHP's jurisdiction.)

Seventy-seven corridors—specific stretches of highway that can also include neighboring high-density zones—emerged as well. After further analysis by project staff, a county in northern California and one in southern California were selected to be targets of anti-DUI campaigns on the basis of a number of factors, including having sufficient corridors subject to CHP jurisdiction to permit an effective effort.             

The TSC work on the project continues as it assists the CHP with assessing optimal enforcement and education campaigns and answering questions like “Why do DUI issues sometimes drop from public attention?” and “How can interest be restored?”  

The TSC is also charged with further understanding the factors that contribute most to rises in alcohol-involved collisions and developing a more detailed picture of the victims. The TSC is considering an attitudinal survey to identify beliefs, habits, and dispositions of drinking drivers regarding perceived danger, laws and penalties, knowledge of past campaigns, and motivating factors for compliance.           

Formal presentations of the results are scheduled for 2006.

Related Links

Driving Under the Influence, the Summer 2003 TSC Online Newsletter.

Download printer-friendly PDF of Newsletter

Download printer-friendly PDF of this article only