Alcohol-Impaired Driving: Average Quantity Consumed and Frequency of Drinking Do Matter

• Published in: Traffic injury prevention Vol. 13; no. 1; pp. 24 - 30
• Main Authors: Birdsall, William C., Reed, Beth Glover, Huq, Syeda S., Wheeler, Laura, Rush, Sarah
• Format: Journal Article
• Language: English
• Published: England Taylor & Francis Group 01-01-2012; Taylor & Francis Ltd
• Subjects: Adolescent; Adult; Age Distribution; Alcohol Drinking - epidemiology; Alcohol Drinking - psychology; Alcohol use; Alcohol-impaired; Alcoholic Intoxication - epidemiology; Automobile Driving - psychology; Automobile Driving - statistics & numerical data; Behavior; Behavioral Risk Factor Surveillance System; BRFSS; Demographics; Drinking; Driving; Drunk driving; Female; Humans; Logistic Models; Male; Middle Aged; Models, Psychological; Risk-Taking; Seat Belts - utilization; Sex Distribution; Traffic accidents & safety; United States - epidemiology; Young Adult; United States
• ISSN: 1538-9588; 1538-957X
• DOI: 10.1080/15389588.2011.629700

Objective: The objective of this article is to estimate and validate a logistic model of alcohol-impaired driving using previously ignored alcohol consumption behaviors, other risky behaviors, and demographic characteristics as independent variables.
Methods: The determinants of impaired driving are estimated using the US Centers for Disease Control and Prevention's (CDC) Behavioral Risk Factor Surveillance System (BRFSS) surveys. Variables used in a logistic model to explain alcohol-impaired driving are not only standard sociodemographic variables and bingeing but also frequency of drinking and average quantity consumed, as well as other risky behaviors. We use interactions to understand how being female and being young affect impaired driving. Having estimated our model using the 1997 survey, we validated our model using the BRFSS data for 1999.
Results: Drinking 9 or more times in the past month doubled the odds of impaired driving. The greater average consumption of alcohol per session, the greater the odds of driving impaired, especially for persons in the highest quartile of alcohol consumed. Bingeing has the greatest effect on impaired driving. Seat belt use is the one risky behavior found to be related to such driving. Sociodemographic effects are consistent with earlier research. Being young (18-30) interacts with two of the alcohol consumption variables and being a woman interacts with always wearing a seat belt. Our model was robust in the validation analysis.
Conclusions: All 3 dimensions of drinking behavior are important determinants of alcohol-impaired driving, including frequency and average quantity consumed. Including these factors in regressions improves the estimates of the effects of all variables.