Background Highly caffeinated 'energy drinks' ( ED) are commonly consumed and sometimes mixed with alcohol. Associations between ED consumption, risk-taking, and alcohol-related problems have been observed. This study examines the relationship between ED consumption-both with and without alcohol-and drunk driving. Methods Data were derived from a longitudinal study of college students assessed annually via personal interviews. In Year 6 (modal age 23; n = 1,000), participants self-reported their past-year frequency of drunk driving, ED consumption patterns (frequency of drinking alcohol mixed with energy drinks [AmED] and drinking energy drinks without alcohol [ED]), alcohol use (frequency, quantity), and other caffeine consumption. Earlier assessments captured suspected risk factors for drunk driving. Structural equation modeling was used to develop an explanatory model for the association between ED consumption patterns and drunk driving frequency while accounting for other suspected risk factors. Results More than half (57%) consumed ED at least once during the past year. Among ED consumers, 71% drank Am ED and 85% drank ED alone; many (56%) engaged in both styles of ED consumption while others specialized in one or the other (29% drank ED alone exclusively, while, 15% drank Am ED exclusively). After accounting for other risk factors, ED consumption was associated with drunk driving frequency in 2 ways. First, a direct path existed from ED frequency (without alcohol) to drunk driving frequency. Second, an indirect path existed from Am ED frequency through alcohol quantity to drunk driving frequency. Conclusions Among this sample, ED consumption with and without alcohol was common, and both styles of ED consumption contributed independently to drunk driving frequency. Results call for increased attention to the impact of different patterns of ED consumption on alcohol-related consequences, such as drunk driving. [ABSTRACT FROM AUTHOR]