BACKGROUND: Studies such on alcohol and injuries have defined alcohol-related injury as an injury with a positive self-report of alcohol consumption in the 6h prior to the event. However, there is very limited data on the pattern of alcohol use over time of day and day of week among the general population. The aim of this study is to estimate the rate of alcohol use by time of day, and day of week for the U.S. general adult (≥ 18 years) population. METHODS: This study employed the design of a retrospective cohort study using data collected from three waves (2005-06, 2007-08, 2009-10) of the National Health and Nutrition Examination Survey (NHANES). Incidence rates of overall drinking (≥ 10 g of alcohol) and incidence rates of heavy drinking (≥ 40 g of alcohol) were estimated for day of week, and time of day (in hours). Multivariable Poisson regression models were used to investigate the difference between weekend nights and weekday nights. RESULTS: The incidence rates (95% confidence interval) of all drinking episodes were 30.5 (29.2-32.0) per 100 person-days and 24.4 (22.8-26.2) per 100 person-days for weekend and the rest of the week, respectively. The incidence rates of heavy drinking episodes were 11.0 (10.2-11.9) and 7.7 (6.8-8.7) for weekend and the rest of the week. Multivariable analysis indicated that risks of overall drinking and heavy drinking were significantly higher (18% and 34%, respectively) during the weekend nights when compared to weekday nights. It was also observed young adults (18-29 years old) were more likely to increase their alcohol use during weekend nights compared to older age groups. CONCLUSIONS: The general US population, especially young adults are exposed to alcohol and its acute effects at a much higher level during the night, and this in-turn increases the risk of alcohol-related injuries during that time.