To estimate the association of driver sleepiness with the risk of car crashes.Methods A population-based case-control study was conducted in Shenyang, a northeastern city in China, between November 2001 and July 2002. The case group comprised 406 car drivers involved in crashes, and 438 car drivers recruited at randomly selected sites, and on the day of week, and the time of day when they were driving on highways in the study region during the study period were used as control groups. Face-to-face interviews with drivers were conducted according to a well-structured questionnaire covering the circumstances of their current trip and their background information.Stanford sleepiness scale and Epworth sleepiness scale were used to quantify acute sleepiness and chronic sleepiness respectively. Results There was a strong association between chronic sleepiness and the risk of car crash. Significantly increased risk of crash was associated with drivers who identified themselves as sleepy (Epworth sleepiness score ≥ 10 vs ＜10; adjusted odds ratio 2.07, 95%confidence interval 1.30 to 3.29), but no increased risk was associated with measures of acute sleepiness. Conclusions Chronic sleepiness in car drivers significantly increases the risk of car crash. Reductions in road traffic injuries may be achieved if fewer people drive when they are sleepy.