I used a case-control study to identify an increased risk of motor vehicle crash (MVC) in cigarette smokers as compared to non-smokers. Information on smoking status and potential confounding factors was obtained using a self-administered mail-out questionnaire sent to 1,000 persons known to have had a MVC, and a control group of 1,100 persons with no record of MVC in the previous five years. Assessment of relative risk estimates (RR) and adjustment for confounding factors was done by logistic regression analysis. Smokers had a 1.5-fold increase in risk for MVC over non-smokers (p = .01). Also, an increased tendency to smoke while driving was associated with greater risk of MVC (X2 trend: p = .01). The basis for this association may be: 1. distraction from driving by the act of smoking. 2. behavioural differences between smokers and non-smokers. 3. carbon-monoxide toxicity. Further study is needed to determine the importance of these factors as components of the increase in risk found.