Background Habitual sleep patterns may independently affect morbidity and mortality. However, the effect of habitual sleep patterns on the risk for stroke and coronary heart disease is unclear. Methods We evaluated the association between sleep duration and daytime somnolence (often or almost always taking daytime naps) with the incidence of stroke and coronary heart disease in a national cohort of 7,844 adults who participated in the First National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. Cox proportional hazards analyses were used to examine these relationships during the 10-year follow-up. Results After adjusting for differences in age, race, gender, education, cigarette smoking, body mass index, serum cholesterol, systolic blood pressure, and diabetes mellitus, the risk for stroke was increased in persons who reported sleeping greater than 8 hours at night compared with persons who slept between 6 and 8 hours (relative risk [RR] = 1.5, 95% confidence interval [CI] = 1.1 to 2.0). Daytime somnolence was also associated with stroke incidence (RR = 1.4, 95% CI = 1.1 to 1.8). Persons who reported both greater than 8 hours of sleep and daytime somnolence were at the greatest risk for stroke (RR = 1.9, 95% CI = 1.2 to 3.1). Similar results were also found for coronary heart disease, although the results did not reach statistical significance in the multivariate adjusted model. Conclusions Habitual sleep patterns have significant effects on the risk for stroke.