BACKGROUND We aimed to explore the association between sleep duration and sleep disorder variables obtained from the National Health and Nutrition Examination Survey (NHANES) and stroke. MATERIAL AND METHODS This cross-sectional study enrolled 10 442 participants from the United States. The outcome variable was stroke diagnosed by a doctor. Sleep disorder variables were assessed via self-report questionnaires and included sleep duration, the time required to fall asleep, frequency of sleep deprivation, frequency of early morning wakefulness, and frequency of middle of the night wakefulness. The odds ratio (OR) and 95% confidence intervals (CIs) were calculated by multivariate logistic regression models. RESULTS Compared with participants with 6 to 7 h of sleep duration, participants with less than 6 h of sleep duration had an increased risk of stroke of 0.97 times (OR=1.97, 95% CI: 1.19-3.29). Participants with a sleep disorder had 0.71 times (OR=1.71, 95% CI: 1.04-2.82) the risk of stroke than those without a sleep disorder. Stroke was significantly associated with sleep deprivation 16 to 30 times a month (OR=1.99, 95% CI: 1.14-3.46), early morning wakefulness 16 to 30 times a month (OR=1.97, 95% CI: 1.20-3.25), and middle of the night wakefulness 16 to 30 times a month (OR=1.81, 95% CI: 1.05-3.09), compared with no sleep deprivation, early-morning wakefulness, or middle of the night wakefulness. CONCLUSIONS Short sleep duration and sleep disorder were associated with an increased risk of stroke, suggesting healthy sleep behaviors may reduce the risk of stroke. However, further studies are needed to confirm the causality and underlying mechanism.