Background: There are strikingly few national data available to describe sex differences in age-specific stroke incidence. Methods: REGARDS is a national, population-based, longitudinal study of black and white participants aged > 45 years old, with oversampling of blacks and residents of the stroke belt. Between 2003 and 2007, 30,239 participants were enrolled and examined; follow-up is every 6 months by telephone for self- or proxy-reported stroke, with retrieval and adjudication of medical records by physicians. This analysis included 27,756 participants with follow up data who had no physician-diagnosed stroke at baseline. Stroke incidence rates were calculated as the number of stroke events divided by the person-years at risk with 95% confidence limits. Proportional hazards models were used to assess the race-specific association of sex with stroke risk by age strata (<65, 65–74, and 75+) after adjustment for socioeconomic factors, and Framingham stroke risk factors. Results: There were 613 incident strokes events over 135,551 person-years of follow-up. Stroke incidence rates increased with age (from 237/100,000 to 1003/100,000), and were higher in men than women in both blacks and whites (left panel of figure). After multivariable adjustment, men had higher risk than women at younger ages (<75) but for the 65–75 age group, the difference is larger for blacks than whites (right panel of figure). Discussion: These national data confirm the patterns in male/female stroke risk observed in the Greater Cincinnati/Northern Kentucky Stroke Study, with smaller sex differences at older ages, and for men, larger excess risk in whites than blacks.