INTRODUCTION: Demographic disparities in recurrent events among stroke survivors have received some attention, but little is known about the impact of gender. We explored whether gender was associated with having a recurrent event, after accounting for health status and SES, in a multi-ethnic stroke cohort. METHODS: The Stroke Warning Information and Faster Treatment study (SWIFT) was a randomized stroke preparedness educational intervention, which prospectively enrolled mild and moderate stroke/TIA patients able to sign informed consent and identified at the NY Presbyterian Medical Center from 2005 to 2010. We defined a recurrent event as having ≥1 of the following post-enrollment: a stroke, TIA, MI, death, migraine, or significant stroke mimic. Gender was dichotomous. Health status included age, race/ethnicity, comorbidities, and CVD history. SES included education, employment, and insurance. We conducted chi-squared tests and t-tests. We used multiple logistic regression to assess the impact of gender, after adjustment for health status and SES, on having a recurrent event. Effect modification of the adjusted gender difference by race/ethnicity was also examined. RESULTS: The cohort included 1203 patients, of which 31% (n=375) had 580 recurrent events. Intervention was not associated with gender (p=0.35) nor having a recurrent event (p=0.07). Females (n=596, 49.5%) were older (p=0.002), had a higher proportion of comorbid hypertension (p=0.019), had less non-Hispanic Whites (p=0.028), were less educated (p=0.022), and had more Medicaid recipients (p<0.001), than males. About 35% of females and 28% of males had a recurrent event (p=0.006). The adjusted odds of a recurrent event were 1.35-fold (95% CI=1.04-1.75) higher for females than for males. Race/ethnicity modified the association. Females, compared to males, were similar for Blacks (OR=1.26, 95% CI=0.68-2.33) and Latinos (OR=0.87, 95% CI=0.60-1.24), yet the difference was substantial among Whites (OR=3.16, 95% CI=1.84-5.41). CONCLUSION: While age has historically been suggested as an explanation for poorer functional scores post stroke, the nature of these recurrent events among women with milder stroke/TIA and its intersection with race/ethnicity needs further exploration.