OBJECTIVE: To describe the prevalence of urinary incontinence during exercise in women, estimate whether exercise intensity is related to leakage severity, and report women’s assessments of incontinence as a barrier to exercise. METHODS: Questionnaires were mailed to 5,130 women aged 18-60 years drawn from National Family Opinion research panels. Physical activity levels were assessed by the International Physical Activity Questionnaire. Urinary incontinence, defined as involuntary leakage in the last 30 days, was assessed with the Sandvik Severity Index and a global measure of bother. Prevalence estimates were adjusted via post-stratification weighting. RESULTS: A total of 3,364 eligible women responded (68%), of whom 34.6% were insufficiently active (95% confidence interval [CI] 32.7-36.5%), 29.7% were sufficiently active (95% CI 27.9-31.5%), and 35.7% were highly active (95% CI 33.8-37.6%). Urinary incontinence prevalence was 34.3% (95% CI 32.5-36.1%). One in seven women experienced urinary leakage during physical activity; this was more common among highly active (15.9%) than less active women (11.8%) (P = .01). After adjusting for age, comorbidities, education, and race, women with very severe incontinence were 2.64 times (95% CI 1.25-5.55) more likely to be insufficiently active than continent women. Incontinence was a moderate or substantial barrier to exercise for 9.8% (95% CI 8.8-10.9%) of women. Of incontinent women, the proportion for whom incontinence was a moderate or substantial barrier to exercise increased with each severity category: 9.2%, slight; 37.8%, moderate; 64.6%, severe; and 85.3%, very severe (P < .01). CONCLUSION: Urinary incontinence is perceived as a barrier to exercise, particularly by women with more severe leakage.