Association between urinary incontinence and device-measured physical activity: A cross-sectional study.

Objectives: Determine the association between urinary incontinence (UI) and physical activity in a well-characterized and racially and ethnically diverse cohort of women. Secondarily, we aimed to determine the association between the frequency of symptoms of UI and UI type (stress, urge, mixed) and accelerometer-based measures of physical activity and sedentary behavior.

Study design: Cross-sectional study of 1,098 women who were enrolled in the Study of Women's Health Across the Nation (SWAN), an epidemiologic study of community-dwelling women representing five racial/ethnic groups and seven geographic sites across the United States. For the primary analysis, presence of UI (<1 vs. >1 leakage episode/month) was the primary predictor of selected measures of physical activity including moderate-intensity physical activity (MVPA) minutes, number of MVPA bouts, duration of MVPA bouts, and sedentary minutes. Unadjusted models, and models adjusted age, race and ethnicity, body mass index (BMI), parity, smoking status, mental health status, total comorbidity score, and accelerometer wear time were performed using multivariate linear regression. The same analytic approach was repeated for frequency of episodes of UI and UI subtypes of stress (SUI), urge (UUI), and mixed (MUI) incontinence.

Results: Average age was 65±2.7 years, and was comprised of 23.5% Black, 12.5% Chinese, 11.3% Japanese, 4.9% Hispanic/Latina, and 47.8% White women. Prevalence of UI was 65%. In unadjusted analyses, presence of any UI was associated with increased sedentary minutes, decreased MVPA bouts, and decreased amount of time spent in MVPA bouts, but these associations were eliminated when accounting for covariates, particularly age and body mass index. However, frequency of UI episodes (> 1/week) was inversely related to duration of MVPA bouts (beta coefficient: -3.0 minutes, 95% CI: -5.77, -0.24 minutes). This association was strengthened for urgency urinary incontinence (beta coefficient: -3.70, 95% CI: -6.64, -0.77). Any significant unadjusted associations in the remaining variables were primarily accounted for by BMI.

Conclusions: Presence of UI was not associated with physical activity estimates. More frequent UUI episodes was associated with less time spent in MVPA. Longitudinal studies are needed to further understand impact of UUI on achievement of physical activity guidelines.