• urinary incontinence;
  • incidence;
  • progression;
  • female;
  • risk factors;
  • cohort

OBJECTIVES: To examine the epidemiology of urinary incontinence (UI) in older women.

DESIGN: Prospective study.

SETTING: Nurses' Health Study.

PARTICIPANTS: Incidence of UI was determined in 23,792 women aged 54 to 79 without UI at baseline. Progression or remission of UI was determined in 28,813 women with UI at least monthly at baseline.

MEASUREMENTS: UI was ascertained according to questionnaires in 2000 and 2002. Rates of incident UI and progression or remission of prevalent UI were calculated. Logistic regression was used to estimate relative risks of UI associated with risk factors.

RESULTS: In women with no urine leakage at baseline, 9.2% reported leakage at least monthly after 2 years. For women with leakage at least weekly, the incidence was 3.6%; of these cases, stress UI had the highest incidence, followed by mixed and urge UI. Relationships between UI and age differed for stress UI, which decreased with age (relative risk (RR)=0.63, 95% confidence interval (CI)=0.43–0.92 for aged 70–79 vs 54–59), and urge and mixed UI, which increased with age (RR=2.28, 95% CI=1.09–4.75 and RR=2.11, 95% CI=1.24–3.61, respectively). For prevalent UI in 2000, 32.1% of subjects with leakage once a month progressed to leakage at least once a week over follow-up. Only 8.9% with frequent leakage in 2000 reported improvement to monthly leakage or less, with 2.0% having complete remission.

CONCLUSION: The incidence of UI is high in older women, and progression from occasional to frequent leaking is common. Urge UI, for which there are limited effective treatments, increases with age, thus research on UI prevention in older women is particularly important.