OBJECTIVE: To test the hypothesis that urinary incontinence (UI) is associated with incident parkinsonism in older adults.
METHODS: We used data from 2,617 older persons without dementia. Assessment included baseline self-report UI and annual structured exam which assessed parkinsonian signs, motor performances, cognitive function, and self-report disabilities. We used a series of Cox proportional hazards models to examine the association of UI with parkinsonism and adverse health outcomes and a mixed-effect model to examine the association of UI with the annual rate of cognitive decline. In decedents, regression models were used to examine if UI proximate to death was related to postmortem indices of neuropathologies.
RESULTS: At baseline, more than 45% of participants reported some degree of UI. Over an average of nearly 8 years of follow-up, UI was associated with incident parkinsonism (hazard ratio [HR] = 1.07, 95% CI = 1.02, 1.12), death (HR = 1.07, 95% CI = 1.03, 1.11), incident ADL disability (HR = 1.11, 95% CI = 1.07, 1.16), and incident mobility disability (HR = 1.07, 95% CI = 1.02, 1.13). UI was not related to incident MCI (HR = 1.02, 95% CI = 0.97, 1.07), incident AD dementia (HR = 1.00, 95% CI = 0.95, 1.05) or to the rate of cognitive decline (Estimate = -.002, standard error = .002, p = .167). In 1,024 decedents with brain autopsy, UI proximate to death was related to PD pathology (Lewy body pathology and nigral neuronal loss), but not Alzheimer's disease pathology or other age-related neuropathologies.
CONCLUSION: UI in older adults is associated with incident parkinsonism and may identify older adults at risk for accumulating PD brain pathology.
