Incident dementia in female Medicare beneficiaries with urinary incontinence versus controls

Objectives: Investigate the association of urinary incontinence (UI) in women aged ≤70 years old with the hazard of incident dementia using Medicare 5% Limited data set (LDS).

Methods: This retrospective, matched, case-control study was conducted using 2011-2020 Medicare 5% LDS data tables. Female beneficiaries ages 67 to 70 years old were excluded if international classification of disease (ICD) codes for neurologic conditions or paralysis were present during a two-year look back period. The UI cohort was identified by an index UI diagnosis between 2013 and 2015, using a 2 year look back period to ensure the first diagnosis in outpatient, inpatient, skilled nursing facility, or carrier files. Controls were matched by age, dual Medicare/Medicaid eligibility, and calendar year. The cumulative incidence of the outcome (1 or more ICD codes for dementia after the index period) or death was estimated over five years. Cox proportional hazard models assessed the association of variables with dementia using p< 0.05 as the threshold for significance. The model was adjusted for index age, dual Medicare/Medicaid status, race, and comorbidities.

Results: The sample of 17,302 women were on average 68.5 (SD 1.1) years old at the index period. The UI (n=8651) and control (n=8651) cohorts were similar at baseline in age, race, dual eligibility, history of alcohol abuse, anxiety, and depression (Table 1). Women in the control group were more likely to be black or other/unknown race (versus white) and have cardiac disease, hypertension, diabetes, or drug abuse at baseline. The 5-year cumulative incidence of dementia in the full sample was 559/17302 (3.7%). Cumulative incidence of dementia was similar in UI and controls (3.5% versus 3.8%, p=0.26); mortality rate was higher in controls (16.4% versus 9.9%, p< 0.001).

UI was not associated with hazard of dementia in the adjusted model [HR 0.90 (0.76, 1.06), p=-0.22]. Incident dementia was associated with older age group, black race, dual Medicare/Medicaid, cardiac disease, diabetes, anxiety, and depression at index period (Table 2).

Conclusions: Among women 67 to 70 years old, those diagnosed with UI were healthier than age-matched controls at baseline and were not more likely to be diagnosed with incident dementia over 5 years. Future investigations of the UI association with dementia or cognitive function should control for the potential confounding effect of health status and anticholinergic medication use at baseline.