Association between urinary incontinence and self-reported hearing loss: findings from two national longitudinal aging studies

Hearing loss (HL) and urinary incontinence (UI) frequently co-occur in aging populations, yet longitudinal evidence on the association between UI and incident HL remains limited. This study examined this association using self-reported HL as the outcome measure across two culturally distinct aging cohorts. Retrospective cohort analysis using data from the English Longitudinal Study of Ageing (ELSA, 2004–2023) and the China Health and Retirement Longitudinal Study (CHARLS, 2013–2020). The included participants were aged ≥ 50 years without baseline HL (ELSA, n = 5,720; CHARLS, n = 987). The exposure was self-reported UI, and the outcome was incident self-reported HL. Multivariable Cox models were used to estimate hazard ratios (HRs), with stratified analyses by age and sex. Predictive nomograms were constructed and evaluated. After multivariable adjustment, UI was longitudinally associated with a higher incidence of self-reported HL in ELSA (HR = 1.182, 95% CI:1.041–1.342, P = 0.010); a similar direction of association was observed in CHARLS, albeit with wider confidence intervals (HR = 1.523, 95% CI:1.001–2.317, P = 0.049). In CHARLS, exploratory subgroup analyses suggested potentially stronger associations in participants aged < 70 (HR = 3.490, 95% CI:1.111–10.960, P = 0.032) and women (HR = 1.666, 95% CI:1.054–2.633, P = 0.029). The UI consistently ranked as a relevant associated feature in the feature importance analysis. The nomograms showed good calibration and positive net benefits. These findings suggest a potential role for UI screening in geriatric assessments to help identify individuals who may warrant prioritized hearing evaluation, though this approach requires further validation in prospective interventional studies and external cohorts.