Hearing loss, social isolation, and frailty-state transitions in older adults: a longitudinal multistate analysis of CHARLS and HRS

Background Frailty is dynamic, but the relationships of hearing loss, social isolation, and their coexistence with transitions between frailty states remain unclear. We examined their associations with frailty progression, recovery, and death in two longitudinal ageing cohorts. Methods We included 4,068 China Health and Retirement Longitudinal Study (CHARLS) participants (mean baseline age, 72.0 years; baseline, 2011; median follow-up, 7.0 years) and 9,769 Health and Retirement Study (HRS) participants (mean baseline age, 75.4 years; baseline, 2010; median follow-up, 8.0 years). Questionnaire-based hearing loss and social isolation were defined using cohort-specific measures harmonized at the construct level. Frailty was categorized as non-frail, pre-frail, or frail using deficit-accumulation frailty indices, with death as an absorbing state. Interval-based complementary log-log models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for seven prespecified transitions, with participant-clustered robust standard errors. Results In CHARLS, hearing loss was associated with greater pre-frail-to-frail progression (HR 1.21, 95% CI 1.05–1.41) and lower pre-frail-to-non-frail recovery (HR 0.70, 95% CI 0.59–0.84). In HRS, hearing loss was associated with greater non-frail-to-pre-frail progression (HR 1.41, 95% CI 1.20–1.67) and pre-frail-to-frail progression (HR 1.21, 95% CI 1.10–1.34), and lower frail-to-pre-frail recovery (HR 0.83, 95% CI 0.74–0.94). Social isolation was associated with greater non-frail-to-pre-frail progression in CHARLS (HR 1.25, 95% CI 1.00–1.55); in HRS, it was associated with lower frail-to-pre-frail recovery (HR 0.84, 95% CI 0.76–0.93) and a greater pre-frail-to-death transition hazard (HR 1.34, 95% CI 1.18–1.52). Compared with neither exposure, both exposures were associated with greater non-frail-to-pre-frail progression in CHARLS (HR 1.67, 95% CI 1.23–2.28) and HRS (HR 1.47, 95% CI 1.19–1.82). Multiplicative interaction was not consistent across transitions or cohorts. Conclusions Hearing loss was most consistently associated with greater frailty progression and lower recovery. Coexisting hearing loss and social isolation identified a higher-risk profile for frailty progression, but the findings did not support consistent multiplicative interaction.