Digital Exclusion and Frailty Among Older Adults: A Longitudinal Analysis From China

Abstract Background China is ageing rapidly while everyday services are migrating online, yet whether exclusion from the digital world predicts frailty in later life remains insufficiently quantified. Existing multicohort evidence relies on a single internet-use item, which may underestimate the true scope of exclusion. Methods We analysed the 2018 and 2020 waves of the China Health and Retirement Longitudinal Study (CHARLS). Community-dwelling adults aged 60 years or older with a calculable frailty index (FI) and complete exposure data were followed for two years. Digital engagement across four domains—internet use, smartphone use, mobile payment and online social activities—was scored 0–4, with full digital exclusion defined as no engagement in any domain. Frailty was defined as an FI ≥ 0.25. Modified Poisson regression estimated adjusted relative risks (RRs) of incident frailty among baseline non-frail participants, and linear regression estimated differences in annualised FI change. Results Among 10,216 participants (mean age 67.8 years; 52.4% women), 61.4% were fully digitally excluded. Over two years, 17.3% of baseline non-frail participants developed frailty. After full adjustment, full digital exclusion was associated with incident frailty (RR 1.42, 95% confidence interval [CI] 1.24–1.63) and with a faster annualised increase in the FI (adjusted difference 0.29 per 0.01 units, 95% CI 0.17–0.41). The risk declined monotonically with each additional engaged domain (P for trend < 0.001). Associations were strongest among rural residents (P for interaction = 0.04) and participants with primary education or less (P for interaction = 0.03). Conclusions Digital exclusion independently predicted incident frailty and accelerated deficit accumulation among older Chinese adults. Integrating digital-inclusion measures into healthy-ageing strategies, particularly for rural and less-educated populations, warrants consideration.