Disability transitions and life expectancy: a national longitudinal study on social isolation and loneliness in China

BACKGROUND: Functional disability among older adults represents a dynamic process characterized by bidirectional transitions between states rather than a static condition. Social isolation and loneliness have emerged as potentially modifiable determinants of functional outcomes, yet evidence specifically examining their effects on disability state transitions remains limited. OBJECTIVE: To examine the independent and interactive associations of social isolation and loneliness with disability state transitions and life expectancy among Chinese adults aged ≥45 years. METHODS: We analyzed data from 20,954 adults aged ≥45 years in the China Health and Retirement Longitudinal Study (2011-2020). Continuous-time multi-state Markov models examined transitions among four states: no disability, mild disability, severe disability, and death. Life expectancy measures were estimated through numerical integration. Four progressively adjusted models evaluated independent and interactive effects. RESULTS: Social isolation significantly increased risks of transitioning from no disability to mild disability (hazard ratio [HR]=1.19, 95% confidence interval [CI]: 1.06-1.33) and from severe disability to death (HR = 1.33, 95% CI: 1.14-1.54), while reducing recovery likelihood from mild disability (HR = 0.88, 95% CI: 0.78-1.00). Loneliness independently elevated mild disability onset risk (HR = 1.16, 95% CI: 1.06-1.27) and decreased recovery probability (HR = 0.89, 95% CI: 0.81-0.97). Their interaction increased mild-to-severe disability deterioration risk (HR = 1.56, 95% CI: 1.13-2.15). At age 45, social isolation reduced total life expectancy by 3.18 years and loneliness by 2.13 years, primarily through compression of disability-free years. CONCLUSIONS: Social isolation and loneliness synergistically influence disability trajectories by affecting both worsening and recovery transitions, with disproportionate impacts on vulnerable subpopulations.