Bidirectional transitions of frailty states in middle-aged and older adults: Findings from two longitudinal cohorts studies

OBJECTIVE: We aimed to characterize frailty dynamics in older adults by applying multi-state Markov models (MSMs) to English Longitudinal Study of Ageing (ELSA) and the China Health and Retirement Longitudinal Study (CHARLS) data to estimate transition intensities. METHODS: Data from ELSA (2004-2012) and CHARLS (2011-2018) cohorts were analyzed using MSMs. Frailty was assessed using the Frailty Index (FI), and covariates like age, sex, marital status, physical activity, and economic status were included. Transition probabilities between frailty states were estimated, and subgroup analyses were performed to assess the influence of covariates on frailty transitions. RESULTS: A total of 8 187 participants from ELSA, and 9 171 participants from CHARLS were included in the analysis. In ELSA, the transition intensity from pre-frail to robust (0.226, 95 % CI 0.217-0.234) exceeded that from pre-frail to frail (0.105, 95 % CI 0.099-0.111), whereas in CHARLS a different pattern was observed. In the covariate analysis, we found that age, marital status, loneliness, physical activity, current drinking status, and economic status exerted both common and cohort-specific effects on frailty transitions in ELSA and CHARLS, with age and marital status influencing bidirectional frailty transitions in both cohorts. CONCLUSION: Frailty is reversible, especially in the pre-frail stage. Transition intensities and influencing factors vary across older adult cohorts in different countries, and beyond age, marital support, physical activity quality, and economic status significantly shape frailty trajectories and should be incorporated into public health policy design.