Multidimensional adversity burden predicts functional deterioration and reduced maintenance of independence in...

Functional decline in middle-aged and older adults involves transitions among functional independence, mild limitation, and severe limitation. Prior studies have often used binary disability endpoints, leaving it unclear how multidimensional adversity burden relates to staged deterioration. Using the China Health and Retirement Longitudinal Study (CHARLS) as the discovery cohort and the English Longitudinal Study of Ageing (ELSA), Survey of Health, Ageing and Retirement in Europe (SHARE), and Mexican Health and Aging Study (MHAS) as validation cohorts, we examined whether baseline adversity burden was associated with subsequent functional-state deterioration and reduced maintenance of independence. Participants were aged ≥50 years. Records with missing age or age coding above 100 years were excluded. CHARLS used a 7-domain adversity score, whereas ELSA, SHARE, and MHAS used a harmonized 4-domain score. Functional status was classified using cohort-specific daily-function difficulty counts as independence, mild limitation, or severe limitation. Adjacent-wave person-interval data were analyzed with generalized estimating equations to estimate deterioration between consecutive waves. Robustness analyses included inverse probability weighting, multiple imputation, deterioration or death, baseline-independence restriction, exclusion of the first follow-up interval, leave-one-domain-out analyses, and a 4-cohort 1-stage pooled model. CHARLS baseline sample included 13,717 participants; 11,518 entered the complete-covariate transition model. Each additional adversity domain was associated with higher odds of functional-state deterioration (odds ratio [OR], 1.323; 95% confidence interval [CI], 1.293-1.352). The association was stronger among participants starting from independence (OR, 1.377) and remained present among those with mild limitation who experienced further deterioration (OR, 1.193). External validation showed consistent associations in ELSA (OR, 1.513), SHARE (OR, 1.373), and MHAS (OR, 1.349). The fixed-effect pooled OR was 1.402 (95% CI, 1.374-1.430), and the random-effects estimate was similar (OR, 1.409; 95% CI, 1.317-1.507). Maintenance of independence was lowest in the high-adversity group. Multidimensional adversity burden served as a longitudinal risk marker for functional deterioration and reduced maintenance of independence. These findings support risk stratification and identification of high-adversity groups for future integrated intervention studies, although further harmonized validation and prospective intervention evidence are needed before the score can be used as an intervention-guiding tool.