Answers are grounded in this paper and cite their sources.
Cohort study explores the link between functional dispersion and disability-free survival, suggesting new risk indicators.
BACKGROUND: Physical, cognitive, and psychological functions contribute to disability risk and are interrelated. This study explored the association between multidomain functional dispersion and disability-free survival in community-dwelling older adults. METHODS: This cohort study used NCGG-SGS data. Functional dispersion was the within-person SD across standardized physical, cognitive, and psychological domain scores; higher values indicated greater between-domain differences. Disability-free survival was time to first certification under the Japanese Long-Term Care Insurance system. Adjusted cause-specific Cox models with restricted cubic splines examined dispersion and incident disability, including mean functional score. Robustness was assessed using bootstrap, Fine-Gray, and log-transformed sensitivity analyses. RESULTS: Among 13,746 participants (baseline 2011-2017), 2,135 developed disability during 5 years (15.5%). Dispersion was non-linearly associated with disability risk (P for non-linearity <0.0001). Using the cohort median dispersion as the reference, adjusted hazard ratios were 0.94 (95% CI, 0.90-0.98) at the 25th percentile and 1.11 (95% CI, 1.03-1.20) at the 75th percentile. Below the median, hazard ratios decreased as dispersion declined, but rose again at the extreme lower tail. Above the median, risk rose modestly and then declined at higher dispersion. Bootstrap and Fine-Gray analyses supported these patterns. CONCLUSIONS: Multidomain functional dispersion showed a non-linear association with disability risk. Multidomain functional dispersion may be useful as an exploratory indicator of heterogeneity in disability risk beyond overall functional level.
No takes yet. Share an insight, caveat, or question.
Morikawa et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: