- Prevalence: functional dependency 22.96%, cognitive impairment 16.04%, depressive symptoms 41.70%, co-occurrence 4.03% among older Chinese adults.
- "3Ds index" shows a dose-response association with all-cause mortality: HRs 1.31, 1.86, 2.16 for scores 1, 2, 3 respectively (P for trend <0.001).
- "3Ds index" is a leading mortality predictor, partly mediated by self-rated health (9.77%–12.37%), and exceeds cardiometabolic multimorbidity.
Zhonghua Yi Xue Za Zhi. 2026 Aug 18;106(30):3164-3173. doi: 10.3760/cma.j.cn112137-20260123-00271.
ABSTRACT
Objective: To describe the epidemiological characteristics of functional dependency, cognitive impairment, and depressive symptoms (the “3Ds”: Dependency, Dementia, and Depression) among older adults in China, and to evaluate the association of “3Ds” co-occurrence and index with the risk of all-cause mortality. Methods: Based on the 2018 and 2021 waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS), adults aged≥65 years were included. To characterize prevalence, the baseline sample was classified into five groups: no “3Ds”(intact physical function, cognition, and mental health), dependency only (functional dependency alone), cognitive impairment only, depressive symptoms only, and co-occurrence of all three. To evaluate the association with mortality, participants were grouped by cumulative “3Ds index” score as 0 (no deficit, reference), 1, 2, or 3 (presence of any one, any two, or all three deficits, respectively). Cox proportional hazards models assessed the effects of the “3Ds” and cardiometabolic multimorbidity (CMM) on all-cause mortality; a Random Survival Forest (RSF) model validated variable importance (VIMP); and causal mediation analysis examined the mediating role of self-rated health. Results: Among 13 682 participants at baseline in 2018 [mean age (84.74±11.39) years], the prevalence of functional dependency, cognitive impairment, and depressive symptoms was 22.96% (3 142/13 682), 16.04% (2 194/13 682), and 41.70% (5 705/13 682), respectively, and the co-occurrence rate of all three was 4.03% (552/13 682). In the mortality analysis of 11 909 participants [(median follow-up 28 (1-40) months, 2 872 deaths], multivariable Cox regression showed that, compared with participants scoring 0 on the “3Ds index”, the risk of all-cause mortality rose significantly across scores of 1 to 3: HR=1.31 (95%CI: 1.19-1.44) for a score of 1, HR=1.86 (95%CI: 1.66-2.09) for 2, and HR=2.16 (95%CI: 1.83-2.55) for 3, indicating a significant linear dose-response relationship (P for trend<0.001); by contrast, CMM was not significantly associated with all-cause mortality (HR=1.02, 95%CI: 0.91-1.13). The RSF model identified the “3Ds index” as a leading predictor of all-cause mortality (VIMP=0.104), second only to age. Self-rated health significantly and partially mediated the “3Ds”-mortality association, accounting for 9.77%-12.37% of the total effect. Conclusions: Co-occurrence of the “3Ds” is prevalent among Chinese older adults, particularly the oldest-old, and shows significant demographic heterogeneity. The “3Ds index”-reflecting impairment in physical function, cognition, and mental health-shows a significant dose-response association with all-cause mortality that is partially mediated by self-rated health, and may serve as a simple composite indicator for early identification of mortality risk in the oldest-old.
PMID:42595511 | DOI:10.3760/cma.j.cn112137-20260123-00271
Share Evidence Blueprint
Save to Google Notes

Search Google Scholar
⭐ My Revision List

