Wang Yu-Cheng, Huang Yi-Ran, Zhang Qinge, Zhang Ling, Feng Yuan, Fan Ping, Su Zhaohui, Cheung Teris, Li Jinghua, Ungvari Gabor S, Ng Chee H, Xiang Yu-Tao
Unit of Psychiatry, Department of Public Health and Medicinal Administration, & Institute of Translational Medicine, Faculty of Health Sciences, University of Macau, Macao SAR, China; Centre for Cognitive and Brain Sciences, University of Macau, Macao SAR, China.
Beijing Key Laboratory of Mental Disorders, National Clinical Research Center for Mental Disorders & National Center for Mental Disorders, Beijing Anding Hospital, Capital Medical University, Beijing, China.
J Affect Disord. 2026 Feb 15;395(Pt B):120750. doi: 10.1016/j.jad.2025.120750. Epub 2025 Nov 24.
The relationship between frailty and depressive symptoms in older adults remains unclear, particularly regarding its nonlinear associations. We investigated whether baseline frailty index (FI) predicts incident depressive symptoms in study cohorts from China and the United States.
Longitudinal data were analyzed from the China Health and Retirement Longitudinal Study (CHARLS, 2011-2020) and Health and Retirement Study in the USA (HRS, 2012-2020), which include 8727 adults aged ≥50 years without baseline depressive symptoms (HRS: n = 4386; CHARLS: n = 4341). FI was constructed from 29 health deficits. Depressive symptoms were assessed using the CES-D scales. The associations between baseline FI and incidence of depressive symptoms were examined with Cox proportional hazards regression and restricted cubic spline analyses. Stratified analyses tested sociodemographic and lifestyle modifications. Domain-specific risks were analyzed separately.
Within a median follow-up of 8.0 years in HRS and 9.0 years in CHARLS, 35.7 % and 55.1 % of participants developed depressive symptoms, respectively. Higher FI predicted increased risk (HRS: HR = 1.06, 95 % CI: 1.05-1.06; CHARLS: HR = 1.04, 95 % CI: 1.04-1.05). Nonlinear analyses revealed threshold at FI = 38.24 in HRS and FI = 17.18 in CHARLS, with stronger associations below thresholds (HR 1.08 and 1.09, respectively). Physical health conditions showed the strongest domain-specific association. In HRS, higher education and wealth strengthened the associations although no such gradients were found in CHARLS.
Frailty strongly predicted the incidence of depressive symptoms with significant threshold effects and cross-national differences. Early identification and intervention to address frailty could reduce the risk of late-life depressive symptoms.
