Utility of biological aging markers for mortality risk stratification in community-dwelling older adults: Insights from the Mr. OS & Ms. OS (Hong Kong) cohort.

Abstract

BACKGROUND Studies regarding the integration of various biological aging (BA) markers and their predictive values for long-term mortality risk remain limited, particularly in Asian older adults. We evaluated the associations of multiple BA markers with overall and cause-specific mortality over a 20-year period. METHODS Data on 4000 older adults (mean age: 72.5 years) were obtained from the Mr. OS and Ms. OS cohort. BA was assessed by the frailty phenotype, clinical deficit-based frailty index (FI), biochemical-enhanced FI (eGFR, homocysteine, hsCRP, 25(OH)-D), and leukocyte telomere length. Mortality was ascertained by the Hong Kong Death Registry. Hazard ratios (HRs) were assessed using Cox and Fine-Gray models and predictive accuracy with Harrell's c-index. RESULTS Over a median follow-up of 18.25 years, 2446 deaths occurred (CVD: 511, cancer: 644). For overall mortality, the HRs (95%CI) of pre-frail and frail groups were 1.24 (1.13-1.35) and 1.66 (1.39-1.98) compared to the fit. Each standard deviation (SD) increase in the clinical or biochemical-enhanced FI was associated with 22% or 23% increased risk of overall mortality (P < 0.001). Longer telomere length reduced overall mortality risk (HR per SD: 0.93, 95%CI: 0.88-0.99). All BA markers except telomere length were significantly associated with CVD mortality. While all BA markers were not significantly associated with cancer mortality. Frailty-related markers showed added predictive values for both overall and cause-specific mortality, while telomere length was specifically predictive for CVD mortality (C-index improvements: 0.32%-7.90%). CONCLUSIONS Biological aging is associated with overall and CVD-cause mortality in older Chinese and may support risk stratification and personalized interventions.