Population aging is accompanied by increasing heterogeneity in health trajectories and mortality risk. We aimed to identify and compare independent predictors of 3-year and 5-year all-cause mortality in Taiwanese adults aged ≥50 years, integrating intrinsic capacity (IC), functional status, laboratory biomarkers, and genetic factors within a unified multidomain framework.

This study used data from the Social Environment and Biomarkers of Aging Study (SEBAS, n = 772), a population-based cohort embedded within the Taiwan Longitudinal Study of Aging. Assessments covered five IC domains, functional status, 13 comorbidities, a comprehensive biomarker panel (inflammatory, metabolic, neuroendocrine, hematologic, nutritional), and genetic markers (ApoE ε4, 5-HTTLPR). Mortality was ascertained through national death registry linkage. Predictors were identified by stepwise selection and multivariable Cox proportional hazards models.

The mean age of participants was 65.3 years (SD 9.3), and 55.7% were men. Distinct predictor profiles emerged for near-term (3-year) and long-term (5-year) mortality. Near-term mortality was dominated by markers of acute vulnerability: low grip strength (HR 4.98, 95% CI 1.88–13.20), depressive symptoms (HR 3.42, 95% CI 1.34–8.71), and proinflammatory biomarkers including interleukin-6 and neutrophil percentage while serum albumin was protective (HR 0.13, 95% CI 0.04–0.42). Long-term mortality reflected a broader accumulation of multisystem burden: low grip strength remained predictive (HR 2.57, 95% CI 1.43–4.62), but was joined by metabolic and renal markers (HbA1c and creatinine), and immune aging indices (monocyte and lymphocyte percentages), whereas higher Instrumental Activities of Daily Living scores were protective (HR 0.78 per unit increase, 95% CI 0.65–0.94).

Mortality risk in aging adults is multidimensional and time-dependent. Near-term mortality was driven by acute physiologic vulnerability, such as nutritional depletion, muscular reserve loss, and active inflammation, whereas long-term mortality reflected cumulative metabolic, renal, and immune-aging burden. These findings supported time-horizon–specific, multidomain risk stratification in aging populations.