Domain-Specific Physical Activity and Its Association With Frailty Syndrome: Evidence From NHANES Using Phenotype and Index-Based Models

Abstract

Background: Frailty syndrome is a common age-related condition associated with increased risks of disability, hospitalization, and mortality. Although physical activity (PA) is a modifiable lifestyle factor, limited research has explored how specific PA domains relate to frailty using validated multidimensional models. This study aimed to examine the associations between 4 PA domains—total PA, occupational PA (OPA), transport-related PA (TPA), and leisure-time PA (LTPA)—and frailty syndrome, as assessed by both the frailty phenotype (FP) and frailty index (FI) models in a nationally representative U.S. sample. Methods: We analyzed data from 59 842 adults who participated in 6 cycles of the National Health and Nutrition Examination Survey (NHANES, 2007–2018). PA levels were self-reported and categorized by weekly duration based on guideline adherence (≥ 150 min/week) and distribution quartiles. Frailty was independently assessed using FP and FI models. Multivariable linear regression, subgroup, sensitivity analysis, and dose–response analyses were conducted. Results: Higher levels of total PA and LTPA were consistently associated with lower frailty scores in both FP and FI models. The strongest inverse associations were observed for LTPA ≥ 300 min/week. In contrast, OPA and TPA showed limited or no associations after adjustment for confounders. Subgroup and sensitivity analyses supported these findings, with particularly pronounced effects among older adults (≥ 80 years) and women. Conclusion: LTPA demonstrated the strongest associations with lower frailty, highlighting its potential as a key target for preventive strategies and public health guidelines aimed at promoting healthy aging.