Age-Based Sex and Race and Ethnicity Percentiles for Physical Functioning Among a Diverse Population From the HABS-HD...

Physical functioning (PF) is critical to quality of life and longevity among older adults and is related to biomarkers of Alzheimer's disease, cardiovascular and respiratory disease, and functional health. Physical functioning varies by age, sex, and race and ethnicity, yet its distribution across these groups is not well characterized. This study aimed to (1) characterize age-PF patterns by sex and race and ethnicity and (2) generate sex-, race and ethnicity-, and age-specific percentile distributions for PF.

Cross-sectional data from the Health & Aging Brain Study-Health Disparities (HABS-HD; n = 6712; ages 50-92 years) were analyzed. Physical functioning was assessed using Timed Up and Go (TUG) time (seconds) and Short Physical Performance Battery (SPPB) total time (seconds) among participants able to complete the tests. Adjusted regression models assessed associations between age and PF. Unadjusted percentiles (5th-95th) within sex × race and ethnicity (non-Hispanic White [NHW], non-Hispanic Black [NHB], and Hispanic) × age categories (50-64, 65-69, and ≥70 years) were reported among the full sample, low-comorbidity participants (no cognitive impairment, diabetes, or assistive device use), and comorbidity-free participants (additionally free of hypertension and dyslipidemia).

Among participants (62.6% female; mean age 66.5 ± 8.7 years), physical function declined with age, with longer TUG and SPPB times across all groups. Among all participants, the mean (SD) time (seconds) for TUG and SPPB were 10.0 (2.9) and 10.6 (2.2), respectively. Among low-comorbidity participants, TUG 95th percentiles (seconds) ranged from 10.93 (NHW females, 50-64 years) to 17.12 (NHB females, ≥70 years). The SPPB 95th percentiles (seconds) ranged from 17.90 (NHW males, 50-64 years) to 34.64 (Hispanic males, ≥70 years). Sensitivity analyses indicated that comorbidities explained a substantial proportion, but not all, of the observed age, sex, and race and ethnicity disparities in PF.

These values provide a descriptive framework for clinicians and researchers to contextualize PF and monitor change over time. Future studies should aim to establish valid percentile prognostic cut-points for various health outcomes.