Longitudinal Sex-Specific Physical Function Trends by Age, Race/Ethnicity, and Weight Status

BACKGROUND: There is a need to assess longitudinal trends in physical function and to understand the prevalence of physical function status in the older population using objectively measured physical function. Therefore, the primary purpose of this study was to examine longitudinal sex-specific physical function trends and prevalence of physical function status by age, race/ethnicity, and weight status in older adults, aged 65 years and older. METHODS: The present study used longitudinal data from the National Health and Aging Trends Study. Physical function was measured annually from 2011 to 2018 using the Short Physical Performance Battery, and classified as poor (0–6), fair (7–9), or good (10–12). Mixed effect regression and generalized estimating equation models were used to estimate unadjusted trends and annual changes in physical function, adjusting for confounders and accounting for complex sampling. RESULTS: Respondents\u27 physical function declined over time regardless of sex, age, race/ethnicity, and weight status (P \u3c.001). Respondents aged 75 to 84 years (males: β = −.18; 95% confidence interval (CI) = −.22 to −.13; females: β = −.23; 95% CI = −.27 to −.20) and 85 years and older (males: β = −.39; 95% CI = −.47 to −.30; females: β = −.33; 95% CI = −.39 to −.28) had a greater annual physical function rate of decline than respondents aged 65 to 74 years. Respondents aged 75 years and older also had a greater annual increase in the proportion classified as having poor physical function than respondents aged 65 to 74 years (P \u3c.001), regardless of sex. Compared with non-Hispanic whites, Hispanic females had a lower annual physical function rate of decline (β =.14; 95% CI = 0.07–0.20), and Hispanic and/or non-Hispanic blacks had slower annual changes in the proportion classified as having good physical function (P \u3c.05). Compared with normal weight males, overweight males had lower annual changes in physical function (P =.003) and proportion of those with poor physical function (P =.004), but no differences between weight status were observed in females. CONCLUSIONS: Physical function declines with age, but temporal changes in physical function and physical function status varied by sex, age, race/ethnicity, and weight status. J Am Geriatr Soc 68:2270–2278, 2020