Potential utility of physical function measures to improve the risk prediction of functional disability in community-dwelling older Japanese adults: a prospective study

BACKGROUND: While gait speed, one-leg standing balance, and handgrip strength have been shown to be independent predictors for functional disability, it is unclear whether such simple measures of physical function contribute to improved risk prediction of functional disability in older adults. METHODS: A total of 1,591 adults aged ≥ 65 years and without functional disability at baseline were followed up for up to 7.9 years. Functional disability was identified using the database of Japan’s Long-term Care Insurance System. Maximum gait speed, one-leg standing time, and handgrip strength were measured at baseline. Cox proportional hazard models were used to estimate the hazard ratios (HRs) and 95 % confidence intervals (CIs) for the association of physical function and functional disability incidence. The incremental predictive value of each physical function measure for risk prediction was quantified using the difference in overall C-statistic, category-free net reclassification improvement (NRI), and integrated discrimination improvement (IDI) index. RESULTS: During follow-up (median: 7.8 years), functional disability was identified in 384 participants. All of the physical function measures were inversely associated with the risk of functional disability, independent of potential confounding factors. The multivariable adjusted HRs (95 % CIs) for functional disability per one standard deviation increment of maximum gait speed, one-leg-standing time, and hand grip strength were 0.73 (0.65–0.83), 0.68 (0.59–0.79), and 0.72 (0.59–0.86), respectively. Incorporation of each of maximum gait speed, one-leg-stand time, and hand grip strength into a basic model with other risk factors significantly improved C-statistic from 0.770 (95 % CIs, 0.751–0.794) to 0.778 (0.759–0.803), 0.782 (0.760–0.805), and 0.775 (0.756–0.800), respectively (all p < 0.05). A model including all three measures had the highest C-statistic of 0.787 (0.765–0.810). The improvements in risk prediction were also confirmed by category-free NRI and IDI index. CONCLUSIONS: Adding any of the three measures to a basic model with other known risk factors significantly improved the prediction of functional disability and addition of all three measures provided further improvement of the prediction in older Japanese adults. These data provide robust evidence to support the practical utility of incorporating these simple physical function measures into functional disability risk prediction tools. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12877-021-02415-3.