[A standardised physical performance test versus the frailty phenotype as predictors of adverse events in octogenarians].

BACKGROUND

The population of adults aged 80 years and older is heterogenous with some being robust and others having a higher risk for adverse events. This study compares the predictive value of two tools used to identify older adults who are at higher risk for adverse outcomes: frailty phenotype according to Fried and a standardized physical performance test.

METHODS

The BELFRAIL population-based cohort of 567 community-dwelling adults aged 80 years and older living in Belgium. Fried frailty phenotype and physical performance test (gait, chair stand, standing balance tests and putting on and off a cardigan). The predictive value of the two tools in predicting mortality (up to 5.1 ± 0.25 years), hospitalization (3.0 ± 0.25 years) and decline in activities of daily living (after 1.7 ± 0.21 years) was compared using reclassification statistics and decision curve analysis.

RESULTS

Frail participants according to Fried phenotype and those in the lowest quartile of the physical performance test score had higher risk for mortality and hospitalization. Harrell C and area under operator curve were similar (< 0.70). Reclassification statistics and net benefit in decision curve analysis showed no significant difference between the two tools in identifying higher risk for mortality, hospitalization and functional decline.

CONCLUSION

In a cohort of community-dwelling adults 80 years and older a standardized physical performance test was as good as the Fried frailty phenotype in identifying higher risk for adverse outcomes.