Physical performance and 13.5-year mortality in elderly women

Abstract

Aims: To examine whether four modifiable physical performance-based measurements predicted 13.5-year mortality in 300 elderly women. Methods: This prospective observational cohort study followed 300 randomly-selected community-dwelling women, with a mean age of 80.9 years at baseline, for 13.5 years. Baseline physical performance measurements were: comfortable walking speed, Timed Up and Go (TUG), functional reach, and one-leg stance. Demographic and health measurements were recorded. Date of death was recorded from the national population registry. Cox proportional hazard models were fitted, with time from baseline assessment to death or censored date as the dependent variable. Results: At follow-up, 71% of the participants had died. All performance-based measurements except one-leg stance were significantly associated with 13.5-year mortality. Only walking speed remained significantly independently associated with mortality when the three significant performance measurements were included in the same multiple proportional hazard Cox model. Conclusions: Comfortable walking speed, TUG and functional reach predicted 13.5-year mortality, and walking speed was the strongest predictor of mortality. Screening elderly women with performance measurements may aid in identifying those at risk and thus targeted interventions may be applied.