Citation

Nofuji Y, Shinkai S, Taniguchi Y, Amano H, Nishi M, Murayama H, Fujiwara Y, Suzuki T. J. Am. Med. Dir. Assoc. 2015; 17(2): 184.e1-1847.

Affiliation

National Center for Geriatrics and Gerontology, Obu, Japan.

Copyright

(Copyright © 2015, Lippincott Williams and Wilkins)

DOI

10.1016/j.jamda.2015.11.003

PMID

26717805

Abstract

OBJECTIVES: Walking speed, grip strength, and standing balance are key components of physical performance in older people. The present study aimed to evaluate (1) associations of these physical performance measures with cause-specific mortality, (2) independent associations of individual physical performance measures with mortality, and (3) the added value of combined use of the 3 physical performance measures in predicting all-cause and cause-specific mortality.

DESIGN: Prospective cohort study with a follow-up of 10.5 years. SETTING: Tokyo Metropolitan Institute of Gerontology Longitudinal Interdisciplinary Study on Aging (TMIG-LISA), Japan. PARTICIPANTS: A total of 1085 initially nondisabled older Japanese aged 65 to 89 years. MEASUREMENTS: Usual walking speed, grip strength, and standing balance were measured at baseline survey.

RESULTS: During follow-up, 324 deaths occurred (122 of cardiovascular disease, 75 of cancer, 115 of other causes, and 12 of unknown causes). All 3 physical performance measures were significantly associated with all-cause, cardiovascular, and other-cause mortality, but not with cancer mortality, independent of potential confounders. When all 3 physical performance measures were simultaneously entered into the model, each was significantly independently associated with all-cause and cardiovascular mortality. The C statistics for all-cause and cardiovascular mortality were significantly increased by adding grip strength and standing balance to walking speed (P < .01), and the net reclassification improvement for them was estimated at 18.7% and 7.5%, respectively.

CONCLUSION: Slow walking speed, weak grip strength, and poor standing balance predicted all-cause, cardiovascular, and other-cause mortality, but not cancer mortality, independent of covariates. Moreover, these 3 components of physical performance were independently associated with all-cause and cardiovascular mortality and their combined use increased prognostic power.

Language: en