Citation

Gerdhem P, Ringsberg KA, Akesson K, Obrant KJ. J. Clin. Epidemiol. 2005; 58(3): 226-232.

Affiliation

Department of Orthopaedics, Malmo University Hospital, SE-205 02 Malmo, Sweden.

Copyright

(Copyright © 2005, Elsevier Publishing)

DOI

10.1016/j.jclinepi.2004.06.013

PMID

15718110

Abstract

OBJECTIVE: Fall risk assessment is important because the consequences, such as a fracture, may be devastating. The objective of this study was to find the test or tests that best predicted falls in a population-based sample of elderly women. STUDY DESIGN AND SETTING: The fall-predictive ability of a questionnaire, a subjective estimate of biologic age and objective functional tests (gait, balance [Romberg and sway test], thigh muscle strength, and visual acuity) were compared in 984 randomly selected women, all 75 years of age. RESULTS: A recalled fall was the most important predictor for future falls. Only recalled falls and intake of psycho-active drugs independently predicted future falls. Women with at least five of the most important fall predictors (previous falls, conditions affecting the balance, tendency to fall, intake of psychoactive medication, inability to stand on one leg, high biologic age) had an odds ratio of 11.27 (95% confidence interval 4.61-27.60) for a fall (sensitivity 70%, specificity 79%). CONCLUSION: The more time-consuming objective functional tests were of limited importance for fall prediction. A simple clinical history, the inability to stand on one leg, and a subjective estimate of biologic age were more important as part of the fall risk assessment.