Article

The purposes of this study were : (1) to develop and
validate a simple clinical scale to stratify risk for recurrent falls in
community-dwelling elderly people based on easily obtained social and clinical
items and (2) to evaluate the added value of 3 clinical balance tests in
predicting this risk. DESIGN: This was a prospective measurement study. METHODS:
A population of 1,618 community-dwelling people over 65 years of age underwent a
health checkup, including performance of 3 clinical balance tests: the
One-Leg-Balance Test, the Timed "Up & Go" Test, and the Five-Times-Sit-to-Stand
Test. Falls were recorded using a self-administered questionnaire that was
completed a mean (SD) of 25+/-5 months after the visit. Participants were
randomly divided into either group A (n=999), which was used to develop the
scale, or group B (n=619), which was used to prospectively validate the scale.
RESULTS: Logistic regression analysis identified 4 variables that independently
predicted recurrent falls in group A: history of falls, living alone, taking
>or=4 medications per day, and female sex. Thereafter, 3 risk categories of
recurrent falls (low, moderate, and high) were determined. Predicted probability
of recurrent falls increased from 4.1% to 30.1% between the first and third
categories. This scale subsequently was validated with great accuracy in group B.
Only the Five-Times-Sit-to-Stand Test provided added value in the estimation of
risk for recurrent falls, especially for the participants who were at moderate
risk, in whom failure on the test (duration of >15 seconds) doubled the risk.
LIMITATIONS: Falls were assessed only once, and length of follow-up was
heterogeneous (18-36 months). CONCLUSIONS: Clinicians could easily classify older
patients in low-, moderate-, or high-risk groups of recurrent falls by using 4
easy-to-obtain items. The Five-Times-Sit-to-Stand Test provides added value to
stratify risk for falls in people at moderate risk.

Langue : ANGLAIS