Article

OBJECTIVES: To determine whether a series of voluntary postural sway tasks could
differentiate and accurately identify the falls-history status of older adults,
and to examine the relations between voluntary sway measures and falls risk.
DESIGN: Case-control study. SETTING: University biomechanics laboratory.
PARTICIPANTS: Healthy community-dwelling older adults (N=51) aged 65 to 94 years
who were divided into nonfaller (n=36), single faller (n=10), and multiple faller
(n=5) groups based on a 12-month history of falls. INTERVENTIONS: Not applicable.
MAIN OUTCOME MEASURES: Participants underwent a falls-risk assessment using the
Physiological Profile Assessment (PPA) and then performed 6 voluntary postural
sway tasks. The tasks included maximum static leans, maximum voluntary sway,
continuous voluntary sway, rapid initiation of voluntary sway, rapid termination
of voluntary sway, and rapid orthogonal switches of voluntary sway between the
anterior-posterior and medial-lateral directions. Center of pressure amplitudes
and reaction time measures were examined using analysis of covariance, Pearson's
correlation, and discriminant function analyses. RESULTS: Multiple fallers had
increased age; increased falls risk; slower initiation, termination, and
orthogonal switch reaction times; and reduced center of pressure amplitude during
sway initiation and continuous voluntary sway compared with nonfallers. Few
differences were observed between the nonfallers and single fallers. Voluntary
sway measures were significantly correlated with each other and with PPA score.
Two postural reaction time measures and age identified 80% of multiple fallers
and 98% of nonmultiple fallers. Similarly, PPA score and age identified 80% of
multiple fallers and 100% of nonmultiple fallers. CONCLUSIONS: The slower and
less effective balance responses of multiple fallers compared with nonfallers and
the comparable sensitivity and specificity of PPA score and reactive voluntary
sway measures indicate that postural reaction time is a strong determinant of
falls risk.

Langue : ANGLAIS