Dynamic Gait Stability, Clinical Correlates, and Prognosis of Falls Among Community-Dwelling Older Adults

Objective: To establish an accurate measure for prognostic assessment of fall risk in community-dwelling older adults, this study examined the prediction accuracy of a dynamic gait stability measure and common clinical tests for slip-related falls among these adults. Design: Participants were tested for their fall-risk likelihood on a slip-test. Setting: Biomechanics research laboratory. Participants: Community-dwelling older adults (N=119; >= 65y). Interventions: Not applicable. Main Outcome Measures: Participants performed a battery of clinical tests, including Berg Balance Scale, Timed Up & Go (TUG) test, static posturography, isometric muscle strength, and bone density. They were then exposed to an unannounced slip during gait. The dynamic stability during unperturbed gait was measured based on the center of mass position and velocity relative to the limits of stability against backward falling. Accuracy of each measure was examined for prediction of slip outcome (fall or recovery). Results: On the slip, 59 participants fell, 56 recovered their balance, and 4 were harness-assisted. Dynamic stability predicted fall outcome with 69% accuracy. Except for TUG and bone density, no other measure could differentiate fallers from nonfallers; TUG predicted 56% of fall outcomes. Conclusions: Reproduction of actual falls provides a new benchmark for evaluating the prognostic power of different performance-based assessment tools. The TUG was able to better predict fall outcome than other clinical measures; however, the new dynamic gait stability measure was more sensitive than TUG in its prediction of falls. Ultrasound bone scan could be used to screen older adults for fall risk.