Difference scores between single-task and dual-task gait measures are better than clinical measures for detection of fall-risk in community-dwelling older adults

Background: As the proportion of older adults in the population increases, so does the associated prevalence of falls, making falls the leading cause of fatal and nonfatal injuries among adults aged >= 65 years. In response, researchers and clinicians seek to develop a clinical tool that accurately predicts fall risk. These Investigations have included measures of clinical mobility and balance tests, strength, physiologically based tests, postural sway, and mean and variability of gait measures. To date, no study has concurrently explored all these measures to determine which measures, alone or in combination, emerge as the most predictive of fall risk. While there is evidence that dual-task gait conditions are sensitive indicators of fall risk, difference scores between dual-task and single-task gait conditions (DS) have not been explored. Research question: This study included outcome measures representing diverse domains (clinical mobility and balance, strength, physiological, postural sway, and mean and variability of difference scores between dual-and single-task gait conditions) to determine the combination of measures that were the most sensitive for retrospectively classifying fallers from non-fallers. Methods: Forty-two (mean: 75.8 yrs +/- 3.3) community-dwelling older adults completed a comprehensive battery of 76 measures and classified into two groups based on self-report of having one or more falls in the previous year. Results: Results suggest that 11 measures captured the salient characteristics of the total cohort (fallers (N=27) and non-fallers (N=15) and that five gait measures were sufficient for correctly classifying fallers and non-fallers with 92.3% sensitivity and 66.7% specificity with a total model classification of 82.9%. Significance: The five variables comprise mean DS of stride timing, stride width, and stride length and DS in variability for stride width and stride velocity demonstrating that difference in performance between dual-task and single-task gait trials was essential for discriminating fallers and superior to other measures.