Predicting Falls in Community-Dwelling Older Adults Using the Repetitive Step Test

Background/Purpose: The Repetitive Step Test (RST) is a newly developed dynamic balance test that requires stepping in multiple directions. It has previously demonstrated excellent test-retest and inter-rater reliability, along with ability to retrospectively classify individuals who had fallen more than once in 12 months though its ability to predict falls has not been evaluated. This study aimed to determine the capability of the RST to identify individuals likely to fall prospectively over 12 months. Methods: 63 community-dwelling adults \u3e65 years old were tested using the following balance and mobility measures: RST, 10 Meter Walk Test (10MWT), Activities-specific Balance Confidence scale (ABC), Four Square Step Test (FSST), and Timed Up and Go (TUG). For the RST, participants completed as many steps as possible in 15 seconds at 50% and 75% of their right leg length in the anterolateral and posterolateral directions bilaterally. Participants completed a fall log that was reported on a monthly basis for 12 months. Results: Based on 12 months of data, receiver operating characteristic (ROC) curve analysis was used to compare adults who fell \u3e 2 times to adults (n=17) who did not fall (n=29). Area under the curve (AUC) results with 95% confidence interval are as follows: RST at 50% 0.40 (0.22-0.58), RST at 75% 0.41 (0.24-0.57), FSST 0.45 (0.28-0.62), TUG 0.49 (0.32-0.66), ABC 0.61 (0.44-0.78) and 10MWT 0.67 (0.51-0.83). Conclusion: At 12 months, the RST and other balance and mobility measures either failed to or poorly predicted the number of falls. Research is needed to determine how to best predict falls in this population, and whether any single balance test can effectively identify individuals who are most likely to fall. Clinical Relevance: Currently, there is no simple measure of balance and mobility that can accurately predict falls in community-dwelling older adults. The RST was ineffective in predicting falls at 12 months. Given the unique characteristics of the RST, it may still be a useful tool for assessing and monitoring bilateral lower extremity performance and stepping ability, especially in individuals who present with asymmetrical impairments (e.g., stroke, multiple sclerosis, Parkinson’s disease).https://ecommons.udayton.edu/dpt_symposium/1005/thumbnail.jp