Step width variability as a discriminator of age-related gait changes

Background: Evidence indicates that adults over 65 demonstrate increased step width variability during walking, correlating with heightened fall risk. However, no established threshold levels exist to define optimal reference ranges for this metric. This study aimed to determine reference thresholds distinguishing normative from excessive step width variability in older adults. Methods: A systematic search of multiple databases through September 2018 identified studies measuring step width variability in both younger and older adults during self-selected walking speeds. Meta-analysis pooled standardized mean differences with 95% confidence intervals. Two threshold-setting approaches, the Youden index method and an uncertainty interval method, were used to identify optimal cutoff points. Results: Analysis of ten studies (304 older adults, 219 younger adults) revealed significantly higher step width variability in older participants (SMD=1.15, 95% CI=0.60-1.70; p=0.001). Single-decision thresholding established excessive variability at 2.14 cm. Two-decision thresholding identified values above 2.50 cm as excessive, while measurements below 1.97 cm represented the optimal reference range. Conclusion: Older adults demonstrate elevated step width variability compared to younger cohorts, and values exceeding 2.50 cm indicate excessive variability. These findings could guide rehabilitation technology development targeting lateral stability improvements during walking.