Purpose: The aim of this study was to compare gait stability and variability between walking conditions and age groups.
Methods: Twenty-six healthy younger and older females participated. Trunk acceleration in the vertical (V), medial-lateral (ML) and anteriorposterior (AP) directions during 5 minutes walking overground and 3 minutes walking on the treadmill at self-selected speed were recorded. Root mean square and standard deviations of acceleration, stride time and its variability, Lyapunov exponents (LE), multiscale entropy (MSE) and harmonic ratios (HR) were computed.
Results: Both age groups showed significantly higher stride time variability and short-term LE in all directions during overground walking. For the older group, overground walking showed higher V and AP standard deviation. Significantly lower values for overground walking were observed for long-term LE (V and ML for the younger group, ML for the older group), HR (ML for the older group) and MSE (V for the older group). Significant age-related differences were found for V long-term LE for overground walking.
Conclusions: The present findings suggest that both linear and advanced computational techniques for gait stability and variability assessment in older adults are sensitive to walking conditions.