How should the margin of stability during walking be expressed to account for body size?

When expressing the margin of stability as a distance, it does not directly estimate the perturbation magnitude needed to change stability states. Additionally, it is unknown how body size may influence this measure. Therefore, we propose other expressions of stability margins, including that of an impulse, a change in center of mass velocity, and a scaled, unitless impulse. The purpose of this study was to determine the influence of body size on these margin expressions using walking data from children and adults. We anticipated that margins expressed as an impulse would have strong correlations with body mass and height, as well as large between-group differences. We predicted that scaling this impulse value would result in small correlations and between-group effect sizes. We calculated each stability margin at minimum lateral values and in the anteroposterior directions at mid-swing and foot strike. In the lateral direction, margins expressed as an impulse had strong correlations with body size (r≥0.58, p<0.01) and large between-group differences (|d|≥1.07, p<0.01). The other expressions did not have strong positive correlations (|r|≤0.20) or large between-group effects (|d|≤0.44). In the anteroposterior directions, impulse margins had strong correlations with body size (|r|≥0.83, p<0.01) and large between-group differences (|d|≥1.74, p<0.01). The scaled, unitless impulse margin was the only variable that resulted in small, non-significant differences (|r|≤0.22, p≥0.24) as well as small between-group effect sizes (|d|≤0.46, p≥0.22). We propose expressing stability margins as an impulse. If scaling is needed, we encourage using the scaled, unitless impulse.