Research Article: Does variability of footfall kinematics correlate with dynamic stability of the centre of mass during walking?

Date Published: May 31, 2019

Publisher: Public Library of Science

Author(s): Niklas König Ignasiak, Deepak K. Ravi, Stefan Orter, Seyyed Hamed Hosseini Nasab, William R. Taylor, Navrag B. Singh, Eric R. Anson.


A stable walking pattern is presumably essential to avoid falls. Stability of walking is most accurately determined by the short-term local dynamic stability (maximum Lyapunov exponent) of the body centre of mass. In many studies related to fall risk, however, variability of step width is considered to be indicative of the stability of the centre of mass during walking. However, other footfall parameters, in particular variability of stride time, have also been associated with increased risk for falling. Therefore, the aim of this study was to investigate the association between short-term local dynamic stability of the body centre of mass and different measures of footfall variability. Twenty subjects performed unperturbed walking trials on a treadmill and under increased (addition of 40% body weight) and decreased (harness system) demands to stabilise the body centre of mass. Association between stability of the centre of mass and footfall parameters was established using a structural equation model. Walking with additional body weight lead to greater instability of the centre of mass and increased stride time variability, however had no effect on step width variability. Supported walking in the harness system did not increase centre of mass stability further, however, led to a significant decrease of step width and increase in stride time variability. A structural equation model could only predict 8% of the variance of the centre of mass stability after variability of step width, stride time and stride length were included. A model which included only step width variability as exogenous variable, failed to predict centre of mass stability. Because of the failure to predict centre of mass stability in this study, it appears, that the stability of the centre of mass is controlled by more complex interaction of sagittal and frontal plane temporal and spatial footfall parameters, than those observed by standard variability measures. Anyway, this study does not support the application of step width variability as indicator for medio-lateral stability of the centre of mass during walking.

Partial Text

An efficient walking pattern is characterised by a variety of distinct gait domains, such as pace, rhythm, symmetry, variability and balance [1, 2]. With age, as well as in subjects with neuromotor deficits, all or some of these domains are perturbed, which ultimately results in an increased risk of falling [3–5]. While measures to quantify variability during walking are known to be sensitive in the discrimination of faller and non-faller subjects, their power to estimate fall risk on an individual basis prior to a first fall, remains unclear [3, 6–8]. It is generally assumed that the movement of the centre of mass (CoM) during walking is maintained (returned back to a steady-state after a perturbation) by effectively negotiating the placement of our feet, formally described as the base of support (BoS), and provides the primary means for stabilizing the system [9–12].

Prior to study participation each of the 20 healthy subjects (10 males and 10 females) provided written informed consent. The mean (SD) age, height and body mass of the participants was 27.0 (4.2) years, 175.7 (8.9) cm, and 71.6 (10.9) kg respectively. None of the subjects presented any form of musculoskeletal or neurological disorder or pain. The entire study protocol was approved by the local ethics committee (ETH Zurich Ethikkommission) and was conducted in accordance with the Declaration of Helsinki.

The aim of this study was to investigate the association between footfall kinematics and the stability of the CoM iteratively over a few specific steps as well as over the entire duration of a walking trial. Our findings provide insights on the role of foot placement in stability of the CoM, as well as whether summary spatio-temporal variability (particularly step width variability) measures are indicators of balance control. Interestingly we found that XCoM can be predicted (up to about 68%) using the step width and length from the current and 4 preceding steps. However, variability of step width and stride time independently, but also in combination summarized over an entire walking trial, were not able to predict dynamic stability of the CoM as indicated by LyE. A combination of step width, stride time and step length variability only predicted 8% of CoM stability variance, and hence questions the application of footfall kinematic variability as a surrogate measure for balance control. Finally, our analysis revealed that the association between step-width (and length) and XCoM was lowest in the 40% weight-reduction condition, while the largest LyE was observed for the 40% weight-addition condition.