Gait variability and stability measures: Minimum number of strides and within-session reliability
Introduction
Ageing and pathology can worsen gait performance at multiple levels and in selective ways [1], and quantitative assessment of gait pattern has been proven to be useful in the early identification and prediction of pathology or cognitive decline [2], [3], [4]. In particular, trunk acceleration-based measures of gait variability and stability are proposed in the literature aiming at quantifying subject specific gait characteristics such as gait impairment, degree of neuro-motor control and balance disorders, in pathologic and healthy subjects, and are often related to fall at risk [5], [6], [7], [8], [9]. However, no standard implementation procedure for these measures is defined, potentially explaining the incoherent conclusions [10], as implementation differences can affect outputs. Thus, a standardization of the implementation parameters is necessary to perform a consistent evaluation. Moreover, these measures must reproduce the same results in the same experimental conditions.
Many strides can be required to obtain reliable measures, but treadmill walking differs significantly from over-ground walking [11]; hence, long walking trials have to be analyzed. The use of Inertial Measurement Units (IMU) allows to obtain both stride time variability and stability measures from trunk acceleration signals during long over-ground outdoor walking trials.
In order to further define implementation features for future effective exploitation of measures in research or clinical practice, an assessment of the repeatability of variability/stability measures is hence needed, together with an assessment of the number of necessary strides. The aim of the present study was to assess the minimum number of required strides and the within-session reliability of 11 temporal variability/stability measures proposed in the literature and applied to stride time and trunk accelerations during over-ground walking.
Section snippets
Methods
Ten healthy participants [28±3 years, 174±11 cm, 67±13 kg] walked in a straight line at self-selected natural speed on a 250 m long dead-end road (about 180 strides), wearing two synchronized tri-axial IMU (Opal, APDM, USA) located on the trunk (at the level of the fifth lumbar vertebra) and on the right shank. Sample size was chosen in agreement with previous literature [12]. Range of accelerometers was ±6 G and sampling rate 128 Hz. Right heel strike instants were obtained from the angular
Results
Measures reached steady values for different numbers of strides, depending on the threshold. For MSE V (τ=1, …, 4) and RQA (AP rr, det, avg, ML rr and V rr, det, avg), 10 strides were sufficient to reach a 10% threshold. MSE (AP, ML, V τ=5,6), RQA (ML det, avg) and sLE V reached a 20% threshold within 10 strides. Other measures showed lower stride number requirement with the increasing of the threshold. lLE required a high number of strides (>110) even for the 50% threshold. RQA (V max, diverg)
Discussion
While variability measures aim at quantifying the degree of variability in the stride time, stability measures directly quantify stability (maxFM, sLE, lLE) or stability-related properties of gait time-series, such as recurrence (RQA), complexity at different scales (MSE), smoothness (HR) or harmonicity (IH). Since no standard implementation procedure is defined for these measures, the aim of this study was to investigate the required minimum number of strides and the within-session reliability
Conflict of interest statement
None declared.
Acknowledgment
This research was funded by the Project “Fall risk estimation and prevention in the elderly using a quantitative multifactorial approach” (Project ID number 2010R277FT) awarded by the Italian Ministry of Education, University and Research (Ministero dell’Istruzione, dell’Università e della Ricerca).
References (31)
- et al.
Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force
J. Nutr. Health Aging
(2009) - et al.
Acceleration patterns of the head and pelvis when walking on level and irregular surfaces
Gait Posture
(2003) - et al.
Estimating fall risk with inertial sensors using gait stability measures that do not require step detection
Gait Posture
(2013) - et al.
Gait variability and fall risk in community-living older adults: a 1-year prospective study
Arch. Phys. Med. Rehabil.
(2001) - et al.
Orbital stability analysis in biomechanics: a systematic review of a nonlinear technique to detect instability of motor tasks
Gait Posture
(2013) - et al.
A kinematic comparison of overground and treadmill walking
Clin. Biomech.
(1998) - et al.
Spatio-temporal parameters of gait measured by an ambulatory system using miniature gyroscopes
J. Biomech.
(2002) - et al.
Recurrence quantification analysis of gait in normal and hypovestibular subjects
Gait Posture
(2012) - et al.
Multiscale entropy analysis of human gait dynamics
Phys. A: Stat. Mech. Appl.
(2003) - et al.
Pelvis–thorax coordination in the transverse plane during gait
Gait Posture
(2002)