Original Research ARTICLE
Less is more – estimation of the number of strides required to assess gait variability in spatially confined settings
- 1Department of Neurology, Charité Medical University of Berlin, Germany
- 2Department of Neurology, Universitätsklinikum Schleswig-Holstein, Germany
- 3Department of Neurodegenerative Diseases, Hertie-Institut für klinische Hirnforschung (HIH), Germany
- 4NeuroCure, Charité Universitätsmedizin Berlin, Germany
- 5Experimental and Clinical Research Center, Max Delbrück Center for Molecular Medicine, Helmholtz Association of German Research Centers (HZ), Germany
- 6Berliner Schule für Geist und Gehirn, Humboldt-Universität zu Berlin, Germany
Gait variability is an established marker of gait function that can be assessed using sensor-based approaches. In clinical settings, spatial constraints and patient condition impede the execution of longer distance walks for the recording of gait parameters. Turning paradigms are often used to overcome these constraints and commercial gait analysis systems algorithmically exclude turns for gait parameters calculations. We investigated the effect of turns in sensor-based assessment of gait variability.
Continuous recordings from 31 patients with movement disorders (ataxia, essential tremor and Parkinson’s disease) and 162 healthy elderly (HE) performing level walks including 180° turns were obtained using an inertial sensor system. Accuracy of the manufacturer’s algorithm of turn-detection was verified by plotting stride time series. Strides before and after turn events were extracted and compared to respective average of all strides. Coefficient of Variation (CoV) of stride length and stride time was calculated for entire set of strides, segments between turns and as cumulative values. Their variance and congruency was used to estimate the number of strides required to reliably assess the magnitude of stride variability.
Non-detection of turns in 5.8% of HE lead to falsely-increased CoV for these individuals. Even after exclusion of these, strides before/after turns tended to be spatially shorter and temporally longer in all groups, contributing to an increase of CoV at group level and widening of confidence margins with increasing numbers of strides. This could be attenuated by a more generous turn excision as an alternative approach. Correlation analyses revealed excellent consistency for CoVs after at most 20 strides in all groups.Respective stride counts were even lower in patients using a more generous turn excision.
Including turns to increase continuous walking distance in spatially confined settings does not necessarily improve the validity and reliability of gait variability measures. Specifically with gait pathology, perturbations of stride characteristics before/after algorithmically excised turns were observed that may increase gait variability with this paradigm. We conclude that shorter distance walks of around 15 strides suffice for reliable and valid recordings of gait variability in the groups studied here.
Keywords: Gait Variability, gait analysis, Healthy elderly, turn detection, Movement Disorders
Received: 11 May 2018;
Accepted: 20 Dec 2018.
Edited by:P Hemachandra Reddy, Texas Tech University Health Sciences Center, United States
Reviewed by:Richard Camicioli, University of Alberta, Canada
Arnaud Gouelle, ProtoKinetics LLC, United States
Copyright: © 2018 Kroneberg, Elshehabi, Meyer, Otte, Doss, Paul, Nussbaum, Berg, Kühn, Maetzler and Schmitz-Hübsch. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: MD. Tanja Schmitz-Hübsch, NeuroCure, Charité Universitätsmedizin Berlin, Berlin, 10117, Berlin, Germany, email@example.com