Detection of walking periods and number of steps in older adults and patients with Parkinson's disease: accuracy of a pedometer and an accelerometry-based method

B. Dijkstra, W. Zijlstra, E.J.A. Scherder, Y.P.T. Kamsma

Research output: Contribution to JournalArticleAcademicpeer-review

180 Downloads (Pure)

Abstract

The aim of this study was to examine if walking periods and number of steps can accurately be detected by a single small body-fixed device in older adults and patients with Parkinson's disease (PD). Results of an accelerometry-based method (DynaPort MicroMod) and a pedometer (Yamax Digi-Walker SW-200) worn on each hip were evaluated against video observation. Twenty older adults and 32 PD patients walked straight-line trajectories at different speeds, of different lengths and while doing secondary tasks in an indoor hallway. Accuracy of the instruments was expressed as absolute percentage error (older adults versus PD patients). Based on the video observation, a total of 236.8 min of gait duration and 24,713 steps were assessed. The DynaPort method predominantly overestimated gait duration (10.7 versus 11.1%) and underestimated the number of steps (7.4 versus 6.9%). Accuracy decreased significantly as walking distance decreased. Number of steps were also mainly underestimated by the pedometers, the left Yamax (6.8 versus 11.1%) being more accurate than the right Yamax (11.1 versus 16.3%). Step counting of both pedometers was significantly less accurate for short trajectories (3 or 5 m) and as walking pace decreased. It is concluded that the Yamax pedometer can be reliably used for this study population when walking at sufficiently high gait speeds (>1.0 m/s). The accelerometry-based method is less speed-dependent and proved to be more appropriate in the PD patients for walking trajectories of 5 m or more. Copyright © The Author 2008. Published by Oxford University Press on behalf of the British Geriatrics Society.
Original languageEnglish
Pages (from-to)436-441
Number of pages6
JournalAge and Ageing
Volume37
Issue number4
DOIs
Publication statusPublished - 2008

Fingerprint

Dive into the research topics of 'Detection of walking periods and number of steps in older adults and patients with Parkinson's disease: accuracy of a pedometer and an accelerometry-based method'. Together they form a unique fingerprint.

Cite this