Accuracy of uploadable pedometers in laboratory, overground, and free-living conditions in young and older adults

Ann M Swartz,Nora E Miller,Elizabeth K Lenz,Christopher J Dondzila,Scott J Strath

doi:10.1186/1479-5868-9-143

Abstract

PurposeThe purpose of this study was to examine the accuracy of uploadable pedometers to accurately count steps during treadmill (TM) and overground (OG) walking, and during a 24 hour monitoring period (24 hr) under free living conditions in young and older adults.MethodsOne hundred and two participants (n=53 aged 20–49 yrs; n=49 aged 50–80 yrs) completed a TM protocol (53.6, 67.0, 80.4, 93.8, and 107.2 m/min, five minutes for each speed) and an OG walking protocol (self-determined “< normal”, “normal”, and “> normal” walking speeds) while wearing two waist-mounted uploadable pedometers (Omron HJ-720ITC [OM] and Kenz Lifecorder EX [LC]). Actual steps were manually tallied by a researcher. During the 24 hr period, participants wore a New Lifestyles-1000 (NL) pedometer (standard of care) attached to a belt at waist level over the midline of the left thigh, in addition to the LC on the belt over the midline of the right thigh. The following day, the same procedure was conducted, replacing the LC with the OM. One-sample t-tests were performed to compare measured and manually tallied steps during the TM and OG protocols, and between steps quantified by the NL with that of the OM and LC during the 24 hr period. Mean error step scores (MES, criterion – device) and 95% Limits of Agreement (LoA) were calculated.ResultsThere were no significant differences between the OM and tallied steps for any of the TM speeds for either the young or older adult groups. The LC significantly underestimated steps for the young adult group during the 53.6 m/min TM speed (MES 31.4 [14.5, 48.3]) and during the OG < normal walking speed (MES 12.0 [0.9, 23.1] (p<0.01 for both age groups). The LC also significantly underestimated steps for the older adult group during the TM speeds of 53.6 m/min (MES 64.5 [45.6, 83.4]), 67.0 m/min (MES 15.1 [6.1, 24.0]), and 80.4 m/min (MES 3.2 [0.6, 5.9]) (p<0.01 for all speeds), in addition to the OG < normal walking speed (MES 14.7 [−13.3, 42.6] (p<0.01). The OM reported significantly lower steps during the 24 hr period for the young adult group by 949.1 steps (t=6.111, p<0.025) and for the older adult group by 612.9 steps (t=2.397, p<0.025).ConclusionBoth the OM and LC pedometers were more accurate as TM and OG walking speed increased. The OM significantly underestimated steps during the 24 hr compared with a standard of care evaluation. Overall, both uploadable pedometers appear acceptable to use in young or old age groups to measure walking behavior.

Highlights

Regular physical activity has long been shown to be host to a variety of benefits related to chronic conditions and diseases, such as diabetes, obesity, hypertension, and heart disease across a variety of populations [1]
Uploadable pedometers aim to further expand on the potential to increase and maintain activity habits to users by offering additional information and features. Such features could include individualized feedback and progress updates on daily walking behaviors, setting visual walking targets and how users compare to such targets, offering behavioral feedback cues based upon uploaded walking behaviors, all of which can be insightful during interventional purposes [5]
Participants in the 20–49 year category were marginally taller, and had a slightly longer stride length compared to the older age group

Summary

Introduction

Regular physical activity has long been shown to be host to a variety of benefits related to chronic conditions and diseases, such as diabetes, obesity, hypertension, and heart disease across a variety of populations [1]. As the technology of these devices has advanced, newer pedometer models are able to connect to a computerized interface, such as a desktop computer, allowing ambulatory physical activity behavior to be uploaded. Uploadable pedometers aim to further expand on the potential to increase and maintain activity habits to users by offering additional information and features. Such features could include individualized feedback and progress updates on daily walking behaviors, setting visual walking targets and how users compare to such targets, offering behavioral feedback cues based upon uploaded walking behaviors, all of which can be insightful during interventional purposes [5]

Objectives

Methods

Results

Discussion

Conclusion