Full length articleStepWatch accuracy during walking, running, and intermittent activities
Introduction
Step counting has been performed for several centuries. Historically, it was used to estimate walking distances (based on an assumed stride length). More recently, however, it has been used to assess ambulatory physical activity in free-living humans. Tryon (1991) has noted that the step can be viewed as a preferred metric for physical activity assessment because it is a natural unit of ambulation. Step counting is an accepted measure of functional status of individuals, particularly at the lower end of the physical activity continuum where walking is the main activity performed and where physical activity questionnaires are less sensitive. For instance, increases in steps per day are viewed as a sign of improvement during the rehabilitation process and they are a valuable metric for discriminating physical activity levels in older adults [1], [2], [3]. Additionally, in healthy adult populations, step counters can serve to motivate clients and facilitate behavior change when coupled with a physical activity goal [4], [5]. However, these consumer-oriented step counters need to be compared against a more rigorously validated research grade device.
In previous studies, waist-worn pedometers have been found to be reasonably accurate for moderate or self-selected walking speeds (e.g. 80 m min−1). However, few studies have reported step count accuracy during running. Two studies report reasonable accuracy for running speeds including 160–215 m min−1, but accuracy at faster speeds and during intermittent activities is largely unknown [6], [7], [8]. Considering the slow speeds involved in intermittent activities of daily living (e.g. cleaning counter tops, vacuuming, and dusting) and high speeds involved in sport pursuits (e.g. tennis and basketball) a device that can accurately tally steps over a wide range of speeds is needed for assessing true daily step counts.
The StepWatch 3 (SW3; Modus Health, Inc., Washington, DC) is an ankle-mounted step counter originally developed for use in individuals with impaired functional status [1]. The SW3 is the most accurate pedometer ever constructed for walking, capturing 98% to 100% of all steps taken from 27 to 107 m min−1 [9]. Hickey et al. (2015) have shown the SW3 is also quite accurate for intermittent household activities, capturing 90% of steps taken while dusting and 102% of steps taken while cleaning a room. However, a serious limitation of this device is that it only captures 68% of steps during running at 161 m min−1 [7]. Other researchers have taken the approach of individualized calibration, adjusting the cadence and sensitivity settings until the device yielded results within 3% of directly observed steps [10], [11]. Using this approach, they were able to obtain accurate SW3 results for both walking and running in young children. However, no set rules for practical application were developed.
Clinicians and researchers can program the SW3 to account for the individual user’s step characteristics and gait speed. Specifically, the cadence and sensitivity settings preprogrammed into the device can be altered to account for these differences. The cadence setting is the length of time (cadence settings x 0.01 s) after a step is taken during which a subsequent step cannot be counted and sensitivity setting is the threshold acceleration that must be exceeded to register a step [12]. Determining the appropriate cadence and sensitivity settings for different speeds might improve the accuracy of the SW3 during running, while at the same time preserving its high accuracy for walking and intermittent lifestyle activities. Improving the accuracy of the SW3 in able-bodied adults could allow it to be used as a criterion for validating other step counters.
Thus, the purpose of this study was to examine the accuracy of the SW3 for determining steps taken by normal, healthy adults during walking, running, and intermittent lifestyle activities. In Part 1, the impact of the cadence and sensitivity settings on step counting error was explored for treadmill walking/running speeds ranging from 26.8 to 268 m min−1, in order to enhance the accuracy. In Part 2, the step count accuracy during intermittent, lifestyle activities was then investigated using the preprogrammed “quick start” settings and the modified settings developed in Part 1.
Section snippets
Participants
Twenty-five participants were recruited from the local community. Fifteen participants performed Part A, and five of them returned to complete Part B. Ten additional participants were recruited for Part B.
Individuals who were unable to run at 268 m min−1, who had lower limb injury or dysfunction, or who were diagnosed with cardiovascular disease were excluded from this study. The research protocol was approved by the local Institutional Review Board (IRB). All participants signed an informed
Results
A total of 25 participants with a mean (SD) age of 26 (8) years, height of 1.73 (0.10) m, body mass of 74.4 (13.8) kg, and body mass index of 24.7 (2.9) kg/m2 took part in this study. In Part 1, results from the one-way ANOVAs detected significant differences within all cadence and sensitivity settings at each speed (p < 0.05). Post hoc testing with Bonferroni adjustments revealed significant differences for several of the cadence and sensitivity settings used at each speed. Table 3, Table 4
Discussion
The main objective of this study was to explore the effects of the SW3 cadence and sensitivity settings on step count accuracy. This was done in order to develop modified settings that would allow for improved accuracy across a wide range of ambulatory speeds. A secondary objective was to compare the accuracy of the modified settings vs. the default settings, during intermittent, lifestyle activities.
To explore the accuracy of the SW3 settings at running speeds, the “Easy Start” selections in
Conflict of interest statement
All authors of this manuscript declare that they have no competing interests or conflicts of interest.
Funding
No financial support was received for the development and preparation of this manuscript.
Authors’ contributions
“Each of the authors has read and concurs with the content in the final manuscript. The material within has not been and will not be submitted for publication elsewhere except as an abstract.”
References (16)
- et al.
Pedometer accuracy in slow walking older adults
Int. J. Ther. Rehabil.
(2012) - et al.
Use of a step activity monitor in determining outcomes
JPO: J. Prosthet. Orthot.
(2006) - et al.
Pedometer accuracy in nursing home and community-dwelling older adults
Med. Sci. Sports Exerc.
(2004) - et al.
Using pedometers to increase physical activity and improve health: a systematic review
JAMA
(2007) - et al.
Wearable devices as facilitators, not drivers, of health behavior change
JAMA
(2015) - et al.
Validity and reliability of the Omron HJ-303 tri-axial accelerometer-based pedometer
J. Phys. Act. Health
(2011) - et al.
Validity of activity monitor step detection is related to movement patterns
J. Phys. Act. Health
(2015) - et al.
Validation of the Kenz Lifecorder EX and ActiGraph GT1 M accelerometers for walking and running in adults
Appl. Physiol. Nutr. Metab.
(2008)
Cited by (9)
Walking for health during pregnancy: A literature review and considerations for future research
2019, Journal of Sport and Health ScienceCitation Excerpt :Other activity monitor locations, including the thigh and ankle, have shown promise for assessment of PA and walking in nonpregnant populations and at slow speeds, but have not yet been tested in pregnant women.95,96 For example, although the ankle-worn StepWatch pedometer has not been validated in pregnant women, it has been used by Kong et al.23 to track walking during pregnancy and has also shown high accuracy for assessing free-living steps in nonpregnant populations,97,98 making it a potentially attractive option for assessing pregnancy PA. Additionally, activity monitors placed on the wrist show moderate or high validity for tracking steps taken in nonpregnant populations, but accuracy appears lower for tracking energy expenditure.99
Comparative Analysis of ActiGraph Step Counting Methods in Adults: A Systematic Literature Review and Meta-Analysis
2024, Medicine and Science in Sports and ExerciseStep Monitor Accuracy during Poststroke Physical Therapy and Simulated Activities
2022, Translational Journal of the American College of Sports MedicinePhysical Activity Monitor Accuracy for Overground Walking and Free-Living Conditions Among Pregnant Women
2020, Journal for the Measurement of Physical BehaviourValidity of smartphones and activity trackers to measure steps in a free-living setting over three consecutive days
2020, Physiological MeasurementDominant vs. Non-dominant wrist placement of activity monitors: Impact on steps per day
2019, Journal for the Measurement of Physical Behaviour