We thank Aguiar, Moore, Ducharme, and McCullough (1) for commending our effort to evaluate the accuracy of seven wearable step counters against direct observation of video recorded steps in the free-living environment (2). We agree that a discussion of the cost/practicality trade-off and development of error tolerances for free-living studies is warranted.
We agree that lower-cost, consumer-grade step counters could be “good enough” for most consumers, behavior change interventions, and possibly physical activity surveillance. For carefully designed studies investigating the health implications of stepping, greater accuracy and precision are desired. The results of our study show that the consumer-grade step counters captured 75.3% to 83.5% of criterion steps which is similar to some of the expensive research-grade devices (i.e., activPAL, 76.9%; ActiGraph moving average vector magnitude (MAVM) algorithm on the wrist, 83.7%) (2). Additionally, the consumer devices have similar levels of accuracy compared to the activPAL and ActiGraph MAVM at the wrist. Furthermore, consumer-grade devices have been shown to detect changes in step counts between low and high active persons in free-living settings (3) implying that these devices can track changes in step counts during physical activity interventions. However, these findings are limited to a healthy population, and caution should be applied with slow walking (≤54 m·min−1 because this results in decreased step count accuracy for several devices (i.e., Yamax Digi-Walker, Fitbit Flex) (4,5). Therefore, device accuracy should be investigated in older adults and clinical populations.
For studying associations between steps per day and health, we advocate for counting every step because there is increasing evidence that suggests light physical activity confers health benefits (6). It is reasonable to hypothesize that light physical activities include brief, intermittent, and slow walking which is typically undercounted by many devices (4,5). Thus, our knowledge of the health implications related to steps is limited. Currently, the only device that does a reasonably good job of counting all steps is the StepWatch, which is worn on the ankle. We agree that its price and wear location, compared to consumer devices, make it an unlikely choice for large cohort studies. However, its superior accuracy should be considered a benchmark for developing step counting algorithms.
Because of the influx of consumer-grade devices used in scientific research, error tolerances must be developed for the free-living environment. Because it is unrealistic to expect consumer devices to estimate steps in the free-living environment with the same degree of accuracy as in treadmill conditions, we agree that acceptable error should be greater for 24-h data. Development of this standard should be based on the consensus of experts in the field, similar to how the Consumer Technology Association developed protocols and standards for laboratory-based studies (7).
Step-based physical activity recommendations for the general public should be based on the accurate assessment of all steps that contribute to health benefits. Future research regarding health benefits associated with stepping would fulfill the gap in the literature suggested by the 2018 Physical Activity Guidelines Advisory Committee Scientific Report (8).
Lindsay P. Toth
David R. Bassett Jr.
Department of Kinesiology, Recreation, and
The University of Tennessee
Salt Lake City, UT
1. Aguiar EJ, Moore CC, Ducharme SW, McCullough AK. Validation of step counters in the free-living context. Med Sci Sports Exerc
2. Toth LP, Park S, Springer CM, Feyerabend MD, Steeves JA, Bassett DR. Video-recorded validation of wearable step counters under free-living conditions. Med Sci Sports Exerc
3. Mendoza AR, Staudenmayer J, Freedson P. A consumer activity tracker is sensitive in detecting change in free-living energy expenditure and steps. Med Sci Sports Exerc
4. Chen MD, Kuo CC, Pellegrini CA, Hsu MJ. Accuracy of wristband activity monitors during ambulation and activities. Med Sci Sports Exerc
5. Hickey A, John D, Sasaki J, Mavilia M, Freedson P. Validity of activity monitor step detection is related to movement patterns. J Phys Act Health
6. Füzéki E, Engeroff T, Banzer W. Health benefits of light-intensity physical activity: A systematic review of accelerometer data of the national health and nutrition examination survey (NHANES). Sports Med
7. Consumer Technology Association. Physical Activity Monitoring for Fitness Wearables: Step Counting. Arlington, VA: Consumer Technology Association Technology & Standards Department; 2016.
8. 2018 Physical Activity Guidelines Advisory Committee. 2018 Physical Activity Guidelines Advisory Committee Scientific Report
. Washington, DC: U.S. Department of Health and Human Services; 2018.