The Validity Of A Commercially-available, Low-cost Accelerometer In A Free-living Setting

Abstract

Wearable activity monitors are a popular tool for the measurement of physical activity in the consumer market. However, much of the existing evidence assessing the validity of these commercially-available activity monitors is laboratory-based. Because these monitors will typically be used by consumers outside of a laboratory setting additional research assessing the validity of wearable activity monitors in free-living environments is warranted. PURPOSE: To assess the ability of a low cost, wearable activity monitor marketed to consumer (Movband 3) to measure physical activity behavior in a free-living setting against a previously-validated, research-grade accelerometer (Actigraph GT1M). METHODS: Sixteen participants (n = five females, 27.0 ± 6.0 years old) were given both the commercially-available Movband (mounted on the wrist) and the research-grade Actigraph (mounted on the waist) to wear over the period of one week during free-living activity. Participants were required to simultaneously wear the two monitors for a minimum of five hours per day on five of the seven days. Participants completed daily diaries indicating the time of day in which the two monitors were worn. Physical activity data from each monitor was then divided by the duration of time participants wore the two monitors. Therefore, physical activity data for each monitor was expressed as activity counts/minute worn for each day. Pearson’s correlation analyses were then performed to assess the relationship between counts/minute of the two monitors for each participant individually across all days the two devices were worn for a minimum of five hours. These individual correlation coefficients were then averaged across all participants and presented as a single value. RESULTS: There was a large, positive effect size for the association (r = 0.57) between Movband and Actigraph activity counts/minute during free- living activity. In other words, as physical activity counts/minute increased in the Movband, activity also increased in the Actigraph. CONCLUSION: The low-cost, commercially- available wearable activity monitor (Movband) appears to provide a valid assessment of physical activity behavior in a free-living setting.