Publication Date

Summer 2016

Advisor(s) - Committee Chair

Jill Maples (Director), Mark Schafer, Cody Morris, and Rachel Tinius

Degree Program

Department of Kinesiology, Recreation and Sport

Degree Type

Master of Science


Commercially available fitness trackers have been found to accurately measure steps and caloric expenditure during walking and running activities. Circuit-style, highintensity functional training (HIFT) has become increasingly popular because it is inexpensive and effective in improving muscular strength and cardiovascular fitness. PURPOSE: To evaluate the accuracy of five accelerometers (ActiGraph GT3X, Nike Fuelband, Fitbit One, Fitbit Charge HR, and Jawbone UP Move) in estimating energy expenditure while performing an acute bout of HIFT. METHODS: Participants (n = 47) underwent baseline testing and at least 48 hours later, each participant completed the main test: a 15-minute workout consisting of 12 repetitions each of 7 different exercises; performed circuit-style by completing as many rounds as possible. During the main test, each participant wore the Cosmed K4b2 portable metabolic analyzer (PMA) and five different accelerometers. RESULTS: Four of the five fitness trackers reported lower (p <0.01) total caloric expenditure values compared to the PMA during the acute bout of HIFT. The waist-mounted device (ActiGraph, 182.55 ± 37.93 kcals) most closely mimicked caloric expenditure compared to the PMA (Cosmed, 144.99 ± 37.13 kcals) as indicated by an insignificant p value (0.056). Systematic differences between the activity monitors were calculated using an Intraclass Correlation (ICC) with an ICC = -0.032. The ICC of F (46,235) = 0.812 (p = 0.799) was not significant at the predetermined 0.05

alpha level. A Repeated Measures ANOVA showed that when compared to the Cosmed, all activity monitors were significantly different at the 0.05 alpha level. The Fitbit One and the Fitbit Charge HR were the only two activity monitors that are not significantly different from one another (p = 0.985). The range of error based on mean absolute percentage errors (MAPE) was lowest for the ActiGraph (15.1%) and highest for the Fitbit Charge HR (22.1%). CONCLUSION: The wrist- and hip-mounted fitness trackers do not accurately assess energy expenditure during HIFT exercise. Supported by: WKU Graduate School, NIGMS 2P20 GM103436-14; Institutional Development Award (IDeA) from National Institute of General Medical Sciences, National Institutes of Health, 5P20GM103436 and the WKU RCAP Grant 14-8007.


Exercise Science | Sports Sciences