Yang Bai1, Gregory J Welk, Yoon Ho Nam, Joey A Lee, Jung-Min Lee, Youngwon Kim, Nathan F Meier, Philip M Dixon. 1. 1Department of Kinesiology, Iowa State University, Ames, IA; 2School of Health, Physical Education, and Recreation, University of Nebraska at Omaha, Omaha, NE; 3MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Cambridge, UNITED KINGDOM; and 4Department of Statistics, Iowa State University, Ames, IA.
Abstract
PURPOSE: This study evaluated the relative validity of different consumer and research activity monitors during semistructured periods of sedentary activity, aerobic exercise, and resistance exercise. METHODS: Fifty-two (28 male and 24 female) participants age 18-65 yr performed 20 min of self-selected sedentary activity, 25 min of aerobic exercise, and 25 min of resistance exercise, with 5 min of rest between each activity. Each participant wore five wrist-worn consumer monitors [Fitbit Flex, Jawbone Up24, Misfit Shine (MS), Nike+ Fuelband SE (NFS), and Polar Loop] and two research monitors [ActiGraph GT3X+ on the waist and BodyMedia Core (BMC) on the arm] while being concurrently monitored with Oxycon Mobile (OM), a portable metabolic measuring system. Energy expenditure (EE) on different activity sessions was measured by OM and estimated by all monitors. RESULTS: Mean absolute percent error (MAPE) values for the full 80-min protocol ranged from 15.3% (BMC) to 30.4% (MS). EE estimates from ActiGraph GT3X+ were found to be equivalent to those from OM (± 10% equivalence zone, 285.1-348.5). Correlations between OM and the various monitors were generally high (ranged between 0.71 and 0.90). Three monitors had MAPE values lower than 20% for sedentary activity: BMC (15.7%), MS (18.2%), and NFS (20.0%). Two monitors had MAPE values lower than 20% for aerobic exercise: BMC (17.2%) and NFS (18.5%). None of the monitors had MAPE values lower than 25% for resistance exercise. CONCLUSION: Overall, the research monitors and Fitbit Flex, Jawbone Up24, and NFS provided reasonably accurate total EE estimates at the individual level. However, larger error was evident for individual activities, especially resistance exercise. Further research is needed to examine these monitors across various activities and intensities as well as under real-world conditions.
PURPOSE: This study evaluated the relative validity of different consumer and research activity monitors during semistructured periods of sedentary activity, aerobic exercise, and resistance exercise. METHODS: Fifty-two (28 male and 24 female) participants age 18-65 yr performed 20 min of self-selected sedentary activity, 25 min of aerobic exercise, and 25 min of resistance exercise, with 5 min of rest between each activity. Each participant wore five wrist-worn consumer monitors [Fitbit Flex, Jawbone Up24, Misfit Shine (MS), Nike+ Fuelband SE (NFS), and Polar Loop] and two research monitors [ActiGraph GT3X+ on the waist and BodyMedia Core (BMC) on the arm] while being concurrently monitored with Oxycon Mobile (OM), a portable metabolic measuring system. Energy expenditure (EE) on different activity sessions was measured by OM and estimated by all monitors. RESULTS: Mean absolute percent error (MAPE) values for the full 80-min protocol ranged from 15.3% (BMC) to 30.4% (MS). EE estimates from ActiGraph GT3X+ were found to be equivalent to those from OM (± 10% equivalence zone, 285.1-348.5). Correlations between OM and the various monitors were generally high (ranged between 0.71 and 0.90). Three monitors had MAPE values lower than 20% for sedentary activity: BMC (15.7%), MS (18.2%), and NFS (20.0%). Two monitors had MAPE values lower than 20% for aerobic exercise: BMC (17.2%) and NFS (18.5%). None of the monitors had MAPE values lower than 25% for resistance exercise. CONCLUSION: Overall, the research monitors and Fitbit Flex, Jawbone Up24, and NFS provided reasonably accurate total EE estimates at the individual level. However, larger error was evident for individual activities, especially resistance exercise. Further research is needed to examine these monitors across various activities and intensities as well as under real-world conditions.
Authors: Samuel R LaMunion; Andrew L Blythe; Paul R Hibbing; Andrew S Kaplan; Brandon J Clendenin; Scott E Crouter Journal: Appl Physiol Nutr Metab Date: 2019-07-03 Impact factor: 2.665
Authors: Yue Liao; Michael C Robertson; Andrea Winne; Ivan H C Wu; Thuan A Le; Diwakar D Balachandran; Karen M Basen-Engquist Journal: Transl Behav Med Date: 2021-03-16 Impact factor: 3.046
Authors: Stephen P Wright; Tyish S Hall Brown; Scott R Collier; Kathryn Sandberg Journal: Am J Physiol Regul Integr Comp Physiol Date: 2017-01-04 Impact factor: 3.619
Authors: Philip M Dixon; Pedro F Saint-Maurice; Youngwon Kim; Paul Hibbing; Yang Bai; Gregory J Welk Journal: Med Sci Sports Exerc Date: 2018-04 Impact factor: 5.411
Authors: Felipe Lobelo; Heval M Kelli; Sheri Chernetsky Tejedor; Michael Pratt; Michael V McConnell; Seth S Martin; Gregory J Welk Journal: Prog Cardiovasc Dis Date: 2016-02-26 Impact factor: 8.194