Comparison of a Wearable Tracker with Actigraph for Classifying Physical Activity Intensity and Heart Rate in Children

• Main Authors: Seoungki Kang, Youngdeok Kim, Wonwoo Byun, JinSu Suk, Jung-Min Lee
• Format: Article
• Language: English
• Published: MDPI AG 2019-07-01
• Series: International Journal of Environmental Research and Public Health
• Subjects: accelerometer; Fitbit; physical activity classification
• Online Access: https://www.mdpi.com/1660-4601/16/15/2663

<i>Introduction:</i> To examine the validity and reliability of the Fitbit Charge HR (FCH), wrist-worn ActiGraph (AG) accelerometers were used for assessing the classification of physical activity (PA) into intensity categories in children. <i>Methods:</i> Forty-three children (n = 43) participated in the study. Each participant completed 3 min bouts of 12 PAs ranging from sedentary to vigorous intensity while simultaneously wearing FCH and AG on both hands, a Polar HR monitor, and a portable indirect calorimeter. Total time spent in different PA intensity levels measured by FCH and AG were compared to the indirect calorimetry. <i>Results:</i> The highest classification accuracy values of sedentary behavior was 81.1% for FCH. The highest classification (72.4%) of light intensity PA was observed with Crouter&#8217;s algorithm from the non-dominant wrist. Crouter&#8217;s algorithm also show the highest classification (81.8%) for assessing moderate to vigorous intensity PA compared to FCH (70.8%). Across the devices, a high degree of reliability was found in step measurements, ranging from an intra-class correlation (ICC) = 0.92 to an ICC = 0.94. The reliability of the AG and the FCH showed high agreement for each variable. <i>Conclusion:</i> The FCH shows better validity for estimating sedentary behavior and similar validity for assessing moderate to vigorous PA compared to the research-grade monitor. Across the devices, the reliability showed the strongest association.