Examination of a Wearable Activity Tracker to Assess Children’s Physical Activity.

Abstract

Newer wearable activity monitors have the capability to measure heart rate (HR) from the user’s wrist using an optical blood flow sensor (i.e., photoplethysmography techniques), and none of the previous studies examining children’s free-living activity utilizing a newly developed wearable activity tracker which includes built-in HR sensor. PURPOSE: To examine PA intensity using the wearable HR monitor. METHODS: a total of 43 children (girls = 18, boys = 25), aged 8 - 12 years, participated in the study. Participants were asked to wear the Fitbit Charge HR (FHR) on their left (FHRL), right wrist (FHRR) and Polar HR (PHR) monitor on their chest, while completing 11 different activities lasted 48 mins, monitoring their HR. Oxygen consumption was measured throughout the routine with the Cosmed K4B2 metabolic analyzer. 10 min average values of the resting metabolic rate were used as one metabolic equivalent of task (MET) to categorize children’s PA intensity. Activity intensities using HR from the FHR were classified by calculating children’s target HR between 50 and 75% of their maximum HR as moderate intensity activity and greater than 75% of their maximum HR as a vigorous intensity activity. An estimate of children’s maximum age-related HR was obtained from the 208 - 0.7 (age) equation. Each intensity classified by the MET determined by HR was compared to the intensity classified by the MET determined by resting metabolic rate (i.e., criterion measure) to examine the measurement agreement. McNemar’s test was used to examine the measurement agreement for paired intensity frequency. RESULTS: After setting PA intensity classification with 50% and 75% of HR, the frequency determined by measured MET from the Cosmed K4B2 were 1002, 407, and 120 for light, moderate, and vigorous intensity. The frequency determined by HR monitor and FHR were as follow: 1076, 394, and 58 by PHR, 1393, 127, and 9 by FHRL, 1393, 129, and 7 by FHRR. The values of weighted Kappa statistics from the McNemar’s test were 0.40 (95% CL: 0.36-0.44) for the intensity by PHR, 0.21 (95% CL: 0.18-0.25) for the intensity by FHRL, 0.19 (95% CL: 0.16-0.23) for the intensity by FHRR. CONCLUSIONS: The agreement of PA intensity classified by the wearable activity tracker to the intensity classified the metabolic analyzer showed fair agreement.