Background: The purpose of this study was to examine the accuracy of heart rate (HR) and energy expenditure (EE) estimated by the Apple Watch Series 1 worn both on the wrist and the upper arm. Methods: Thirty healthy, young adults (15 females) wore the two monitors while participating in a maximal exercise test. Criterion measures were obtained from the Parvo Medics TrueOne 2400 Metabolic Cart and an electrocardiograph. Results: The HR estimations of the arm-worn (AW) Apple Watch had the highest agreement with the electrocardiogram, with mean absolute percent error (MAPE) of <2.5% for the entire sample, for males, and for females, at all exercise intensities. The HR estimations of the wrist-worn Apple Watch had MAPEs ranging from 3.61% (females at very light intensity) to 14.97% (males at very vigorous intensity). When estimating EE for total exercise bout in the entire sample, the arm-worn Apple Watch overestimated EE, with a MAPE of 39.63%, whereas the wrist-worn underestimated EE, with a MAPE of 32.28%. Both the arm- and wrist-worn overestimated EE for females and underestimated EE for males. Conclusion: Wearing the Apple Watch Series 1 on the upper arm versus the wrist improves the MAPE for HR estimations, but does not improve MAPE for the EE calculations when compared to a criterion measure.
© Copyright 2019 Journal for the Measurement of Physical Behaviour. Human Kinetics. All rights reserved.