JMIR Publications

ABSTRACT

Background:

Smartwatches have the potential to track physical activity during treatment of childhood cancer. Validation of a Withings smartwatch to monitor steps and heartrate in children with cancer during treatment has not yet been performed.

Objective:

Therefore, the aim of this study is to examine the validity of a Withings smartwatch to monitor steps and heartrate during treatment of childhood cancer.

Methods:

Children (8-18 years) under active cancer treatment wore a smartwatch for 12 weeks and were asked to wear an ActiGraph simultaneously for one week to validate step monitoring and to be connected to a heartrate monitor under resting conditions to validate heartrate monitoring. Linear mixed models to test the difference in steps per hour, steps per day and heartrate per minute between the references and the smartwatch with participant ID as random effect were estimated. The repeated measurements were visualized in scatterplots and Bland-Altman plots.

Results:

The mixed models of the difference in steps per hour and per day between the ActiGraph and the smartwatch revealed a significant difference between the two devices (P<.001) where the smartwatch underestimated step count. Results based on the mixed model of the difference in heartrate per minute between the heartrate monitor and the smartwatch revealed no difference between the two devices (P=.24).

Conclusions:

The Withings smartwatch has good validity to monitor heartrate but underestimates step count compared to the reference device during treatment of childhood cancer. Therefore, children with cancer might use the Withings smartwatch to monitor heartrate but should be cautious when interpreting step count of the smartwatch.