Physical activity monitors have become increasingly popular in recent years and researchers have started integrating these devices into artificial pancreas (AP) systems to improve the management of T1D. The accuracy of these devices at measuring energy expenditure (EE) may be important for the function of AP systems. Most research to date has determined the accuracy of activity monitors during steady state aerobic exercise in healthy individuals. In this study, the accuracy of the Fitbit Ionic and Garmin vívosmart 3 at measuring EE was assessed against indirect calorimetry (Cosmed K5) during 5 forms of non-steady state activities in individuals with and without T1D. Fourteen adults (Age: 25.8 ± 8.1 year; BMI: 24.1 ± 3.4 kg/m2; 8 T1D) performed a VO2 peak test, resistance exercise, activities of daily living, and high-intensity interval training on treadmill and cycle ergometer, while wearing a Fitbit and Garmin watch. A significant difference in accuracy was displayed by Garmin between individuals with T1D and healthy controls (T1D: 22.6 ± 35.4%; Control: -15.6 ± 24.0%, P<0.0001), but no such difference was exhibited by Fitbit (T1D: 13.5 ± 35.8%; Control: 19.5 ± 37.0%) (Figure).

In summary, the Garmin vívosmart 3 overestimates EE during physical activity in individuals with T1D as compared to healthy controls. This difference may need to be addressed before integrating these devices into AP systems.

R. Pooni: None. S. McGaugh: None. R. Patel: None. M. Riddell: Advisory Panel; Self; Xeris Pharmaceuticals, Inc. Consultant; Self; Lilly Diabetes. Research Support; Self; Dexcom, Inc., Insulet Corporation, Novo Nordisk Inc., Sanofi. Speaker's Bureau; Self; Medtronic MiniMed, Inc., OmniPod. Stock/Shareholder; Self; Zucara Therapeutics Inc.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at