Abstract
Despite their extensive functionality in the area of physiological data collection and high user adoption, commercial fitness trackers are rarely used in a medical context to improve costly patient care. Therefore, this paper investigates under which circumstances fitness trackers are applicable to give reasonable behavior change advice for patient’s individual health management. To do so we defined daily routine application scenarios and evaluated the accuracy of 15 commercial fitness trackers in a field study for step count, continuous heart rate and sleep duration measurement. In this context we conclude that the sensor technology of the devices is partly accurate enough, depending on the test scenario and device. To effectively enhance individual health management, we suggest a definition of context-aware application scenarios and standardized procedures for accuracy tests. However, the challenges of the current devices regarding susceptibility to outliers and low software and hardware transparency need to be considered.
Recommended Citation
Witte, Anne-Katrin; Blankenhagel, Kim Janine; Korbel, Jakob J.; and Zarnekow, Rüdiger, "How accurate is accurate enough? - An Evaluation of Commercial Fitness Trackers for Individual Health Management" (2019). AMCIS 2019 Proceedings. 29.
https://aisel.aisnet.org/amcis2019/healthcare_it/healthcare_it/29
How accurate is accurate enough? - An Evaluation of Commercial Fitness Trackers for Individual Health Management
Despite their extensive functionality in the area of physiological data collection and high user adoption, commercial fitness trackers are rarely used in a medical context to improve costly patient care. Therefore, this paper investigates under which circumstances fitness trackers are applicable to give reasonable behavior change advice for patient’s individual health management. To do so we defined daily routine application scenarios and evaluated the accuracy of 15 commercial fitness trackers in a field study for step count, continuous heart rate and sleep duration measurement. In this context we conclude that the sensor technology of the devices is partly accurate enough, depending on the test scenario and device. To effectively enhance individual health management, we suggest a definition of context-aware application scenarios and standardized procedures for accuracy tests. However, the challenges of the current devices regarding susceptibility to outliers and low software and hardware transparency need to be considered.