Experiences measuring sleep and physical activity patterns across a large college cohort with fitbits

In the past few years, a wide variety of highly capable and inexpensive wearable health sensors have emerged. One of the interesting aspects of such sensors is the capability for researchers to longitudinally and automatically quantify important health behaviors, such as physical activity and sleep, with little intervention required by the participant. While the accuracy of these devices has been evaluated in laboratory settings, there exists little public data with respect to user compliance and the consistency of the resulting measurements at a large scale. The focus of this paper is to share our experience in distributing five hundred Fitbit Charge HR devices across a group of college freshmen and to introduce the resulting dataset from our study, the NetHealth Study. We find that when users are compliant, they tend to be exceptionally so, having an average compliance of 86%. User non-compliance does play a role, however, reducing the overall average compliance rate to 67%. We discuss various reasons for non-compliance and also briefly highlight preliminary monitored characteristics of physical activity and sleep in our student population.