JMIR Publications

ABSTRACT

Background:

Wearable devices can provide insight on health and well-being using longitudinal physiological signals.

Objective:

In this prospective study, we evaluate the performance of a previously developed consumer wearable physiology-based respiratory infection detection algorithm on a healthcare worker group.

Methods:

The algorithm used resting heart rate, respiratory rate and heart rate variability measures during the sleeping period from a wearable to predict the presence of COVID-19 or other upper respiratory infections.

Results:

In a cohort of 559 participants from January 6th to July 20th 2022, 31 instances of COVID-19 infection were confirmed by polymerase chain reaction (PCR) testing, 14 instances of COVID-19 confirmed by home test and in total 80 instances of respiratory virus (COVID-19 or other respiratory viruses confirmed with PCR or home test) were observed. For the 31 confirmed cases of COVID-19 infection, 28 received a positive alert within 8 days prior to the PCR test. For the larger set of confirmed respiratory infections (i.e., COVID-19 or other respiratory infections using PCR or home test), 63 received a positive alert within the 8 day window. Across all the cases, the estimated false positive rate on a prediction per day basis was 2% and positive predictive value ranged from 4% to 10% on this specific population with an observed incidence rate of 198 cases per week per 100k. Detailed examination of questionnaires filled out after receiving an alert revealed physical or emotional stress events such as intense exercise, poor sleep, stress or excessive alcohol consumption could result in a false positive.

Conclusions:

Thus, the real-time alerting system provides advance warning on respiratory viral infections as well as other physical or emotional stress events that could lead to physiological signal changes. This study shows the potential of wearables with embedded alerting systems to provide information on wellness measures.