CovIdentify Study Aims to Monitor COVID-19 Spread Through Wearable Smart Devices

A new study by Duke researchers may help predict early cases of COVID-19 infections in people and ultimately monitor the spread of the coronavirus by using data generated by wearable smart devices.

The research project, called CovIdentify, is led by co-PIs Dr. Jessilyn Dunn and Dr. Ryan Shaw. One of the challenges researchers and clinicians have noted with COVID-19 is the rate of unidentified infections circulating in the community and continuing the spread. According to the study’s website, there are 60.5 million people in the U.S. using 117 million wearable devices, such as Fitbits. The CovIdentify team believes these devices could be used to develop digital biomarkers for infectious disease.

Previously, Dr. Dunn had worked on a project that used data from wearable devices to detect influenza and other respiratory diseases. Dr. Shaw’s work has focused on using data from wearable devices for patients with chronic illnesses such as diabetes and hypertension – risk factors associated with COVID-19. They believe their work ideally positions their team to tackle this project.

“Our goal is to be able to detect signals of COVID-19 in a person before they know they’re even sick,” Dr. Dunn said. “Using wearables, we are able to see signals for different illnesses. The more pronounced these physiological symptoms are, the better chance we’ll have to detect them.”

Along with other Duke researchers and staff at MEDx, the Pratt School of Engineering, the School of Nursing, and the Department of Biostatistics and Bioinformatics, Dr. Dunn, Dr. Shaw and their team have been working with the CTSI’s Recruitment Innovation Center and Mobile App Gateway teams to create a mobile platform for her study and recruit study participants.

The first phase of the study officially launched the first week of April. Participants are asked to share their demographics and medical information and complete a daily survey asking about people they’ve come in contact with and whether or not they feel sick. These data are stored on secure servers to protect participant privacy. Currently, the study can collect data directly from Fitbit and Garmin wearables, and users will soon be able to connect a multitude of other wearable devices in the coming weeks.

“We want to prevent the spread of this illness,” Dr. Dunn said. “This disease has strange trajectories that vary dramatically. Some patients seem to be doing better and then crash, some don’t have many symptoms. We want to understand why and determine earlier signs for when someone is going downhill.”

Those interested in learning more and participating in the study can visit