Long COVID Mobile Monitoring Study Hunts for Answers

So, researchers are starting to supply subjects with their own monitors. The RECOVER Initiative expects to give out 10,000 sensors to people who are eligible based on race/ethnicity, income, and other demographic factors (rural residents for example). After two months, all people in the RECOVER study over the age of 13 will be eligible for the sensors.

The federal program builds on earlier research at places like The Scripps Institute, a center of research into remote monitoring. The institute supplied 7,000 monitors to people in an arm of the All of Us study, a 5-year-old multi-site cohort that aims to collect medical information from 1 million people. 

The devices went to people who have been historically underrepresented in biomedical research, said Scripps researchers, who plan to give out more this year. 

Last month, Vanderbilt University researchers published a study on the tracking data that found a significant post-COVID-19 drop in physical activity. But the data is incomplete because many people can't always afford these devices. Most of the people in the study were "White, young, and active," they wrote.

Researchers at an "All of Us" site at Vanderbilt University, which also used a Bring your Own Device approach, realized that they produced biased results. They reported their findings at the Pacific Symposium on Biocomputing in January.

"[The] majority of participants who provided Fitbit data reported being white and employed for wages," they said. "However, these data represent participants who had their own Fitbit devices and consented to share EHR [electronic health record] data."

Their solution: The program has begun providing Fitbit devices to all study participants who do not own one or cannot afford one. 

Now, the web page for the All of Us study asks visitors to "Learn about the All of Us WEAR study. You could get a Fitbit at no cost! … As a part of the WEAR Study, you could receive a new Fitbit to wear at no cost to you. All of Us will be able to get the data the Fitbit collects. This data may help us understand how behavior impacts health."

Jennifer Radin, PhD, an epidemiologist at Scripps Research Translational Institute, is heading up the DETECT study, which is a remote monitoring research project that has enrolled over 40,000 people who have their own sensors — be it a smartwatch or Fitbit. She was looking at remote monitoring for disease before COVID emerged.

Radin said she began researching remote sensing after working in public health and dealing with outdated data collection systems. 

"They typically rely on case reports that are recorded by pen and paper and faxed or mailed in," she said. "Then, they have to be entered into a database. "

In addition to offering objective data on a subject's physical response to the infection, she said, the data collection can be long-term and continuous. 

DETECT collects data on resting heart rate, which is unique to every person, and activity levels. Both measures are meaningful for those with long COVID. Her research found differences in sleep, heart rate, and activity between those with COVID and those without.

Joseph Kvedar, MD, is a Harvard Medical School researcher and the editor of  NPJ Digital Medicine. He's been studying digital health systems and called clinical research a "beachhead" for the use of data from monitors. But he also said problems remain that need to be worked out. The quality of the devices and their Bluetooth connections are better. But different devices measure different things, and a counted step can vary from person to person, he said. And the problems of the early days of electronic health records have not been fully resolved.

"We haven't gotten to this universal language to connect all these things and make them relevant," he said. 

The All of Us researchers are working with the RECOVER project to address some of those issues. Usually not focused on a single condition, the All of Us researchers are testing a machine-learning approach for identifying long COVID.