The results of several ambitious studies testing wearables as early predictors of for Covid-19 are in — and they suggest that data from devices including Apple Watches, Fitbits, and Oura smart rings may be useful for flagging some infections in people before they even feel ill.
Recently published research from ongoing efforts at three high-profile institutions in the Golden State — the University of California in San Francisco, Stanford University, and Scripps Research Translational Institute in San Diego — indicate that wearables can detect a bump in heart rate or temperature, the most consistent signs that the body is mounting a response to an external threat before symptoms appear. Feeding those data to algorithms that crunch large amounts of information provides a sort of traffic map for the spread of the virus — and could prove a useful tool in the pandemic response in the months to come.
While the Scripps and Stanford studies are device-agnostic — they accept data from people with Apple Watches, Fitbits, and Garmins — the UCSF study is a partnership with smart ring maker Oura and analyzes data from those devices only.
Although each effort is being conducted separately, all of the studies center around a common principle that by establishing a baseline set of biometrics for every study participant — including temperature, heart rate, and activity and sleep levels — researchers can detect deviations that are suggestive of illness.
“Forget precision medicine. This is precision health,” said Michael Snyder, Stanford University School of Medicine genetics department chair and director of genomics and personalized medicine. Snyder is leading the school’s research on wearables and coronavirus.
Oura’s findings, published Monday in Scientific Reports, showed that temperature data from its rings could be used to foreshadow coronavirus infections in 76% of participants with the virus roughly three days before they felt sick. The study, conducted in collaboration with UCSF, enrolled 65,000 people over five months and involved analyzing data from 50 people who said they had Covid-19.
For Stanford’s study, published in November in Nature Biomedical Engineering, researchers used deviations in people’s resting heart rate to accurately identify roughly 2 out of every 3 people with Covid-19 four to seven days before they felt symptoms. The study recruited more than 5,000 people and examined data from 32 people who reported a confirmed Covid-19 diagnosis.
The findings were similar to those from Scripps, which in October published a study in Nature Medicine with close to 31,000 participants — of whom 54 reported testing positive — showing that by pairing wearable data including sleep and activity levels, heart rate, and temperature with self-reported symptoms, researchers could predict Covid-19 with about 80% accuracy, a significantly better result than if they had tracked people’s symptoms alone.
Most of the researchers don’t anticipate consumers being able to turn on a virus-detection feature on their wearable just yet. Before taking such a step, all of them are conducting further studies designed to refine the systems, including work to reduce the rate of false-positives, which happen when a tool falsely flags a healthy person as sick. And of course, many people do not have access to a wearable, some of which cost upward of $400.
Snyder and his colleagues at Stanford launched a new wearable prediction study last week that aims to enroll 10 million people over the next few months. The study being conducted at Scripps in San Diego continues into 2022.
Oura, for its part, has already parlayed its research into big deals with professional sports teams with the promise of keeping players and staff safe. The company has partnerships with the NBA, WNBA, UFC, and NASCAR, all of whom are currently using an Oura platform designed to assess the risks of illness among players and staff. Oura is also outfitting Seattle Mariners’ players and staff with rings, and said the Las Vegas Sands casino and hotel group is providing employees with Oura rings to predict their risk of infection as part of its reopening plans.
“Frankly the main reason companies have partnered with us is this mutual understanding that this is still early days, we’re trying to figure all this out, and we’re trying to do as much as possible,” said Harpreet Rai, Oura chief executive officer.
The researchers involved in the UCSF-Oura study hope to further streamline their algorithm in the months ahead and begin thinking about ways temperature tracking could be used for Covid-19 detection and beyond.
“It’s not just, ‘this will work for Covid and nothing else.’ When you’re having inflammation or a fever, how that integrates with medicine, those are conversations we’re having,” said Benjamin Smarr, an assistant professor of bioengineering and data science at the University of California, San Diego, and a researcher involved in the UCSF-Oura study.
While Stanford’s Snyder awaits the results from the university’s larger ongoing study, he said he feels confident his work is already helping to make a dent in the fight against Covid-19 — at least by flagging potential infections in his research participants and encouraging them to stay home and take care of themselves.
“We think we can have an impact literally right now,” said Snyder.