COVID-19 detection: Could your Apple Watch or Fitbit help slow the coronavirus pandemic?
Growing evidence suggests that a smartwatch or wearable such as a Fitbit could help warn wearers of a potential COVID-19 infection prior to a positive test result.
Wearables such as the Apple Watch, Samsung Galaxy smartwatch, Fitbit and other devices can collect heart and oxygen data, as well as sleep and activity levels. Researchers are studying whether a body's health data might signal an oncoming COVID-19 infection.
A COVID-19 infection may not be imminent for a person whose heart or activity data suggests a potential infection. But the increased likelihood – and the ability to alert the patient to get tested and possibly quarantine – could provide a vital tool in preventing the spread of the disease and tracking it, researchers say.
Such findings, if proven out, could lead to remote medical alerts for other possible viruses, flu and undue stress.
Even if ongoing experiments don't lead to a rollout during the current coronavirus pandemic, "if we are hit by a different pandemic or a new pandemic, we have a means … to potentially deploy in real time some of these tools we have been working on," said Zahi Fayad, a professor of radiology and cardiology at the Icahn School of Medicine at Mount Sinai Hospital in New York.
Privacy tip:How to delete cookies from your phone
Fayad is one of the principal investigators there who have been analyzing Apple Watch data from about 500 people in the hospital's health care system. Those who developed COVID-19 had significant changes in their heart variability rate, a measure of nervous system function, compared to the rest of those monitored, they found.
Mount Sinai isn’t the only place with research ongoing into wearables and their potential use in identifying and tracking COVID-19. Earlier identification of a potential COVID-19 patient could reduce the time they are in public, potentially infecting others.
The University of Washington School of Medicine is currently studying how heart rate and blood oxygen readings captured by Apple Watches could provide early signs of the flu and COVID-19. At Purdue University, there's an ongoing study of 100 participants measuring data from Samsung Galaxy smartwatches and electrocardiogram chest sensors to compare and assess smartwatch data accuracy.
“There won’t be a point where a smartwatch can tell you that you’re COVID-19 positive, but it could potentially say, ‘Within the next couple of days, you might be getting sick and should go get tested,’” said Craig Goergen, an associate professor of biomedical engineering at Purdue overseeing the research team, in an announcement of the project.
As part of the Stanford Healthcare Innovation Lab's COVID-19 Wearables Study, researchers have begun alerting participants in real time through the lab's MyPHD app about significant changes in their heart rate or other data. Since they started the alerts in December, researchers have been able to alert 70% of patients who later tested positive for COVID-19 prior to the onset of symptoms, or as they began, said Michael Snyder, professor and chair of genetics at the Stanford School of Medicine.
The study of nearly 5,300 participants, begun in March 2020, found that those infected with the coronavirus had changes in their heart rate and other health data collected by the lab's MyPHD app from Apple Watches, Fitbits and devices from other makers including Garmin.
The findings aren't foolproof, but 81% of the 32 patients who became infected with COVID-19 had changes in heart rates, time spent sleeping and daily steps, the researchers wrote in the journal Nature Biomedical Engineering.
Most (63%) of the COVID-19 patients had changes that could have led to early detection, before the onset of symptoms, they found. Typically, their analysis found identifiable changes four days before the onset of COVID-19 symptoms.
Stanford is seeking more participants; find out about the study on the trial's website.
“Smartwatches and other wearables make many, many measurements per day – at least 250,000 – which is what makes them such powerful monitoring devices,” he says in an online presentation of the research. “My lab wants to harness that data and see if we can identify who’s becoming ill as early as possible – potentially before they even know they’re sick.”
Could Fitbits be a predictive fit, too?
Fitbit is continuing a COVID-19 study with more than 100,000 Fitbit users in the U.S. and Canada enrolled since announcing the trial in May 2020. The study has found breathing rate, resting heart rate and heart rate variability are all useful metrics for indicating the onset of illnesses including COVID-19.
Fitbit is also conducting research with the Department of Defense, the Department of Veteran Affairs and NASA on COVID-19 detection and participating in the Stanford study and one being conducted by the Scripps Research Translational Institute in La Jolla, California.
Fitbit finetunes strategy:Wearable maker signals it is time to take on Apple and Samsung with new smartwatches
The Institute's DETECT (Digital Engagement & Tracking for Early Control & Treatment) has 36,000 people contributing data from Fitbits, Apple Watches and Garmin devices to its MyDataHelps app. The researchers found that resting heart rate and sleep and activity data could predict about 80% of positive COVID-19 cases among those reporting symptoms. They also made sure to keep false positive rate small, a key step towards the potential implementation of such system, they reported in October 2020 in the journal Nature Medicine.
The Scripps researchers hope to get as many as 100,000 people to download the app and help expand the study. Being able to predict those who might develop COVID-19 or other coronaviruses could lead patients to get testing and quarantine to avoid spread.
“We know that common screening practices for the coronavirus can easily miss pre-symptomatic or asymptomatic cases,” says Jennifer Radin, an epidemiologist who is leading the study. “And infrequent viral tests, with often-delayed results, don’t offer the real-time insights we need to control the spread of the virus.”
A concern: The researchers noted that the pool of patients might not be representative of the U.S. population because not everyone can afford wearable tech and older Americans may be less likely to use the devices. Privacy and security of health data would be major challenges to a system designed to alert device wearers or conduct contact tracing after infections.
There's a good pool of potential participants as about one-fifth (21%) of Americans use a smartwatch or fitness tracker regularly, Pew Research has found.
The Mount Sinai researchers hope to eventually add a real-time alert element to their research using the Warrior Watch Study App, which collects data from the Apple Watch and Apple Health app. A total of 13 people were diagnosed with COVID-19 during the initial study period, begun last April and published in the Journal of Medical Internet Research, involving 297 health care workers for an average of 42 days.
Those who were diagnosed had significant heart rate variability differences, measured over 24-hour periods, the week before they were diagnosed, as well as differences from others in the study. "This was seen even in asymptomatic people diagnosed with COVID-19 who didn’t have symptoms at all from COVID," said Dr. Robert Hirten, a gastroenterologist and assistant professor of medicine, and along with Fayad, the project's co-principal investigator.
A change in the heart variability rate does not necessarily signal COVID-19 infection, but "does provide you that potential that you may be exposed," Fayad said. Then the patient can isolate. "It’s an easy thing to do and we know how effective it is."
Follow Mike Snider on Twitter: @MikeSnider.