By casting a broad, digital net to pick up the signs of coronavirus outbreaks, researchers hope to build an early warning system that uses data from people’s wearable activity trackers.
Now, initial data from a study by the Scripps Research Translational Institute showed that devices such as smartwatches and Fitbits may be capable of spotting COVID-19 cases, and could aid in public health efforts to slow the spread of the disease.
Launched in March, the DETECT study and its smartphone app collects daily activity data from participants and matches it with their self-reported symptoms, as well as any diagnostic tests they may take.
Results from the study’s first six weeks, spanning over 30,000 participants, showed that significant changes in a person’s normal heart rate, sleep quality and daily movement could help pinpoint new infections.
“One of the greatest challenges in stopping COVID-19 from spreading is the ability to quickly identify, trace and isolate infected individuals,” said the study’s first author, Giorgio Quer, director of artificial intelligence at Scripps.
“Early identification of those who are pre-symptomatic or even asymptomatic would be especially valuable, as people may potentially be even more infectious during this period. That’s the ultimate goal,” Quer said. The study’s results were published in Nature Medicine.
These methods could be combined with more common, but less effective, ways of screening for COVID-19, such as temperature readings and asking about a person’s travel history—approaches that are likely to miss asymptomatic cases. In addition, fevers alone were recorded in less than a third of COVID-19 patients during their admission to a hospital, the researchers said.
In addition, “infrequent viral tests, with often-delayed results, don’t offer the real-time insights we need to control the spread of the virus,” said Scripps epidemiologist Jennifer Radin.
In the study, the data model was able to predict which people were positive for COVID-19 with about 80% accuracy—typically as they began to sleep more, and move less. But this also included specific deviations from daily patterns that pointed more toward a coronavirus infection than other illnesses, they said.
“What’s exciting here is that we now have a validated digital signal for COVID-19. The next step is to use this to prevent emerging outbreaks from spreading,” said Eric Topol, Scripps’ director and founder. “Roughly 100 million Americans already have a wearable tracker or smartwatch and can help us; all we need is a tiny fraction of them—just 1% or 2%—to use the app.”
Earlier this year, Fitbit—which collaborates with Scripps’ DETECT study—delivered preliminary results from a similar study of its own algorithms, which has enrolled more than 100,000 participants.
At the time, the system provided 70% specificity, or a 30% rate of false positives, but it could prove valuable in prompting people to seek testing earlier and to isolate themselves sooner to avoid transmitting the virus.
The company also showed that its devices could detect nearly half of COVID-19 cases at least one day before the participant reported any symptoms, such as fever, cough or muscle aches.
And this week, Fitbit received an award of nearly $2.5 million from the U.S. Army’s medical research division, to move forward with the algorithm’s development and potentially use it to screen military personnel for the disease.
“Our research shows that our bodies start to fight the disease before more visible symptoms appear and we believe Fitbit can reliably detect these signals, giving us an incredible opportunity to get ahead of this virus and help alert people that they could be sick before they unknowingly spread it to others,” said Amy McDonough, general manager of Fitbit Health Solutions.
The award will help launch a prospective study with Northwell Health’s Feinstein Institutes for Medical Research to help validate the early detection program. Fitbit also said it will work with the FDA and other regulators to determine the best path for bringing these features to the general public.