Scripps researchers find COVID-19 signals in daily activity data from Fitbits, smartwatches

By casting a wide range of digital nets to pick up signs of a coronavirus outbreak, researchers want to build an early warning system that uses data from people’s wearable activity trackers.

Currently, initial data from a study by the Scripps Research Translational Institute could help devices such as smartwatches and Fitbits find cases of COVID-19, helping public health efforts to slow the spread of the disease. Indicates that there is.

Launched in March, the DETECT study and its smartphone app collect daily activity data from participants and collate it with self-reported symptoms and diagnostic tests that participants may take.

Results from the first 6 weeks of a study spanning more than 30,000 participants suggest that significant changes in a person’s normal heart rate, sleep quality, and daily movements may help identify new infections. I showed that there is.

“One of the biggest challenges in stopping the spread of COVID-19 is the ability to quickly identify, track and quarantine infected individuals.” Said Giorgio Quer, director of artificial intelligence at Scripps, the lead author of this study.

“Early identification of people who are presymptomatic or even asymptomatic is especially valuable, as people can become more infectious during this period. That is the ultimate goal,” Quer said. Said.The results of the study were published in Nature medicine..

Relation: Fitbit has posted early findings showing that trackers can identify cases of COVID-19 before symptoms settle.

These methods can be combined with more common but ineffective methods of screening COVID-19, such as temperature readings and questions about a person’s travel history. This is an approach that is likely to miss asymptomatic cases. In addition, less than one-third of COVID-19 patients recorded only fever during hospitalization, the researchers said.

“The infrequent virus testing and often delayed results do not provide the real-time insights needed to control the spread of the virus,” said Scripps epidemiologist Jennifer Ladin. I will.

In this study, the data model was able to predict who was positive for COVID-19 with approximately 80% accuracy. This is usually due to more sleep and less movement. However, they said that this also included certain deviations from the everyday pattern of pointing to coronavirus infections over other illnesses.

“What’s exciting here is that we have a verified digital signal for COVID-19. The next step is to use it to prevent the spread of new outbreaks,” said Scripps Director and Founder. Eric Topol says. “About 100 million Americans already have a wearable tracker or smartwatch to help us. All we need is just 1% or 2% to use the app.”

Relation: NIH Funds Seven Digital Health Projects for COVID-19

Earlier this year, Fitbit, which collaborated with Scripps’ DETECT study, provided preliminary results from a similar study of its own algorithm that enrolled more than 100,000 participants.

At the time, the system provided 70% specificity, or 30% false positive rate, but it helps encourage people to seek testing sooner and quarantine faster to prevent virus infection. There is a possibility.

The company also showed that the device can detect nearly half of COVID-19 cases at least one day before participants report symptoms such as fever, cough, and muscle aches.

And this week, Fitbit received an award of about $ 2.5 million from the US Army’s Medical Research Division. This is because the algorithm is being developed and could be used to screen military personnel for illness.

“Our research shows that our bodies begin to fight the disease before any visible symptoms appear, and we believe Fitbit can reliably detect these signals, which is ahead of the virus. You have a great opportunity to warn people that you can get sick unknowingly. Spread it to others. ” Said Amy McDonough, General Manager of Fitbit Health Solutions.

The award will help initiate a prospective study with The Feinstein Institutes of Medicine in Northwell Health and help validate early detection programs. Fitbit also said it would work with the FDA and other regulators to determine the best way to make these features publicly available.