Fitbit Study Uses mHealth Wearables to Detect COVID-19 Symptoms

Early results from a study launched in May by Fitbit indicate mHealth wearables can help providers identify people infected with the coronavirus – often before symptoms are evident.

According to data gathered from more than 100,000 Fitbit users on the company’s telehealth platform, nearly half of the roughly 1,000 users who developed COVID-19 could be identified as having the virus one day before the onset of symptoms, with 70 percent specificity.

“This is important because people can transmit the virus before they realize they have symptoms or when they have no symptoms at all,” Conor Heneghan, the company’s director of research and algorithms, said in a blog on the company’s website. “If we can let people know they should get tested a day before symptoms begin, they can isolate and seek care sooner, helping to reduce the spread of COVID-19.”

The study is also helping researchers identify which metrics can signal the onset of the virus, as well as which symptoms will most likely lead to hospitalization.

The Fitbit study is one of several using connected health devices and platforms to monitor people for signs of COVID-19 and help providers with care management plans for those who are infected and in isolation, either in a hospital or at home.

In March, San Diego-based Scripps Research Translational Institute launched DETECT, a national effort to gather data from smartwatches and activity trackers to analyze how activity, heart rate, sleep patterns and other connected health data can be integrated into a public health surveillance program. Participants are asked to upload their data through the MyDataHelps app.

“In light of the ongoing flu season and the global pandemic of COVID-19, we see enormous opportunity to enhance disease tracking for improved population health,” said Jennifer Radin, PhD, an SRTI epidemiologist who is leading the study. “One way to do this is to leverage and analyze the rich health data that’s already being collected by the millions of Americans who regularly use wearable devices.”

At around the same time, the Stanford Healthcare Innovation Lab launched the COVID-19 Wearables Study, which aims to determine whether data gathered from wearables, including heart rate, skin temperature, and blood oxygen saturation, can be used to predict the onset of a disease like COVID-19 before symptoms are evident.

Fitbit, which is part of a consortium with SRTI and Stanford Health designed to gather studies and evidence on how telemedicine technology can be applied to research projects targeting the pandemic, launched its own study through the Fitbit app in May.

According to Heneghan, researchers have so far found that breathing rate, resting heart rate and heart rate variability (HRV) are the best metrics for tracking illness, with HRV decreasing and breathing rate and resting heart rate increasing in people infected with the virus – sometimes one week before the person even notices symptoms.

In addition, the study has found that fatigue is the most common symptom reported by people with COVID-19, evident on 72 percent of those involved in the study who had the virus; other common symptoms are headache (65 percent), body aches (63 percent), a decrease in sense of taste or smell (60 percent) and coughing (59 percent).

Fever was detected in only 55 percent of the infected participants, Heneghan noted, throwing into doubt the effectiveness of temperature checks in identifying people with COVID-19.  

The study also notes that shortness of breath and vomiting are the most common symptoms in people who were eventually admitted to a hospital for treatment.

Finally, Heneghan reported that, based on the results so far from the study, people with mild cases of COVID-19 recover in about eight days, while those with moderate cases who recover at home take about 15 days to do so. Severe cases, or those requiring hospitalization, average 24 days to recovery, though many cases can last for months – not to mention those dying of the virus.

Through these programs, researchers are hoping to create algorithms that can help care providers quickly identify patients in need of treatment. In addition, they’re looking to develop workflows and standards of care management that incorporate mHealth devices and telehealth platforms.