DTC

Research Supports Viability of Using Health Data from Consumer Wearables to Diagnose COVID-19

 

That Fitbit, smartwatch or other device that many of us wear around our wrist to count our steps, calories burned and other fitness metrics on a daily basis may prove to be an ideal tool for transmitting other significant diagnostic information about our health. And a new study by Scripps Research Translational Institute published in Nature suggests that such information may include whether we have COVID-19. Specifically, the study finds that a smartphone application (or “app”) in combination with passively collected physiologic data from wearable devices, such as fitness trackers, is capable of determining whether a person reporting symptoms is positive or negative for COVID-19.

The Diagnostic Challenge

When it comes to preventing the spread of COVID-19, widespread testing and time are of the essence. Cases must be rapidly and reliably diagnosed so that persons who have or have had recent exposure to somebody has the virus know this information and can go into immediate self-isolation before they expose other people. However, current COVID-19 testing technology lacks the necessary scalability, rapidity and reliability to make this possible.

In addition, current COVID-19 screening at the point of care relies on survey questions about symptoms, recent exposure and travel history (and in some cases body temperature measurements). These screening methods are primarily based on identifying COVID-19 by its symptoms. However, approximately 40 percent to 45 percent of SARS-CoV-2 infection cases are pre-symptomatic or asymptomatic cases.

Wearables and COVID-19 Diagnosis

The idea of using Fitbit and other wearable devices for diagnostic purposes is nothing new. The Nature study authors claim that Smartwatches and activity trackers can improve the ability to objectively characterize each individual’s unique baseline for resting heart rate, sleep and activity and can therefore be used to identify subtle changes in that user’s data that may indicate that they are coming down with a viral illness. Previous research indicates that this “method, when aggregated at the population level, can significantly improve real-time predictions for influenza-like illness.”

The Nature study conducted by researchers from the Digital Engagement and Tracking for Early Control and Treatment (DETECT) is one of the first to consider wearables for COVID-19 diagnosis. The aim was to investigate whether the addition of individual changes in sensor data to symptom data can be used to improve our capability to identify COVID-19-positive versus COVID-19-negative cases among participants who self-reported symptoms.

The Study’s Findings

After analyzing data from more than 30,000 participants from March 25 to June 7, 2020, the researchers concluded that adding individual changes in sensor data improves models based on symptoms alone for differentiating symptomatic persons who are COVID-19-positive and symptomatic persons who are COVID-19-negative.

Of the 30,529 participants enrolled, 78.4 percent connected their Fitbit devices to the study app, 31.2 percent connected the data from the Apple HealthKit, while 8.1 percent connected data from Google Fit (as the percentage numbers suggest, some individuals connected to multiple platforms). Among the 3,811 participants who reported that they were experiencing symptoms, 54 reported testing positive and 279 reported testing negative for COVID-19.

The researchers found that individual changes in physiological measures captured by most smartwatches and activity trackers are, in fact, capable of significantly improving the distinction between symptomatic individuals with and without a diagnosis of COVID-19 beyond symptoms alone.

The Nature study affirms an earlier study cited by the researchers finding that symptom data in over 18,000 SARS-CoV-2-tested individuals captured via a smartphone-based app was helpful in distinguishing between individuals with and without COVID-19. Although the Nature study results are encouraging, the authors were careful to note that they are based on a relatively small sample of participants.

The Fitbit Study

Wearable makers have also been active in exploring the diagnostic potentials of their products. Before the pandemic, San Francisco-based Fitbit applied its own tracking health and wellness metrics technology to develop an algorithm to detect breathing rate, resting heart rate and other factors. The original intention was to alert users to signs of flu infection. But when the public health crisis began, Fitbit pivoted and adapted the concept for COVID-19.

The algorithm was studied only in a retrospective setting, and there was a need for a prospective study to validate it in a real-world setting. Accordingly, Fitbit recently announced that it is collaborating with Northwell Health’s Feinstein Institutes for Medical Research on a study to validate the Fitbit COVID-19 early detection algorithm. The Fitbit study is supported by a $2.5 million award from the U.S. Department of Defense through the medical technology enterprise consortium. The award is part of the consortium’s efforts to keep military personnel healthy by detecting the virus before symptoms emerge.

Several thousand frontline and custodial Northwell staffers are expected to participate in the study. Once the study is initiated, enrolled Northwell employees will be given a Fitbit smartwatch. Upon notification of signs of potential illness, they will be given COVID-19 tests for verification.

Takeaway

Momentum for applying wearables health and wellness measuring technology for purposes of widespread, consumer-based COVID-19 diagnostics is building rapidly, as is the scientific evidence to support the validity of the concept. Were it to succeed, the approach of wedding consumer wearables to COVID-19 detection and differentiation could go a long way in resolving the current rapid testing challenge. Moreover, the early research suggests that using data generated by wearables may help identify infection clusters before wider community spread occurs.

CLOSE TO VIEW ARTICLE x

You have 2 articles left to view this month.

Your 3 Free Articles Per Month Goes Very Quickly!
Get a 3 month Premium Membership to
one of our G2 Newsletters today!

Click on one of the Newsletters below to sign up now and get unlimited access to all articles, archives, and tools for that specific newsletter!

Close

EMAIL ADDRESS


PASSWORD
EMAIL ADDRESS

FIRST NAME

LAST NAME

TITLE

COMPANY

PHONE

Try Premium Membership

(-00000g2)