Scientists at Scripps Research in California are using smart watches and activity trackers to explore the longer lasting effects of Covid-19 on the body.

By tapping into rich data from wearable sensors, researchers have found that Covid-19 impacts last two to three months, on average, and they’ve been able to draw connections to the severity of early symptoms and the speed of recovery.

Through their work, the team is also bringing greater clarity to the poorly understood subset of long Covid, where some experience health issues for many weeks or months after being infected with the coronavirus.

“Wearable sensor data from smart watches and activity trackers provide a wealth of data about individual baselines, from resting heart rate and sleep patterns to normal daily activity levels,” said Jennifer Radin, an epidemiologist with the digital medicine division at Scripps Research Translational Institute, and leader of the study. “These data are giving us new and better ways to measure how the body changes during infection and how it recovers, which is especially important for a new disease like Covid-19 that we’re still learning so much about.”

The research is part of the Scripps ongoing Detect study, which leverages data from wearable devices to track viral infections. Already, the team has shown it is possible to discriminate between Covid-19 and other viral infections using such data along with reported symptoms.

The latest findings, which look specifically at protracted effects of Covid-19 that can be measured with an activity tracker such as a Fitbit, appear in the scientific journal JAMA Network Open.

In an analysis of longer-term sensor data from the Detect study, which launched in March 2020, Radin and her colleagues found a prolonged physiological impact of Covid-19 infection that lasted about two to three months on average, but varied substantially among the 234 participants in the study who reported symptoms and tested positive for Covid-19. The variation may reflect different levels of inflammation or dysfunction of the autonomic nervous system, which regulates many involuntary bodily functions, according to Radin.

But overall, those with Covid-19 took significantly longer to return to their baselines with more sleep, less activity and higher resting heart rates than individuals who reported symptoms but tested negative for Covid-19. The difference was most marked for resting heart rate. Data showed that some people also experienced a brief period when their heart rate was slower than normal – a condition known as transient bradycardia – about nine to 15 days after infection, which is consistent with other medical observations.

Notably, a small subset of the study population with Covid-19 had a distinctly longer disease trajectory than the others. About 14 per cent of them maintained a resting heart rate of over five beats per minute greater than their baseline, which remained elevated for more than 133 days. During their most acute phase of Covid-19, individuals in this group also self-reported a significantly higher frequency of cough, body ache and shortness of breath compared with other groups.

“Our data suggest that the severity of early symptoms and a larger initial resting-heart-rate response to Covid-19 may be a predictor of how long it takes for individuals to physiologically recover from this virus,” Radin said. “In the future, with larger sample sizes and more comprehensive participant-reported outcomes, it will be possible to better understand why some people recover faster or differently than others.”

According to the Centers for Disease Control, long Covid encompasses many symptoms that can last weeks or months after first being infected with the virus, or can appear weeks after infection. Some of the commonly reported effects are fatigue, brain fog, headache, loss of smell or taste, heart palpitations, difficulty breathing, cough, fever, muscle pain and depression or anxiety. For this study, the focus was on the physiological and behavioural factors that can be measured reliably with fitness trackers.

The Detect study continues to enrol participants as it looks to capture longer-term data on Covid-19 and other viral infections such as influenza. One sub-study, led by Giorgio Quer, is mining wearable data to learn how the body responds to Covid-19 vaccines, looking at responses of those who have had Covid-19 and those who have not. As the data sets grow, so will the opportunity for meaningful new discoveries, Quer said.

Individuals can participate in the Detect study by downloading the research app MyDataHelps, available in Android and iOS. In addition to sharing wearable data, participants are also asked to self-report their symptoms, test results, vaccines and demographic data, and share electronic health record data when possible. More than 38,000 people have enrolled, and researchers hope to increase participation to over 100,000.

The JAMA Network Open article is authored by Jennifer Radin, Giorgio Quer, Edward Ramos, Katie Baca-Motes, Matteo Gadaleta, Eric Topol and Steven Steinhubl. Funding for the research was provided by the National Center for Advancing Translational Sciences at the National Institutes of Health and the Rockefeller Foundation.