My sleep tracker first caused me to panic early in the pandemic. Even though most of my data from the previous night (how much time I spent in light, deep, and REM sleep) was fairly normal, my breathing rate was twice as fast as usual. I was testing SleepScore, GGT’s pick for the best sleep-tracking app, and I noticed that even though most of my data from the previous night (how much time I spent in light, deep, and REM sleep) was fairly normal, my breathing rate was twice as fast as usual. A short Google search verified what I previously suspected: fast breathing could be a sign of a respiratory infection. “Oh, my god, I have the coronavirus,” my thoughts jumped to the spur-of-the-moment conclusion.
It wouldn’t be the first time that a sleep tracker was used to detect disease. Sleep trackers are devices that measure how long you sleep and how much time you spend in each sleep stage to determine the quality of your sleep. However, many sleep trackers, particularly wearable sleep and fitness trackers, collect a lot more information, such as your nocturnal heart rate, HRV, body temperature, and breathing rate, as well as your sleep duration. Researchers have been analyzing this massive amount of data to see if they can spot (often subtle) differences in people’s baselines that can forecast the development of infections like COVID-19, particularly in asymptomatic or pre-symptomatic cases.
Since 2014, Michael Snyder, PhD, a Stanford University professor of genetics who has founded and consults with a number of precision medicine companies, has been researching the ability of wearable sleep and fitness trackers (such as the Oura Ring, Fitbit trackers, and the Apple Watch) to detect diseases, primarily using heart-rate data. He’s even used the tactics on himself, using a smartwatch to identify his own Lyme disease: “Basically, my heart rate went up and my blood oxygen plummeted before I was symptomatic,” he says.
Snyder and his team have been working on detecting COVID-19 since the outbreak began. The researchers recently published findings revealing that by analyzing data from popular wearable sleep and exercise trackers, they were able to discover COVID-19 cases (on average) four days before symptoms began and seven days before diagnosis. The study didn’t just look at COVID-19; influenza and other respiratory infections were also found two days before symptoms appeared.
38 of the 50 Oura Ring wearers who contracted COVID-19 had a high temperature detected by their tracker before experiencing symptoms they could recognize.
I evaluated the Oura Ring last year, a wedding band-like device that monitors your heart rate, heart rate variability, respiration rate, and body temperature as you sleep. The Rockefeller Neuroscience Institute (RNI) at West Virginia University has completed a research on the ring’s ability to track COVID-19. Your heart rate and breathing rate will likely increase when you’re sick, while your heart-rate variability will likely decrease. RNI states that it predicted instances of COVID-19 three days before the beginning of symptoms with over 90% accuracy using an app, an AI model, and data from the Oura Ring (the study has not been peer-reviewed).
Meanwhile, researchers at the University of California, San Francisco, have used the Oura Ring’s skin-temperature-tracking capabilities to detect fever early (Oura Health provided less than 10% of the study’s funding, and some of the researchers are affiliated with the company, according to a company representative). In a peer-reviewed study of 50 Oura Ring wearers who had COVID-19, 38 of them had an increased temperature detected by their tracker’s continuous monitoring before exhibiting symptoms they could recognize. The NBA issued rings to players participating in the Orlando, Florida, “bubble” this summer because of the Oura Ring‘s initial promise in detecting asymptomatic and pre-symptomatic COVID-19.
However, there is a downside to all of these technologies’ potential. A tiny research of 27 individuals with chronic heart disease conducted by the University of Copenhagen in 2020 indicated that wearing a fitness tracker can increase anxiety, especially when users don’t fulfill their fitness objectives or have frightening health indicators (something I could relate to after my respiratory-rate scare). The researchers came to the conclusion that in order for these gadgets to be most useful, users would require help from health experts to properly understand the data.
The gadget may also take an inaccurate reading as a result of a fault or user error. That could have been the case with my elevated reading, which turned out not to be an early symptom of infection. My early-morning health scare was most likely caused by GGT’s policy of testing many sonar sleep trackers at the same time. (It’s understandable that my reading was incorrect because there were twice as many sonar waves reflecting off my chest.)
After months of testing, I’m not convinced that these gadgets are useful for this type of predictive health tracking. Confronting so much data from the devices and apps felt unsettling rather than reassuring at a time when I was already hyper-vigilant about my health. Clearly, more reliable and accessible testing, as well as safety procedures like masking and vaccination, are more effective approaches to combat the current pandemic. For the time being, I’ll keep using sleep trackers to learn more about my sleeping habits and (maybe) predict when I’ll catch a cold.