Research into COVID-detecting smartwatches still unconvincing

At the start of the COVID-19 pandemic, smartwatch advocates and wearable tech companies believed the devices could help detect disease. They wanted to flag people who might have the disease using things like their heart rate and oxygen levels. The strategy could still be a reasonable way to track the disease, but two years later the promise has not materialized – the research is still underdeveloped, according to a new review published last week in The Lancet.

The review looked at 12 research studies and 12 proposed study protocols published in 2020 and 2021 that attempted to find patterns in data collected by devices such as the Apple Watch, Fitbit and Whoop. Most of these studies focused on people who had previously tested positive for COVID-19. The researchers looked for patterns in wearable data from the few days before someone became ill, rather than following healthy people and trying to predict who would get sick. None of the studies was a rigorous clinical trial, the authors of this new study noted. None of the existing research has been tested to see if a wearable device could actually lead to earlier detection of COVID-19.

Most algorithms designed to guess COVID-19 from wearable data focused primarily on symptomatic illnesses, the study found. Four tried to detect an infection before a person started showing symptoms, with varying success – they were able to detect between 20 and 88% of infections. The models became less accurate the more they tried to predict illness several days in advance. “The accumulating evidence suggests a trade-off between the accuracy of a model and its ability to identify SARS-CoV-2 infection before the onset of symptoms,” the review authors wrote. It would also make the devices less useful as detectors for COVID-19 – part of the promise of this kind of strategy is to flag sick people early enough so they can get tested and isolate themselves before they can transmit. disease to others.

There is good evidence that physiological signals such as changes in body temperature, heart rate variability and other measures are associated with a person becoming ill. But that includes other illnesses, not just COVID-19, and most of the studies in this review didn’t distinguish between COVID-19 and things like the flu. Research at Fitbit found overlaps between flu data and COVID-19 data, said Conor Heneghan, chief research officer at Fitbit. The edge Last year. “My instinct is that it’s going to be hard to reliably tell them apart,” he said.

Using wearable devices as detectors for COVID-19 or other diseases also raises equity issues, the authors of the new review wrote. The studies included in the review had low racial diversity, so it is unclear whether the models would work equally well in non-white populations. This is all the more concerning as research shows that wearables often work differently and less accurately on darker skin tones. And none of the models assessed in the review took the menstrual cycle into account, although there are changes in body temperature and other variables associated with different stages of the cycle.

Despite the limitations of existing research, it’s still possible that wearable devices will end up being a good way to track and monitor disease. We just need to do better research to prove it and to find the best way to use the devices in these situations. Experts believe it would still be useful to have even a basic tool that could alert someone that they might be getting sick.

“It’s just a warning that something is out of your normal range, and it might be something to watch out for,” said Jennifer Radin, epidemiologist in the Scripps Research Translational Institute’s division of digital medicine. The edge Last year.

Leave a Reply

Your email address will not be published.