Consumer wearable devices, like smartwatches, are ubiquitous and can nudge us to move if we have been sedentary for too long. Smartwatches also help us track steps, heart rate, and sleep duration. The sophistication of data from smartwatches is also rapidly evolving. The large datasets generated by smartwatch users are increasingly mined for research purposes and show some exciting signals on how this kind of data can be used at a population level.
A new paper by Tejawini Mishra and colleagues from Stanford University published in Nature Biomedical Engineering examines smartwatch data’s role in detecting pre-symptomatic aspects of COVID-19.
The study wanted to explore if this data can monitor the onset of symptoms since smartwatches have a growing role in monitoring vital signs at the individual level. The authors accessed data from 5,300 participants; 32 of them had COVID-19, so they further analyzed their activity and physiological data. Of this smaller sample, 81% showed changes in heart rate, activity rate, measured by steps per day, and sleep patterns; 22 were detected at the symptom presentation, and four of the cases were seen nine days before symptoms showed.
The potential value of this data is it could act as an early warning system to smartwatch users. Like today, you will receive a nudge to get up and move if you have been sedentary; an alert could fire if your data were reflecting early signs of having COVID-19. This alert could support more initial signals to self-isolate and follow the CDC’s protocols for being safe, like masking and social distancing. The authors state that 63% of the sample who had COVID-19 could have been alerted if such a system existed. The value of these large datasets to support public health approaches is as yet untapped.
Smartwatch data can be of value at the individual and population-based level. Public, private partnerships to explore this kind of data seem like a prudent step for public health to explore as local and national data capabilities are expanded to keep people safe. At the core of all this are a constant need to consider data use, privacy, and transparency as low trust is a hallmark of the World’s current state. Such a massive effort to demonstrate this data’s value will also consider the general public’s perspective.
Thanks for reading – Trina
(Opinions are my own)
Mishra, T., Wang, M., Metwally, A.A. et al. Pre-symptomatic detection of COVID-19 from smartwatch data. Nat Biomed Eng (2020). https://doi.org/10.1038/s41551-020-00640-6