Read on Twitter
We have been running one of the world’s most impactful research labs for 11 years.
We just published a landmark study on COVID-19.
https://bit.ly/StanfordCovidStudy
Here is a thread on what we learned
We just published a landmark study on COVID-19.
https://bit.ly/StanfordCovidStudy
Here is a thread on what we learned
1/ The current covid wave is bad.
We are in our third wave of hospitalizations... and it’s the biggest one to date, much bigger than the peak in April and the peak in July. https://bit.ly/CovidThirdWave
We are in our third wave of hospitalizations... and it’s the biggest one to date, much bigger than the peak in April and the peak in July. https://bit.ly/CovidThirdWave
2/ Our current testing system is broken.
Currently, we rely on temperature or nose, saliva, or blood tests to diagnose COVID-19.
The problem is… these tests don’t scale for a population 328 million of with many, many vectors of transmission.
Currently, we rely on temperature or nose, saliva, or blood tests to diagnose COVID-19.
The problem is… these tests don’t scale for a population 328 million of with many, many vectors of transmission.
3/ Testing is broken bc these tests:
a) aren’t low-cost enough for everybody to use all the time
b) aren’t conducted by everyone, everyday — only occasionally
c) require wait period after infection for accurate results — in the meantime you may be unknowingly spreading virus
a) aren’t low-cost enough for everybody to use all the time
b) aren’t conducted by everyone, everyday — only occasionally
c) require wait period after infection for accurate results — in the meantime you may be unknowingly spreading virus
4/ We think a scalable testing solution… is smartwatches.
Consumer wearables are:
a) accurate enough for research use ( https://bit.ly/WearablesGuidelines)
b) already widely deployed
c) can be used to establish individual baselines of health.
Consumer wearables are:
a) accurate enough for research use ( https://bit.ly/WearablesGuidelines)
b) already widely deployed
c) can be used to establish individual baselines of health.
5/ We at Stanford have proven this before.
Our lab has previously shown that smartwatches and simple pulse oximeters can be used for the early detection of Lyme disease days before symptoms appear... https://bit.ly/HealthInfoTrackedFromWearables
Our lab has previously shown that smartwatches and simple pulse oximeters can be used for the early detection of Lyme disease days before symptoms appear... https://bit.ly/HealthInfoTrackedFromWearables
6/ and in retrospective studies, we actually showed that heart rate and skin temperature can be used to detect viral respiratory infections, including asymptomatic infections:
This is exactly what we wanted to establish with COVID-19.
This is exactly what we wanted to establish with COVID-19.
7/ In our COVID-19 wearables study, we sampled nearly 5,300 participants
Of those 5,300, 32 became infected with COVID.
Of the 32 infected... 26 of them had alterations in their heart rate, number of daily steps or time asleep.
Of those 5,300, 32 became infected with COVID.
Of the 32 infected... 26 of them had alterations in their heart rate, number of daily steps or time asleep.
8/ That’s 8 out of 10 cases that showed measurable differences in smartwatch readings. https://bit.ly/CovidPreSymptomsTrackedWithWatches
9/ Even more importantly, of the COVID cases we were able to diagnose via smartwatch…
...88% of them were detected before or at symptom onset, aka before you start feeling sick, and want to get a test
...88% of them were detected before or at symptom onset, aka before you start feeling sick, and want to get a test
10/ In four cases, we were able to detect the virus at least NINE DAYS earlier than symptoms.
11/ Why is early detection so important?
Early detection means infected patients are not spreading the disease while they wait for a positive result from other tests.
Smartwatches are worn daily, which means this method is monitoring daily, not occasionally like other tests.
Early detection means infected patients are not spreading the disease while they wait for a positive result from other tests.
Smartwatches are worn daily, which means this method is monitoring daily, not occasionally like other tests.
12/ Could smartwatch testing prevent COVID-19 spreading at scale?
Using retrospective smartwatch data, we show that 63% of the COVID-19 cases could have been detected before symptom onset in real time.
Using retrospective smartwatch data, we show that 63% of the COVID-19 cases could have been detected before symptom onset in real time.
13/ How does smartwatch testing work?
We monitor 3 variables: heart rate, daily steps, and sleep
We then built an algorithm that compares your values over time against your PERSONAL baseline, not inaccurate population averages.
We monitor 3 variables: heart rate, daily steps, and sleep
We then built an algorithm that compares your values over time against your PERSONAL baseline, not inaccurate population averages.
14/ The warning system in our algorithm is built around around a two-tier warning system (you don’t want the scary movie you watched on Netflix to register a false positive)
https://bit.ly/DesignforCovidWatchStudy
https://bit.ly/DesignforCovidWatchStudy
15/ The kicker: we are giving this algorithm away FOR FREE so everyone with a smartwatch will be able to use it.
We think this is THAT important of a tool in our collective fight against COVID.
We think this is THAT important of a tool in our collective fight against COVID.
16/ What’s next?
We need to expand our study past our original 5,300 individuals to improve our detection method and show that this works at scale.
We need to expand our study past our original 5,300 individuals to improve our detection method and show that this works at scale.
17/ How can you help with this project?
Share this post, let people know what is going on.
Consider being part of the study. Learn more about that here.
Or… https://bit.ly/JoinCOVIDStudy
Share this post, let people know what is going on.
Consider being part of the study. Learn more about that here.
Or… https://bit.ly/JoinCOVIDStudy
18/ Consider a donation to help us fund this research. Learn more about that here. https://bit.ly/ContactStanfordLab
