Become a MacRumors Supporter for $25/year with no ads, private forums, and more!

MacRumors

macrumors bot
Original poster
Apr 12, 2001
55,405
17,761


A new study by Mount Sinai researchers has found that an Apple Watch can effectively predict a positive COVID-19 diagnosis up to a week before current PCR-based nasal swab tests (via TechCrunch).

mount-sinai-covid-apple-watch-study.jpg

Published in the peer-reviewed Journal of Medical Internet Research, the "Warrior Watch Study" involved several hundred Mount Sinai healthcare workers using a dedicated Apple Watch and iPhone app for personal health data monitoring and collection.

All participants were also required to fill out a daily survey over several months to provide direct feedback about potential coronavirus symptoms and other factors, including stress.

The data collection ran from April through September, and the main point of focus for researchers was heart rate variability (HRV), a key indicator of strain on the nervous system. This data point was combined with reported symptoms associated with the disease, such as fever, aches, dry cough, and loss of taste and smell.
The Warrior Watch Study was not only able to predict infections up to a week before tests provided confirmed diagnoses, but also revealed that participants' HRV patterns normalized fairly quickly after their diagnosis, returning to normal roughly one to two weeks following their positive tests.
The researchers hope that the results can help anticipate outcomes and remotely isolate individuals from others who are at risk, without having to perform a physical exam or administer a swab test, preventing potential spread before someone is highly contagious.

According to TechCrunch, the study will in future expand in order to look at what other wearables can reveal about the impact of COVID-19 on the health of health care workers, including how things like sleep and physical activity may relate to the disease.

In related research that is ongoing, Apple is currently partnering with investigators at the Seattle Flu Study and faculty at the University of Washington's School of Medicine to explore how changes in blood oxygen and heart rate can be early signals of the onset of influenza and COVID-19.

Previous independent Apple Watch studies have shown that the smartwatch's heart sensors may be able to detect early signs of diabetes and provide early warning signs of atrial fibrillation.

Article Link: New Study Suggests Apple Watch Heart Rate Sensor Can Predict COVID-19 Up to a Week Before a Swab Test
 
Last edited:

sorgo

macrumors 6502
Feb 16, 2016
495
617
Wow, this is pretty massive news if accurate. Hopefully the technology continues to be put to use in a relatively benign and progressive way that builds upon not only warning someone of the possibility of a potential coronavirus case but of other maladies as well—all while, most importantly, protecting user privacy with the utmost earnestness.
 

adrianlondon

macrumors 68040
Nov 28, 2013
3,390
3,978
Switzerland
Surely it's "predicting" any illness? I know runners who regularly wear HRMs have ben using this fact for years. Resting heart rate higher than usual? Heart rate noticeably higher than usual when doing a typical workout? Body is probably fighting something. Or that pre-workiut double espresso has yet to work its way through your system.

I guess the news here is the amount of people now wearing HRMs, which is a good thing.
 

RedTheReader

macrumors 6502
Nov 18, 2019
253
524
If true, that could be pretty useful information. Maybe doing things like drinking more water and fixing sleep schedules (or generally getting more rest) can help people get over the sickness faster. Maybe it won’t stand a chance of getting bad for them.
 

ethanwa79

macrumors regular
Sep 13, 2014
226
819
The Apple Watch could very well end up being one of the most significant medical devices ever made in human history, especially if they end up adding blood-sugar monitoring and more early warning diagnosis like this COVID-19 study. Could end up saving countless lives from all kinds of various conditions.
 

AlexESP

macrumors regular
Sep 7, 2014
226
679
In fact covid and many more things. I’ve always seen my resting HR rise before I actually get sick, and with covid it was the same. But yes, specifically for this one it’s surprising how quickly it decreased from peak (75) to absolutely normal (45) in one day.

8FAABC4A-67CA-413A-BD39-9A30818AACD1.jpeg
 

svanstrom

macrumors 6502a
Feb 8, 2002
787
1,735
🇸🇪
Surely it's "predicting" any illness? I know runners who regularly wear HRMs have ben using this fact for years. Resting heart rate higher than usual? Heart rate noticeably higher than usual when doing a typical workout? Body is probably fighting something. Or that pre-workiut double espresso has yet to work its way through your system.

I guess the news here is the amount of people now wearing HRMs, which is a good thing.
Yes, it predicts a wide range of [somethings]; which could be an infection, stress, not enough sleep, and so on.

But, whatever the cause is, the warning is valid.

Think of is as a warning light in your car; you can probably keep on driving like normal, but if you do you potentially take the problem from a quick fix all the way to a complete failure/crash/burnout.

What we're seeing now is that people wearing heart/health monitors like these makes it possible to turn on these warning lights waaay before they were noticed in the past.

So let's extrapolate from this, and guess a future feature that might be possible due to these findings:

The current Covid-19 apps focuses on tracking outbreaks, alerting those that potentially have been exposed; but imagine if these apps could use the heart rate data to tell people, long before they feel it themselves, that they just in case should stay at home.

We're talking about, based on what they say here, potentially a whole week of a spreader being out of the system.

Potentially this could in an anonymised format also be reported to the local health authorities; making it possible to predict major outbreaks based on the population's heart rates. They could even prepare/stand down healthcare services by predicting what will/might happen a week into the future; making it much easier to not burnout the healthcare workers.

These watches, combined with the magic of scientists, could be amazingly revolutionary.
 

Precursor

Cancelled
Sep 29, 2015
1,091
1,066
Istanbul
Surely it's "predicting" any illness? I know runners who regularly wear HRMs have ben using this fact for years. Resting heart rate higher than usual? Heart rate noticeably higher than usual when doing a typical workout? Body is probably fighting something. Or that pre-workiut double espresso has yet to work its way through your system.

I guess the news here is the amount of people now wearing HRMs, which is a good thing.
It's quite the opposite. Resting heartbeat of athletes is much lower than usual.
 

adrianlondon

macrumors 68040
Nov 28, 2013
3,390
3,978
Switzerland
It's quite the opposite. Resting heartbeat of athletes is much lower than usual.
The opposite of what? They can measure an increase the same as everyone, once it is higher than usual.

I think you took my "usual" to mean some kind of population average and not the "usual" heart rate of the person being measured.
 

mani

macrumors member
Sep 15, 2007
43
5
But is this really specific to covid? I'd imagine that many infections cause change in HRV.

by the way there seems to be a misunderstanding among all previous commentators about what HRV is. HRV is expressed in miliseconds and roughly speaking it's the time between successive heart beats (as measured by the Apple Watch ECG). this is not the same as the heart rate in beats per minute where you take averages across several heart beats.
 

svanstrom

macrumors 6502a
Feb 8, 2002
787
1,735
🇸🇪
But is this really specific to covid? I'd imagine that many infections cause change in HRV.

by the way there seems to be a misunderstanding among all previous commentators about what HRV is. HRV is expressed in miliseconds and roughly speaking it's the time between successive heart beats (as measured by the Apple Watch ECG). this is not the same as the heart rate in beats per minute where you take averages across several heart beats.
Would you accept if I say that using HR as a passively measured indicator could be used to conditionally sort of bootstrap a person into doing the manual HRV measurements; upon which we practically have gotten into the part where the HRV-based information is relevant?

(Just adding a step here to bridge talk about HR and HRV.)
 
  • Like
Reactions: Captain Trips

sanook997

macrumors regular
May 29, 2012
162
88
Bangkok


A new study by Mount Sinai researchers has found that an Apple Watch can effectively predict a positive COVID-19 diagnosis up to a week before current PCR-based nasal swab tests (via TechCrunch).

mount-sinai-covid-apple-watch-study.jpg

Published in the peer-reviewed Journal of Medical Internet Research, the "Warrior Watch Study" involved several hundred Mount Sinai healthcare workers using a dedicated Apple Watch and iPhone app for personal health data monitoring and collection.

All participants were also required to fill out a daily survey over several months to provide direct feedback about potential coronavirus symptoms and other factors, including stress.

The data collection ran from April through September, and the main point of focus for researchers was heart rate variability (HRV), a key indicator of strain on the nervous system. This data point was combined with reported symptoms associated with the disease, such as fever, aches, dry cough, and loss of taste and smell.
The researchers hope that the results can help anticipate outcomes and remotely isolate individuals from others who are at risk, without having to perform a physical exam or administer a swab test, preventing potential spread before someone is highly contagious.

According to TechCrunch, the study will in future expand in order to look at what other wearables can reveal about the impact of COVID-19 on the health of health care workers, including how things like sleep and physical activity may relate to the disease.

In related research that is ongoing, Apple is currently partnering with investigators at the Seattle Flu Study and faculty at the University of Washington's School of Medicine to explore how changes in blood oxygen and heart rate can be early signals of the onset of influenza and COVID-19.

Previous independent Apple Watch studies have shown that the smartwatch's heart sensors may be able to detect early signs of diabetes and provide early warning signs of atrial fibrillation.

Article Link: New Study Suggests Apple Watch Heart Rate Sensor Can Predict COVID-19 Up to a Week Before a Swab Test
The Apple Watch continues to prove it is Apple’s most underrated product.


A new study by Mount Sinai researchers has found that an Apple Watch can effectively predict a positive COVID-19 diagnosis up to a week before current PCR-based nasal swab tests (via TechCrunch).

mount-sinai-covid-apple-watch-study.jpg

Published in the peer-reviewed Journal of Medical Internet Research, the "Warrior Watch Study" involved several hundred Mount Sinai healthcare workers using a dedicated Apple Watch and iPhone app for personal health data monitoring and collection.

All participants were also required to fill out a daily survey over several months to provide direct feedback about potential coronavirus symptoms and other factors, including stress.

The data collection ran from April through September, and the main point of focus for researchers was heart rate variability (HRV), a key indicator of strain on the nervous system. This data point was combined with reported symptoms associated with the disease, such as fever, aches, dry cough, and loss of taste and smell.
The researchers hope that the results can help anticipate outcomes and remotely isolate individuals from others who are at risk, without having to perform a physical exam or administer a swab test, preventing potential spread before someone is highly contagious.

According to TechCrunch, the study will in future expand in order to look at what other wearables can reveal about the impact of COVID-19 on the health of health care workers, including how things like sleep and physical activity may relate to the disease.

In related research that is ongoing, Apple is currently partnering with investigators at the Seattle Flu Study and faculty at the University of Washington's School of Medicine to explore how changes in blood oxygen and heart rate can be early signals of the onset of influenza and COVID-19.

Previous independent Apple Watch studies have shown that the smartwatch's heart sensors may be able to detect early signs of diabetes and provide early warning signs of atrial fibrillation.

Article Link: New Study Suggests Apple Watch Heart Rate Sensor Can Predict COVID-19 Up to a Week Before a Swab Test
Unfortunately this functionality or app will never see the light of day. The ability to use a phone camera to predict wether a mole could be Melanoma has an accuracy of near 90%, but it remains understudy by the FDA etc.

I simply don't understand why things like this are not made available immediately, there is zero harm in getting a possibility indicator so you can get a further evaluation.
 

arn

macrumors god
Staff member
Apr 9, 2001
16,286
5,533
Unfortunately this functionality or app will never see the light of day. The ability to use a phone camera to predict wether a mole could be Melanoma has an accuracy of near 90%, but it remains understudy by the FDA etc.
I don't know anything about the tech, but there's a lot of math behind screening tests.

I think u want a screening test to catch most of the real diseases. So assuming you mean it detects 90% of melanoma, that means 10% of people with melanoma being told they don't have melanoma would probably be a bad result (since they wouldn't seek treatment).
 

svanstrom

macrumors 6502a
Feb 8, 2002
787
1,735
🇸🇪
I don't know anything about the tech, but there's a lot of math behind screening tests.

I think u want a screening test to catch most of the real diseases. So assuming you mean it detects 90% of melanoma, that means 10% of people with melanoma being told they don't have melanoma would probably be a bad result (since they wouldn't seek treatment).
At least in 🇸🇪 online doctors have become quite a thing. It's covered by the healthcare system just like a regular visit to the doctor, but you do a video call instead.

So, instead of an app guessing about your mole, you get an actual doctor to do it.

The interesting thing of course comes when you loop back to the original story about the tech solution, and combine the two; making the image analysis help the doctor not miss something. Then it could also go into your records to be the basis on a later analysis; where a doctor could miss a potential problem if he's not made aware to the fact that the mole has changed since the last inspection.
 

Roller

macrumors 68030
Jun 25, 2003
2,671
1,660
It's worth reading the paper if you want the full picture of what the researchers did. As they note, there is some published data showing that HRV may be predictive of infection, which is what prompted them to do this. They do note several limitations to their study, which is true of all research:

"First, there was a small number of participants who were diagnosed with COVID-19 in our cohort limiting our ability to determine how predictive HRV can be of infection. However, these preliminary findings support the further evaluation of HRV as a metric to identify and predict COVID-19 and warrant further study. An additional limitation is the sporadic collection of HRV by the Apple Watch. While our statistical modelling was able to account for this a denser dataset would allow for expanded evaluation of the relationship between this metric and infections/symptoms. The Apple Watch also only provides HRV in one time-domain (SDNN), limiting assessment of the relationship between other HRV parameters with COVID-19 outcomes. Lastly, an additional limitation is that we relied on self-reported data in this study, precluding independent verification of COVID-19 diagnosis."

But it's interesting and supports the need for further work on the value of wearables in detecting and monitoring disease.
 

RedTheReader

macrumors 6502
Nov 18, 2019
253
524
An additional limitation is the sporadic collection of HRV by the Apple Watch. While our statistical modelling was able to account for this a denser dataset would allow for expanded evaluation of the relationship between this metric and infections/symptoms.
My understanding is that the Apple Watch doesn’t monitor more frequently because of battery life limitations, right? If so, it only further strengthens the argument that the biggest improvement to the watch would be drastic battery increases… somehow.
 
  • Like
Reactions: centauratlas

Ken88

macrumors regular
Dec 28, 2020
161
109
I STILL have my doubts of this finding..will need more info from doctors and scientists before 100% becoming a believer. People often believe EVERYTHING they read online or newspaper these days.
 
  • Like
Reactions: Canezmd

PsykX

macrumors 65816
Sep 16, 2006
1,472
1,216
There are now countless studies that have been able to predict X health problemwith the Apple Watch.

But once these studies are over, we never hear about them anymore... 🤨

Why fon't these end up being new Apple Watch features or at least 3rd-party apps?
 

Canezmd

macrumors member
Feb 6, 2011
79
219
This headline is misleading... The watch CANNOT predict infection a week before the swab detects covid. The watch is not more sensitive than the swab. If those folks had gotten a swab when the heart rate variability changed, they would've likely been positive on the swab. (PCR is super-sensitive.)

It's just that people who get a swab AND were wearing a watch had increased heartbeat variation

Also, it seems like what they mean by "7 days before" is a window including the 7 days before infection, not actually 7 days before the first positive test.

This makes sense as the time from exposure to symptoms is an average of only 5 days with many people experiencing symptoms 2-3 days after exposure. People I know personally had symptoms 2.5 days after exposure.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.