WHOOP Podcast - The Importance of Respiratory Rate Tracking During The COVID-19 Pandemic

Episode Date: April 1, 2020

At WHOOP, we have been gathering information on coronavirus and we have already uncovered fascinating insights into this terrible disease. WHOOP is investigating a potential connection between changes in respiratory rate and COVID-19 symptoms and is partnering with the Cleveland Clinic and Central Queensland University to study this possible link. Emily Capodilupo joins Will Ahmed to discuss why WHOOP is tracking coronavirus (1:16), the details of our research efforts (2:32), what we have learned so far (4:32), the importance of detecting who is sick (8:12), how data can help (12:21), what we know about respiratory rate and how consistent it is (13:17), why accurate information is tricky to obtain during this time (16:25), understanding exercise intolerance (21:02), and how you can track your respiratory rate using WHOOP (23:33). Plus, Will answers your questions in this week's mailbag (26:00).Support the showFollow WHOOP: www.whoop.com Trial WHOOP for Free Instagram TikTok YouTube X Facebook LinkedIn Follow Will Ahmed: Instagram X LinkedIn Follow Kristen Holmes: Instagram LinkedIn Follow Emily Capodilupo: LinkedIn

Hello, folks. Welcome to the WOOP podcast. I'm your host, Will Ahmed, the founder and CEO of Woop, where we are on a mission to unlock human performance. We build technology across hardware and software and analytics that's designed to help you better understand your body. And certainly that's been an important thing this time of year with coronavirus. And many of the of you have been reaching out to us to understand how your whoop data is being reflected in this crazy time, whether that's stress, whether that's more sleep or less sleep. And in particular, if there's anything that you might be able to tell in your whoop data that relates to COVID-19. So is there anything about your whoop data and how it's changing that could be predictive

of whether or not you have COVID-19 or whether you're at risk for getting COVID-19? We've gotten a ton of messages about this, and of course, as promised, this is the topic of today's podcast with Emily Capital Lupo, our brilliant VP of data science and research. Now, we're going to get to Emily in one second. The timeline to understand here is that the week of March 7th, we initially added the Whoop Journal into the Whoop app. The Whoop Journal, of course, allows you to select all sorts of different behaviors and see how they affect. affect your body. We then, a few days later, so on March 13th, added the ability to tag COVID-19. So we became the first consumer app that allowed for COVID-19 tracking in the app. And the response that we saw was pretty amazing. We had hundreds of positive responses in

the first 24 hours, and those have only since grown. We've had WOOP members reach out saying they wanted to volunteer their data to help us really understand what does WOOP data look like in the context of COVID-19. So this podcast, we're going to go into those findings. We're going to talk about a research partner just announced that we're going to be working closely with the publish these results. It's really exciting. It's really fascinating.

We're learning very quickly as much as we can about this horrible disease. And without further ado, I'm going to introduce Emily. who's going to tell us more about this. Emily, welcome back to the podcast. Thanks for having me, Will. Emily, tell us about the partnership that we announced today with Cleveland Clinic and CQU. Yes, so we've partnered with the Appleton Institute at Central Queensland University in Adelaide, Australia. We chose them as a partner because we've been working with them for the past year and a half in order to understand different aspects of sleep

in order to develop a couple new features that will be coming out in the next couple months and years. Because we had this pre-existing relationship, it was really easy for us to pivot into investigating this COVID data, and we were able to take advantage of the fact that they were already familiar with loop data, which allowed them to move really quickly on this analysis. Then we brought in our collaborators at the Cleveland Clinic in order to bring some more expertise around infectious disease and epidemiology, because our partners in CQU are primarily focused on sleep. So we want to just specifically bring in a little bit more COVID-19 specific

expertise. Frankly, the process for publishing research is normally quite long, quite slow, Emily, right? Talk about how, given the pandemic, everything seems to have sped up. Ordinarily, if you were sort of starting from scratch here, there's so much legal paperwork that sort of goes into all of this before you can do anything. So there's sort of legal paperwork around like the rights to share data about who owns the data and all this kind of stuff. You typically spend months before you ever collected a single data point before you could ever do anything.

We were able to skip that because we already had all these various agreements in place with our partners at CQU, but also in general, it just seems that, yeah, that people are just moving a lot faster, that institutional review boards and ethics approval is sort of getting fast-tracked for anything COVID because there's just such a hunger for information. Like, you know, it's remarkable how much we don't know about this disease. Yeah, it's amazing. Well, so what have you learned, Emily? I know this is what people are dying to hear. What have you learned about WOOP data before, during, and after COVID-19? Yeah, so I think like one of the most interesting things that we've seen is just sort of the role that

respiratory rate is playing in all of this. So, you know, we've had respiratory rate available in our web app for over a year at this point. We've been tracking it for years. And it's not really something that we've drawn a lot of attention to in the loop world. But what we're seeing is that it's one of the most interesting metrics responding to COVID-19. And that's because COVID-19 is a lower respiratory tract infection so the actual like infected cells are the alveoli and that's sort of the point of contact in your lungs between the blood and your lungs and so if as those avioli get damaged we start to see that gas exchange so your ability to sort of get oxygen into the blood and get carbon dioxide out of the blood becomes less and less efficient and so in order to

compensate for that loss and efficiency you have to take more breaths. And so respiratory rate skyrockets pretty early in the course of like a COVID-19 disease progression. And so it's like the first sign that we see. And what's interesting is that it's a, you know, it's a little bit of a unique disease progression. Things like the common cold don't show up first with pulmonary symptoms. So essentially what you're saying is if someone gets exposed to COVID-19, there's a chance that their respiratory rate may actually elevate before they even feel symptoms. Because I think in part what's so complicated about COVID-19 is this idea of you being contagious before you feel symptoms, correct?

So in all viruses, it's kind of important to understand how these work. So the virus comes in and viruses can't reproduce on their own. So they enter your cells, which they call the host cell, and they hijack your cell functionality and turn it into a virus-making machine, reproduce themselves in your cell, and then they get out of your cell by exploding the cell. So the cell bursts open, and now all the little new viruses are out,

and then go off to find new cells to infect. So let's say, like, you have one virus in you, like a little bit of an extreme example, right? it goes into a cell and it births that cell. Our bodies are great at like having absolutely no response to a single missing cell. And so it actually takes a while until what they call like the viral load, so like how much virus is in your body to get up to a point that like enough cells have been birthed that the organ can't like just make up for the missing cells.

So it's like if you have whatever, let's say 100 cells, like if you're missing one, then the other 99 can compensate. If you're missing two, the other 98 can compensate, but at some point they can't compensate until organ function starts to suffer. And once organ function starts to suffer, that's sort of what we start to experience as symptoms. But by the time you have symptoms,

that means that like some critical mass of these cells has been damaged. And eventually we replace them, but if there's enough virus, you're sort of damaging new ones faster than you're replacing them. And so that's when you get like thicker and thicker. And so that period where you don't have a critical viral load, so you're not yet experiencing any decline in organ function,

therefore no symptoms, it's called the incubation period. Now, in a lot of diseases during the incubation period, you're not contagious. You can't give the disease to other people. One thing that makes the SARS-CoV-2, which is the name of the virus that causes COVID-19, particularly dangerous, is that during that asymptomatic incubation period,

That's when actually they think most of the transmission is occurring. And so people are making other people sick before they even know that they're sick themselves. So there's a lot of interest in being able to detect that people are sick before they have symptoms so that they can be quarantined because they're very much dangerous even though they're not yet aware that they're sick. And so one of the problems that we're facing here in the U.S. is like a horrible lack of access to testing. There are tests available out there that like if somebody thought tested before they were symptomatic, they could test positive, but it's very, very hard to get tested right now.

And so the sort of general guidelines from the CDC and this whole idea around like social distancing is that we all need to assume that we have COVID-19 or we have this like SARS COVID-2 virus in us, even if we're asymptomatic because some percentage of us do. And that percentage who has no idea who they are are very dangerous to sort of those, especially the high-risk people around them. And so one thing that WOOP is very interested in looking into is like what can we identify during that incubation period, which by the way, in some people it's up to 14 days. So it's actually like pretty long. Really long, really long.

Yeah, so it's typically shorter than that. So some people are going from exposure to the onset of symptoms and closer to like three to six days, but it can be up to 14 days. And there's some reports that say maybe a little bit longer, but 14 days is like that recommended. quarantine time after exposure because that's typically like if you're going to if an exposure is going to turn into an active infection it'll happen within those 14 days so being able to identify somebody as early as possible like as basically as close to that period of infection as possible would sort of minimize the likelihood that somebody thinking that oh there's no way I'm contagious kind of goes out into the world and infects other people so let me see if I've got this straight right

So in most moderate to severe cases, right, COVID-19 is presenting a lower respiratory tract infection, which is similar to bronchitis, some strains of the flu, right? And since the flu and allergies primarily impact tissues of the upper respiratory tract, elevated respiratory rate is a less common symptom of things like the flu or normal sickness, but elevated respiratory rate could be the smoking gun for identifying COVID-19. Yeah, it's something that we're interested in understanding. So typically when you get a cold, so when you think of like upper respiratory tract infections, everything above the larynx.

And so you're going to have symptoms that are going to be above the collar. So that's like the runny nose, the sore throat, the headache, like all those kinds of things. sometimes like while your immune system is suppressed especially in like older people like when older people die of the flu it's actually because like while their immune system suppressed with the flu they get like a secondary infection which turns into pneumonia but when that happens they've like had the flu for a few days and it's like it gets worse and worse whereas like because lower respiratory tract infections are literally in your lower respiratory tract so in your lungs you see the respiratory symptoms. at the onset of the disease progression. So you're going to see it much earlier. One of the things that's sort of worth mentioning is that in the U.S., it's still flu season. So before all of the COVID pandemic,

you know, most people who are getting sick with coughing, fatigue, all these different symptoms, the obvious culprit was the flu. And the flu didn't go away just because COVID got added to all of this. And so, you know, sort of being able to have some metrics that potentially differentiates between these things is really interesting. And I think it's worth mentioning, too, that, like, whoop, we're not a medical device.

Our goal is not to diagnose, you know, any disease or condition, especially not COVID-19 or the flu. But sort of what is interesting is if you have a baseline, you've been on whoop for, you know, a few days prior to getting exposed, and then now you know that your respiratory rate is elevated, that's information that you can give to your doctor. And it's just another data point that's going to help them make, like, the right call. And especially in a world where, like, a lot of primary care doctors are trying to go to, like, telemedicine, they're trying to avoid office visits whenever possible. And we don't want somebody who has the flu to go to the hospital and potentially get exposed to COVID-19.

And now all of a sudden they have both, right? And so it's just like another data point that's potentially valuable. And one of the things that we've seen in our data that I think is just sort of an interesting thing to point out is just that, like, with people who are, so the typical resting respiratory rate is between 12 and. and 20 breaths per minute. So that's like the time it takes to inhale and then exhale would be one breath. And we do that, you know, roughly every five seconds or so. And so if somebody whose athletic has a resting respiratory rate of, say, 14 breaths per minute,

and then it goes up to 17 breaths per minute, percentage-wise, that's actually a really meaningful increase, but they're still kind of right in the middle of that normal range. So if you just had a single point reading of like 17 respirations per minute or breast per minute, that wouldn't necessarily raise any particularly interesting flags. But if you pair that with the context of like, oh, but my baseline is 14, all of a sudden that's just another interesting data point that can kind of add to this conversation. And one of the things that we've noticed about respiratory rate that I think is sort of particularly relevant here is that in healthy individuals, respiratory rate varies very little from data.

today. Most users, they're going to vary, like, less than one breath per minute over, like, you know, months. And so when it goes up from like 14 to 17, that's statistically extremely significant. Now, what have we seen for respiratory rate changes within specific whoop members? I know we've got a couple folks that we're allowed to talk about. This one whoop member, he actually posted on Reddit, and we found his Reddit post. And then, ended up contacting him and he did give us permission to share his story here today. He got exposed on March 7th and then felt totally fine. Woke up Tuesday morning, March 10th.

He goes to do his typical morning workout and fails the workout. And this is actually a sort of interesting early warning sign about an infection because exercise intolerance is like a pretty typical before you have symptoms at rest. When we exercise, we try to access our full lung capacity for a healthy individual at rest. You're not even trying to use your full lung capacity. And so you don't necessarily notice that, like, they're sort of less at the margin. When you go to work out, because you're trying to use all of it, you sort of notice that there's some of it's not there. So he goes to work out.

He fails the workout. Can't complete it. Goes to work. End up like coming home early because, you know, he's exhausted. goes to bed and that night spikes 103 degree fever and within a few days got tested, tested positive. What we saw in this user's case was that in the 10 days prior to exposure, his respiratory rate, it ranged from like 15.4 to 16.1. So that's actually like a really, really tight range.

And then that night that he got sick and spiked that fever, it jumps up to 18.1 and it goes to 18.5 the following day. And so like that, when you go from like that. Yeah, huge, huge is 17% increase. We typically see like one, two percent night overnight, and that's meaningless. 17% is extremely significant. And so we saw this, and he was as far as we know, one of the very first loop members to get COVID-19 and also to be tested.

So one thing that's a little bit tricky in some of this research is because testing is so unavailable right now. There's a lot of noise in the data where it's like there's a lot of people who think they have COVID-19, But, you know, do they actually have a flu? Is this actually something, you know, a cold? This is actually the air is dry and your throat's tickling and turns out should be nothing. I think there's sort of a lot of noise out there. And so, you know, this really early on, we saw this extremely significant respiratory rate response. And so that kind of turned us on to this idea that respiratory rate might actually be a really useful metric to be tracking during this pandemic.

And one thing also that, you know, I think Woop is particularly excited out and proud about. just earlier this year, Woot became the first wearable to have their respiratory rate validated by a third party. We're actually still the only wearable who's had their respiratory rate validated by a third party, and we're very, very good at it. So we're within one breath of truth during sleep. And so it's something that we can put a lot of stake in because we know that when we say this is your respiratory rate, you know, we have this very rigorous validation, which we'll link in this show notes for this podcast showing that like when we say your respiratory rate was 15 and now it's 18 like you can actually trust that that changed the part such a good point now what else did we see in this individual's data and maybe we've

been seeing in general resting heart rate yeah so his resting heart rate went up a lot too you know i think we could just as easily talk about the interest in tracking resting heart rate during this time. I think the reason why we're so focused on respiratory rate is because resting heart rate is a little bit more of a non-specific symptom. So resting heart rate going up, yeah, that does mean that some things up, but there's tons and tons of things that could cause your resting heart rate to go up, right? If your resting heart rate will go up if you had a really tough workout, it'll go up if you went to bed drunk, it'll go up if you're sick with pretty much anything. And so because it's a non-specific indicator, while we definitely expect resting heart rate

to go up. We saw it with this one user's data. We've seen it with our other confirmed cases of COVID-19. What I really don't want people to do is to listen to this podcast, see their resting heart rate go up a little bit and then freak out, right? Because there's so many things that could cause it that like to sort of turn to, oh, I must have COVID-19. Yeah, right. Yeah. So while we definitely expect it to go up, we've seen it go up. And we're definitely looking to better understand the sort of response to resting heart rate. Similarly, we're seeing heart rate variability go down quite a bit, resting heart rate and heart rate variability because they're so responsive to so many other things, even when respiratory rate stays constant. There are less of what we're focusing on right now.

So if I summarize this properly, having an elevated resting heart rate and a lower heart rate variability seems to be consistent with people who have tested positive for COVID-19. However, it's nonspecific in the sense that those two things could also mean you overtrained or could also mean that you have a common cold or could also mean that you're hung over, right? So it doesn't necessarily, if those are the only two things that you saw on your data, your immediate reaction should not be, I have COVID-19. It's just a sign that something's going on in your body.

And I will add that if you saw that your respiratory rate increases, that's also not a super specific symptom in that I wouldn't jump to COVID-19 there either. It's just that fewer things elevate your resting respiratory rate than elevate your resting heart rate and lower your heart rate variability. So it's like slightly more specific but still definitely non-specific. Like if you have TB, if you have bronchitis, you're going to see that elevated respiratory rate. If you have asthma, you know, allergies, all of those things sometimes can, depending on how they're presenting. And then also like if you go to altitude, so I know a lot of people are like fleeing the cities to get out of more densely populated areas.

So if you flew went to altitude, you'd expect to see your resting respiratory rate go up just because the air is thinner. So there are other causes of an increased respiratory rate. We're definitely not claiming that that's a COVID-specific thing. It's just slightly more specific than resting heart rate and HIV changes would be. The other thing that you identified here that I thought was interesting is this idea that exercise is harder than usual. And that that might happen before you're consciously aware of other symptoms. And would you frame that as more specific?

Is that more in the camp of respiratory rate in terms of being more specific towards COVID-19? Or is that more in the camp of increased resting heart rate, decreased heart rate variability, which is more in the camp of something's up with your body, but not specifically COVID-19? It's consistent with the earlier stages of a lower respiratory tract infection. Because what's happening during incubation is that you're slowly losing organ function. And at rest, it's much easier to compensate for those damaged cells that, you know, the virus has infected in birth than it is, when you're sort of trying to exercise and therefore access your full capacity. So healthy people at rest are not using like all of their lungs ability.

And so you might not notice that you've lost that little bit at the top. But when you go to exercise, you will notice that. So before you're sort of like fatigued and that your lungs are like having any kind of symptoms at rest, you'll see these symptoms during exercise. But again, I'd expect to see the same thing with bronchitis, the same thing with tuberculosis, the same thing with like any number of lower respiratory tract infections, the other pneumonia type diseases, none of this is to replace proper COVID-19 testing, but I wouldn't necessarily expect to see exercise intolerance in the same way,

you know, with other infections, like before you had symptoms. So fascinating. So exercise intolerance and increased respiratory rate are more consistent with a lower respiratory infection, which is obviously specific to COVID-19, versus increased resting heart rate, decreased heart rate variability, which are a more general sign of things that your body may be run down or may be sick for other reasons. So I think for people listening who are on whoop, those are some things to keep in mind as you look at your data. I think Emily made some very good points there. Again, whoop is not a medical device. We're not intending to diagnose COVID.

COVID-19 or any other disease, and you should talk to your doctors. This is just what we've been seeing in the data, and we do have a fairly large data set to look at. If WOOP members listening to this are wondering, well, how can I look at my respiratory rate? Good news. In the next 24 hours, respiratory rate will now be in the WOOP app. So you can go into your sleep on a nightly basis, and you'll be able to see respiratory rate as one of the breakdowns in your sleep analysis. And that is, of course, when we capture respiratory rate, correct, Emily? Yeah, so what we report is the median across the entire sleep. But for those people who don't want to wait 24 hours,

it's actually been available in the web app for over a year now. So if you go to app.com and log in and then go over to the sleep page, you can not only see your respiratory rate from last night, but you can look at trends over the last, you know, year plus for those of you have been on whoop that long. Now, for Woop members listening to this, you'll, of course, be able to see COVID-19 tracking in the WOOP journal. If you do have COVID-19, we ask that you toggle that on, and there's a whole symptom

tracker now to go with it. So we're building a big de-identified data set on this. We're going to report back to our WOOP members, hopefully to society at large, to collect as much data as possible around this fucking disease. And as a result, be able to provide feedback to all of you on how to stay safe and stay healthy out there. So keep an eye out for that. That's in the WOOP journal. We're constantly making updates to make this data as robust and as accurate as possible. Any WOOP members or anyone listening to this who has questions about our study or our analysis or in general? You can

always email membership services, which is support at WOOP.com, and they will answer questions about your data. That is part of the service that we provide for all WOOP members. You can also, find us at the locker at whoop.com, which is where you can email us questions specific to this podcast. Our handles at Woop. You can find me at Will Ahmed. We will also be answering all of your questions. This is something that's super important to us here at Woop. And we're doing our very best to help understand it. Emily, thank you so much for coming on as always. And you're going to be back shortly to tell us more about what we're learning here. Thanks for having me, Will. Thank you, everyone, for listening. Emily does a phenomenal job, I think, explaining

some of these complicated concepts. I hope you enjoyed that. I'm going to answer some whoop questions here shortly. But before I do, I want to remind all of our listeners that you can get 15% off a WOOP membership by using the code Will Ahmed, that's W-I-L-L-A-H-M-E-D at checkout, and that will get you 15% off a WOOP membership. Marsha Williams asks, I am curious about different competitors versus WOOP. My tool is telling me that training's unproductive while WOOP tells me I'm doing a good job matching training with recovery. so-and-so also gives me vastly different sleep statistics along with a lower resting heart rate.

It would be nice to understand how Whoop is different from competitors. I'm sort of paraphrasing Marcia's question here, and I'm intentionally not including the product that she's comparing us to. I think I'll just say this broadly speaking. Whenever you're looking at wearable technology, make sure that the technology has third-party validation. We've published our third-party validation. it shows that we are exceptionally accurate in measuring sleep, resting heart rate, heart rate variability, now respiratory rate, which we just spent a lot of time talking about. So if you compare whoop to a product that does not have third-party validation, that is probably why you are seeing

a difference. Accuracy is really important to us. It's why we don't, you know, do a ton of other things. We don't, we're not a smart watch. We don't do phone calls. We don't allow you to call an Uber. But what we do do really well is we collect data and we collect data really accurately. We collect about 50 to 100 megabytes of data on a person per day, which is by far the most data of any product on the market. So that's a little bit of whoop versus competitors. Marcia, thank you for the question. Okay, Kathy asks or recommends, I should say, perhaps you could do a podcast on how various types of exercise affect sleep, things that could offset the higher cortisol level and things that could worsen it.

Love my whoop data. Thank you, Kathy. All right. Well, thank you, Kathy. We will definitely look into that. And Julian asks, I have set up a team today for over 50s athletes. I'm writing because I think doing a podcast and in fact some positioning for older athletes would be great. There's a lot of conversation around aging and aging well and the role of exercises particularly well known in helping this.

Great feedback, Julian. And we also have seen a growing interest in WOOP for the over 50 population, and we will do a podcast on aging. No problem. All right, folks, well, that's it for today. I hope everyone enjoyed today's podcast. Stay healthy and stay in the green. My best to you and your family during these crazy times. Thank you.