The Peter Attia Drive - #305 ‒ Heart rate variability： how to measure, interpret, and utilize HRV for training and health optimization ｜ Joel Jamieson

00:00:00.000 Hey, everyone. Welcome to the Drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.540 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.520 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.720 wellness, and we've established a great team of analysts to make this happen. It is extremely

00:00:31.660 important to me to provide all of this content without relying on paid ads. To do this, our work

00:00:36.960 is made entirely possible by our members, and in return, we offer exclusive member-only content

00:00:42.700 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.940 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.200 of the subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.020 head over to peteratiyahmd.com forward slash subscribe.

00:01:04.180 My guest this week is Joel Jameson. Joel is the CEO and founder of Morpheus Labs and eight weeks out.

00:01:11.140 Morpheus Labs aims to work with trainers and individuals to maximize training results using

00:01:15.740 a combination of data science and physiology, primarily through heart rate and heart rate

00:01:20.680 recovery training systems. The system is used by a number of professional sports teams in the NFL,

00:01:26.000 NBA, MLS, NCAA, and more. Eight Weeks Out is a company that helps coaches, athletes, and fitness

00:01:31.680 enthusiasts improve their strength, conditioning, and performance. In this episode, we speak about what

00:01:36.460 sparked Joel's personal interest in the world of heart rate variability and the history of heart rate

00:01:40.940 variability development over time. We break down the science of HRV and how HRV is calculated. There are

00:01:47.380 many different methods and how the interplay between the sympathetic and parasympathetic nervous system

00:01:52.400 affects your heart, the reliability of tracking HRV, and ultimately what it is that HRV is telling us

00:01:59.420 about these autonomic nervous systems. We talk about the decline of HRV with age and what drives this

00:02:05.180 change and how much of it is within our control versus genetically predetermined. We then talk about

00:02:10.460 Morpheus, which is a product that ultimately led to my meeting Joel. We talk about my skepticism around

00:02:17.480 Morpheus when I first began to use it and ultimately why I came to believe that it is a really valuable

00:02:23.220 tool for people when they're training, especially people who might not be as interested in, for example,

00:02:30.060 using lactate testing or other really advanced forms of testing to fine-tune their training zones.

00:02:35.640 We talk about the impact of lifestyle choices on HRV and its significance for overall health

00:02:40.880 and how to use the data from HRV to inform daily choices. Finally, we talk about HRV within the

00:02:46.940 broader context of other health metrics and where it sits in the hierarchy of measurable insights.

00:02:52.780 Final point I'd like to make is that while we speak extensively about Morpheus, I want to make sure

00:02:57.700 everybody understands I have no financial affiliation whatsoever with Morpheus. I'm not an investor in

00:03:03.360 the company. I'm not an advisor to the company. We have no affiliate deal with Morpheus. Of course,

00:03:07.880 we have no affiliate deal with any company. In other words, there is no financial remuneration of

00:03:13.040 any sort that exists between me and this company. I am simply a huge fan of this company and I speak

00:03:19.060 about it often, recommend it to a number of my patients because of my belief in its efficacy in

00:03:24.940 helping people achieve their exercise goals. We do discuss a couple of other companies in this

00:03:30.620 podcast that I do have relationships with. These have been disclosed previously and they're all

00:03:35.640 on my disclosure page, but I would like to again reiterate them here. I am currently a scientific

00:03:40.780 advisor to the company AteSleep and I am a passive investor in the company Aura. Those two companies

00:03:47.340 do have a mention in this podcast as well. So without further delay, please enjoy my conversation with

00:03:53.800 Joel Jameson. Hey Joel, thanks for coming out to Austin. I've been looking forward to this discussion

00:04:03.580 for some time. We've not met before, but had what seems like an endless stream of email communication.

00:04:08.860 So I always appreciate your willingness to not just respond to all my questions, but the thoroughness

00:04:13.300 with which you do so. This is a topic that as we were discussing just a few minutes ago, I think

00:04:18.460 everybody has heard of it. People have a vague sense of what it is, but once you get beyond a very

00:04:25.620 superficial description of it, most people I think don't really understand it. And certainly most people

00:04:32.440 don't understand how to use the data. And I would absolutely include myself in that category. So the

00:04:38.820 topic of course of heart rate variability is near and dear to your heart. Maybe before we dive into the

00:04:44.920 weeds of this stuff, maybe just give folks a bit of a sense of your background and what brought you

00:04:50.700 to the study of this and over what period of time? It's really interesting to me to see the growth of

00:04:55.180 it because I've been using it now for 20 years. And the story of it's really fascinating to me because

00:04:59.620 I was in my early twenties and I had just graduated here at Virginia, Washington, interned there and done

00:05:05.140 some strength conditioning and then was progressing to the Seahawks to work with the same coaches.

00:05:09.500 And there was a track coach named Randy Huntington. And most people probably never heard of Randy,

00:05:13.360 but he was the USA track and field jumps coach. He coached Mike Powell, who broke Carlos's long

00:05:18.400 jump world record. In 1993. Something like that. Yeah. So Randy was around for a long time. He was a

00:05:22.940 tremendous coach and he was from my area. And I just kind of was talking to him one day and I asked him

00:05:27.620 some general advice, you know, as a young coach, what would you suggest and resources and all sorts of

00:05:32.480 stuff. And he writes a phone number down for me and says, you need to call this guy. And I was kind of

00:05:37.920 like, okay, you know, whatever you say, Randy. And so I called this guy in this thick Russian accent

00:05:43.140 answers the phone and says, his name is Val. And I still don't really know what I'm calling this

00:05:48.000 guy for. I just know that Randy told me to. And so he says, I'll be in your area. I'll meet you

00:05:52.820 at the airport. And I still, I'm like, okay, I'm just kind of playing along and I'm not really sure

00:05:57.400 what the whole point of this is, but Randy says, call this guy. So I call this guy and he just says,

00:06:01.340 I will show you the technology. And I'm again, kind of like, okay. So I go down to the SeaTac airport

00:06:07.080 and I see this kind of big Russian looking guy with a trench coat. I mean, it looks like it could be

00:06:12.180 straight out of a movie, introduced himself and he's like, lay down on the couch. I hope you're

00:06:16.800 now out of the airport or. No, I'm in the hotel by the airport. And so again, I'm just in the dark

00:06:22.240 of what is going on. And he says, lay down on the couch. And so I laid down on the couch and he's

00:06:27.060 like, take your shirt off. This point I'm looking around, like, is there some kind of practical

00:06:30.960 joke? Like is Randy just messed with you? This is when you have a lot of faith in Randy. He doesn't

00:06:34.260 have a lot of faith in Randy and paid off. And so he does this big briefcase and he pulls out this

00:06:38.940 big laptop and he starts plugging in wires and all this sort of stuff. And he pulls out these

00:06:44.100 electrodes. He starts putting them on my chest and he does basically an ECG, asking me birthdate and

00:06:48.800 weight and all this sort of stuff. And he's like, don't move. And so I sit there for two and a half,

00:06:52.960 three minutes and I see all this stuff on the computer happening. And again, still just completely

00:06:57.100 in the dark. What is this guy doing to me? Like, I have no idea. I mean, after about three minutes,

00:07:01.660 he's like finished. And he starts to kind of tell me about my recovery status and my readiness,

00:07:06.560 my metabolic profile, my central nervous system. Starts talking about all this stuff that

00:07:10.120 didn't really make a whole lot of sense of how you would know. And he was like,

00:07:13.160 you're much more strength oriented and your cardiovascular system is not very good, which

00:07:16.880 is accurate at the time. Maybe it didn't take a computer to see that because I was definitely

00:07:21.420 on the strength side. But he started just telling me the story of HRV and heart rate variability.

00:07:26.940 And this was, again, 20 plus years ago. It was not something that people were aware of. And I had never,

00:07:32.640 of course, heard of it. And the idea that you could take something out of a laptop,

00:07:35.980 connected to my body and have any idea of physiologically what I was as an athlete or

00:07:40.700 as a human being was completely new and seemed foreign to me and immediately was like, I need

00:07:45.560 this. Because it seemed to me like there's so much of a black box when it comes to fitness.

00:07:51.100 Sometimes I do a workout and I get better. Sometimes I do workouts and I don't. Why?

00:07:55.920 What is the right magic answer here to always get the workouts that I want to see are the results.

00:08:00.620 And so he started talking about the story of heart rate variability. And this is really

00:08:04.340 fascinating because you don't read this when you look at the Western literature.

00:08:07.760 So you probably know it goes back way 1700s. They're aware of B2B intervals. Chinese have 1.00

00:08:14.020 used pulse medicine a long time. But the Russians were pretty ahead of the game as far as application 0.99

00:08:19.440 of HRV. So in the late 1950s, they were aware that from an ECG, you could pull out these B2B intervals

00:08:27.440 and get something more than just heart rate. And so when they would send the first human

00:08:32.000 being in space, Yuri Gagarin, they were able to send back the ECG and some respiratory data

00:08:37.440 and see fundamentally what happened to people when you shot them into space, which they really

00:08:42.240 didn't have a way of gauging without this. And so they saw as soon as he went into space,

00:08:45.960 his HRV went way up, probably because you have changes in blood pressure, eat up less gravity,

00:08:51.000 so you don't have to have as much muscle activity. And they started using this literally all the

00:08:54.860 way back in the 1960s, which is pretty fascinating because you didn't really see that in the Western

00:08:58.600 literature for decades later, really in a meaningful way. Then in the 1980s, they were

00:09:04.660 dominant as far as all the Olympic sports, right? They just crushed us. Now they had a very elaborate

00:09:10.300 drug program. They had a very elaborate training program. They had a lot of different things that

00:09:14.380 gave them that advantage. But in the mid 1980s, they started basically figuring out, can we use this

00:09:19.680 tech for sport performance? And so they put together an engineering team and they started

00:09:24.660 collecting data on thousands of Russian athletes of all levels from their school age kids all the

00:09:29.760 way up to the Olympic athletes. And they collected just populational norms. And they started building

00:09:34.020 the system that was meant to monitor training and be used for this purpose. Don't get me wrong,

00:09:39.480 they had really high volumes, but they did blood monitoring like weekly or monthly. They were

00:09:43.640 constantly testing. They, from an organizational standpoint, their communist structure gave them

00:09:48.820 a very hierarchical way to organize this sort of stuff and monitor it. So anyway, they spent years

00:09:54.360 working on this platform. And then before they could finish it, the Soviet Union collapsed and

00:09:59.200 kind of the whole team that was involved in this just dispersed and left Russia. And just by chance,

00:10:05.100 a lot of them had been involved in track and field, different sports. They ended up reconvening

00:10:09.340 in a track meet in Eugene and kind of just talking about this past project. And they decide that they want

00:10:13.840 to get together and keep working on this because they'd never brought it to fruition. And they did.

00:10:18.300 And that was ultimately the first system that I think was available. I mean, it was the first system

00:10:22.160 that was available commercially with the intent of being used for sport and fitness. And that was what

00:10:27.580 I was being introduced to at the time. This was the result of this project being finished.

00:10:31.640 The problem is they wanted $35,000 for it. It was a very research heavy medical type system that was

00:10:39.060 not easy to interpret. I didn't have $35,000, but I convinced them to let me help them introduce it

00:10:44.280 to teams across the U.S. and gain some exposure for them as we worked at a deal. And I started using

00:10:49.320 it. But it gave you 12 or 14 different metrics of heart rate variability. It required you, like I said,

00:10:56.120 connect electrodes to people before you'd measure them. And really, it was that introduction to it where

00:11:01.400 I started this whole journey. And 20 years of looking at data and coaching people and trying to

00:11:06.840 understand what the data was telling me and how it aligned with all these other metrics has really

00:11:10.640 just led me down this path of how you get to heart rate variability today. And it's certainly been a

00:11:15.000 large change, but it's kind of that crazy journey of starting one day in a laptop in a hotel next to

00:11:21.080 the airport. And now it's on everyone's phones and watches and everything else. But I spent 20 years

00:11:26.420 coaching with it. And that's really the difference is I was a coach. I opened a gym. I worked with lots

00:11:30.860 of fighters and athletes and different teams and military groups using data, looking at HRV,

00:11:36.060 the whole nine yards. So it's been a long time, but it's really fascinating to see it grow.

00:11:41.320 So let's talk a little bit about the actual measurement. So in the example you gave when

00:11:46.820 you were first introduced to it, it was done off an EKG. I assume three leads would be sufficient.

00:11:52.340 It was six back then.

00:11:53.400 Okay.

00:11:53.800 But you can certainly get it from three, obviously.

00:11:55.360 Yeah. So would we say, Joel, that that's still the gold standard for how to measure HRV?

00:12:01.040 Absolutely. I mean, if you're talking about medical grade research quality,

00:12:04.600 you want the cleanest signals, you want the most signals. ECG, three lead, six lead is by far the way

00:12:09.560 to go.

00:12:09.860 This might be a little too in the weeds, but given how technical this topic is, do you want to explain

00:12:15.360 how an EKG works? Because I think it will be relevant to distinguish between what an EKG is

00:12:22.460 doing, what a chest strap is doing, what an optical sensor is doing on the forearm, on the wrist,

00:12:29.320 or on the finger. These are all going to be basically the tools that technology companies are using

00:12:35.240 to measure HRV. But as you and I have discussed and gone deep on this, there's a total difference

00:12:41.860 in the fidelity of the signal depending on where the signal is acquired. And given that our audience

00:12:46.720 here is an appreciative audience for nuance, I think it might be worth explaining from the gold

00:12:52.000 standard all the way down, how these signals are acquired, what's happening physiologically

00:12:55.960 and electrochemically that's enabling the capture of the signal.

00:12:59.880 Sure. I mean, you can kind of group these into two things, right? One is the electrical signal of

00:13:03.380 the heart itself. And that's what we're measuring with an ECG or EKG or the chest strap. You're

00:13:08.800 literally measuring the polarization and repolarization of the heart as the chambers are

00:13:12.800 beating. And you get this electrical signal that gives you the QRS complex. And we're honing in on

00:13:17.860 where those beat-to-beat intervals are. Because ultimately, to get HRV, we need the exact amount of

00:13:23.480 time from one heartbeat to the next, because that's what we're quantifying. So if you have an

00:13:27.740 electrical signal, you get a very clear, clean signal that you can pull out those exact

00:13:33.040 beat-to-beat intervals. And that's where we fundamentally get heart rate variability from.

00:13:37.580 Is it always done R to R? Because that's the cleanest signal?

00:13:40.300 Yeah, it's always done R to R. So you just have to be able to identify where is the peak of the

00:13:44.100 R interval. The more accurately you can identify, here's the peak of the R wave, here's the peak

00:13:47.900 of the R wave, the more accurately you can get that.

00:13:50.680 And if my memory serves me correctly, because it's been so long, right? So the P 0.99

00:13:53.900 is the polarization of the atria. And then the QRS is the ventricular, is it the repolarization or

00:14:02.660 the contraction?

00:14:03.320 I believe it's the contraction.

00:14:04.440 Yeah. And then the T wave is the repolarization.

00:14:07.080 So you're basically, the R wave is giving you, if my memory serves me correctly, I'm sure there's

00:14:10.840 a cardiologist listening who's going to scream right now. But that's the peak electrical signal

00:14:15.900 of the contraction of the ventricle.

00:14:17.960 Exactly. Yeah. I'm not a cardiologist either, but from my memory, that's correct as well.

00:14:21.820 But yeah, you're getting this exact electrical signal that's showing us where that peak is

00:14:26.580 happening. And because it's a high resolution and it's electrical, we can pick that out pretty

00:14:30.680 easily, especially the more leads you have, the more you're going to be able to get that.

00:14:34.380 The difference between that and an optical sensor.

00:14:37.240 And tell me, by the way, if I'm wearing a polar chest strap, which is what I wear when I'm on my bike,

00:14:43.680 I have a chest strap. How is the fidelity of that compared to an EKG?

00:14:48.300 It's pretty close, honestly. It's very close. As far as picking out the actual

00:14:51.640 peak of the R wave, it's going to be within a millisecond, which is more than enough. Now,

00:14:55.980 obviously, if you have a full six late EKG, you're going to get even more, but you don't need it for

00:15:00.760 HRV as long as you can identify that peak of the R wave precisely within one or two milliseconds of

00:15:06.820 what it's actually at. And that's where the gold standard is, is it's from the ECG.

00:15:10.460 And just to give folks a sense of that, a millisecond, a thousandth of a second

00:15:13.560 is the unit that HRV is typically being measured in. So if a person is looking at their HRV and they're

00:15:19.120 seeing a number that says 60 milliseconds, you're saying with a chest strap, you would put a plus or

00:15:26.480 minus of one or two milliseconds on any reading that comes out?

00:15:30.300 As long as it's a good chest strap. Now, the caveat is good skin contact and those sorts of

00:15:34.760 things. If it's moving around or it's not in the right place, you can lose some of that, which of

00:15:38.480 course, you have a better chance of getting the signal correct with actual electrodes. That is really

00:15:43.340 the gold standard and that's how it was done for decades. And that's how most 90 plus percent of

00:15:48.860 the research has been done with either the EKG or with chest straps, because that's really been

00:15:54.360 the gold standard of how it's measured. The use of these PPG or optical sensors really has only been

00:15:59.120 the last five, six years they've been around and traditionally their accuracy was just questionable

00:16:05.500 when it came to it. And they don't get the same electrical signal. They're measuring changes in

00:16:10.480 blood volume through the skin. Basically, electrodes shine the LED light down into the skin. It reflects

00:16:15.340 differently based on the blood flow flowing through the arteries below it. And so you're getting the

00:16:20.400 pulse and they actually call it pulse rate variable. It's not really heart rate variability. If we want to

00:16:24.480 get technical, it's pulse rate variability, but it's showing us the same thing. It's showing us that cardiac

00:16:29.480 cycle. Now, there seems to be a big difference between the wrist and the forearm. So on my bike,

00:16:38.880 if I'm riding indoors, well, actually I'm doubling up. So I'm wearing, people are going to be like,

00:16:44.900 what is wrong with this Peter guy? He has so many stupid devices. It'll all come to full circle 1.00

00:16:49.180 through this. When I'm on my bike indoors, if I'm outdoors, I'm just wearing my polar chest strap

00:16:54.200 because it pairs perfectly with the system, with the bike system I'm using outdoors.

00:16:59.740 Indoors, I ride with my Morpheus chest strap and my Wahoo optical sensor. The reason is I'm using

00:17:08.020 two different programs. The Wahoo sensor on my forearm, which is optical, is pairing with my

00:17:14.420 computer and that program I'm using there in erg mode. But the reason I'm using the Morpheus chest

00:17:21.660 strap is I'm using the Morpheus program on my phone. But the reason I bring all that up, Joel, is to say

00:17:27.920 they're perfectly in sync. The chest strap, the gold standard, and the optical sensor on my arm,

00:17:34.540 never off by more than a beat. And I can see them in real time concurrently. Conversely, when I'm

00:17:41.040 rucking, when I'm putzing around, I wear a Garmin GPS watch that measures heart rate. It's a very

00:17:48.460 high-end watch. It's about a $700 watch. It is categorically a piece of garbage. I would call it

00:17:54.920 a random number generator for heart rate. It can't come close to estimating my heart rate. There are

00:18:01.280 times I look down and it says I'm at 170 beats per minute when I know I'm below 100. Conversely,

00:18:08.500 there are times when I'm probably at 160 beats per minute and it says I'm at 110. So as far as I can

00:18:15.800 tell, it serves absolutely no purpose. Occasionally it's accurate, I'm sure, but it's so inaccurate so

00:18:21.980 often that I would never rely on it. I'm using it for GPS. I'm almost annoyed that it's a feature that

00:18:27.980 is there. They're both optical sensors. Why the difference?

00:18:32.500 Yeah. One is location, as you mentioned. To get a good resolution, you need good blood flow below

00:18:38.280 the surface and you need the lack of movement. The biggest problem with PPG sensors, optical

00:18:43.220 sensors as a whole, is they get what are called motion artifacts. And any kind of movement starts

00:18:48.160 introducing noise into the signal. Because again, we're not getting electrical signal. We're just

00:18:52.600 getting this blood flow going beneath the surface that we're using the LEDs to detect for heart rate.

00:18:57.040 When you start moving around, you get lots and lots of motion artifacts and it just becomes

00:19:01.740 much more difficult for those sensors to detect it accurately, particularly in like acyclical

00:19:06.240 movements. Anything where your arm is moving around at random, higher heart rates, darker skin colors,

00:19:11.540 lots of things throw off PPG sensors.

00:19:13.840 So my darker skin is obviously a disadvantage, presumably.

00:19:16.840 Just in general, tattoos, all of these things. So optical sensors in particular struggle with higher

00:19:22.360 intensities, higher movements, higher heart rates, all those sorts of things.

00:19:25.880 Now my optical sensor on the bike, even though admittedly I'm not really moving, my upper body

00:19:30.660 isn't obviously moving. Is it superior because it's less movement or is it superior because it's

00:19:37.180 on much larger blood vessels?

00:19:38.680 Both, right? I think the, yeah. So there's a company called Valencell that we use. It's done a lot of

00:19:42.460 research on this because they produce the sensors and they've looked at any location, bicep, arm,

00:19:47.040 calf, all of the above. Even your wrist, you have bone movement. Even if you're not really moving,

00:19:51.920 your wrist can still be flexing and extending and that, in just that wrist movement will cause

00:19:56.280 motionary effects. You get much cleaner blood flow on the forearm. You get just much less

00:20:00.580 movement and torsion as you're moving. So you get just a much better overall signal on the forearm

00:20:05.980 in general than you're going to get from the wrist. They've looked at accuracies of garments and whoops

00:20:09.960 and you don't get very good accurate data at all as you've seen when you're doing exercise. And even

00:20:15.220 when you're doing somewhat cyclical exercise, you can still get, as you've seen, completely garbage

00:20:19.840 numbers that make no sense because the sensor just can't pick up the blood flow very accurately.

00:20:24.620 And like I said, tattoos, dark skin makes it far worse. So in general, the desk trap is always going 1.00

00:20:31.240 to be the gold standard. But if you're going to wear an optical sensor, the forearm where you can get

00:20:34.640 good blood flow is going to be by far the best place to be able to put it. And you can even

00:20:38.640 manipulate where in the forearm you tend to get the best signal, the best results.

00:20:43.020 Yeah. I don't know if I'm doing it correctly, Joel. I tend to apply it right beneath the

00:20:47.560 antecubital fossa where I know the artery is running.

00:20:51.200 Yeah, that's what you want.

00:20:52.100 And I'm sort of like assuming that I'm going to get the best signal there. And I also put it on

00:20:56.300 pretty snug. I mean, I was just doing that, not thinking about motion artifact, but it sounds like

00:21:00.640 that you would encourage that.

00:21:02.280 You want a good enough skin contact that I can read. You don't want to smash it in there.

00:21:05.340 I'm not using it as a katsu band. 0.99

00:21:07.060 Yeah, exactly.

00:21:07.840 I'm not doing BFR with my...

00:21:09.280 You should get a forearm pump and you probably have it a little too tight.

00:21:10.800 So it's almost like there's really three. I know you said there's two buckets. You could really say

00:21:14.180 it's anything on the chest, amazing gold standard.

00:21:17.820 Always going to be good as long as...

00:21:18.700 Anything on the forearm. I think we've established if you do the forearm right,

00:21:21.940 at least heart rate to heart rate is comparable. We'll talk about the HRV variability. And then

00:21:27.460 anything below the forearm is sort of nonsense.

00:21:30.940 It's not great. It's definitely not great, especially if you're lifting weights,

00:21:33.860 if you're doing interval training, if you're doing anything high intensity, you're doing

00:21:36.460 change of direction. It's garbage. You just get very poor data and I wouldn't rely on it. 0.99

00:21:41.760 It might be accurate sometimes and sometimes it'll be way off. The one that seems to be,

00:21:46.160 I wouldn't say accurate, but the most accurate, the less accurate is the Apple Watch. And I think

00:21:50.660 what they're doing, since they can actually have ECG, I think they're just interpolating a bunch

00:21:54.900 of data. So when they see bad data, they just kind of replace it with what they think the data

00:21:59.060 actually should be.

00:22:00.480 I see. So you're saying the Apple Watch might be a step ahead of other wrist-based devices based on

00:22:06.320 sampling an algorithm.

00:22:07.980 You can detect when the junk data is there. You don't have to display it. The other ones do.

00:22:11.760 But they have enough previous data to know that your heart rate didn't go from 110 to 160 in two

00:22:15.860 seconds. So I think they'd start building the algorithm out to interpolate that.

00:22:19.920 Yeah. It's odd that Garmin tolerates that, for lack of a better word, because I'll see it do that,

00:22:24.760 right? I'll see it go from 100 to 150. And I'm like, that's not even physiologically possible.

00:22:29.820 Why wouldn't you sample that out or ask a second order question? Yeah. Great point.

00:22:34.240 Okay. Let's now talk about the very confusing subject of how one calculates heart rate variability,

00:22:44.220 because let's again reiterate what we're talking about. So if anybody has seen an EKG,

00:22:50.720 everybody watching us has, you've got your little PQRST, and you just line up a strip of those.

00:22:58.920 And you imagine you were doing this in the olden days, you'd have a set of calipers. You'd literally

00:23:03.420 measure across R to R to R to R. So let's pretend we have a minute's worth of data, Joel,

00:23:10.060 and a person's heart rate is, they're laying down and resting. So they're at 60 beats per minute.

00:23:16.080 So the approximate beat to beat interval is one second.

00:23:21.360 On average, sure.

00:23:21.960 Yeah. On average, or 1000 milliseconds. What's happening at the physiologic level that makes it

00:23:30.020 such that there is variation? And how is that measured and calculated from the raw data? And

00:23:37.940 let's start with the gold standard and assume you have an EKG.

00:23:40.900 Yeah. As you mentioned, you're starting with this gold standard of, okay, we can accurately pinpoint

00:23:45.700 where are these R to R intervals. And so we pull out what are called the R to R intervals,

00:23:50.160 surprisingly, and we'll plot those. Now, from there, you do what's called correction. Basically,

00:23:56.180 you have to filter data there for filter ectopic beats, which are beats that don't actually arise

00:24:00.480 in a single atrial node. You fill out if there is any noise in the signal or anything like that,

00:24:04.620 and you end up with this clean set of RR intervals.

00:24:07.960 So let's say I gave you 60 of them. And it's again, it's a person who's at rest. So on average,

00:24:13.540 it's 1000 milliseconds between them, but I'm going to give you 60 numbers that vary from 900

00:24:20.340 to 1100 milliseconds. So this is where things get interesting, because when we talk about HRV,

00:24:26.980 we just usually give a number and that number can be different. But a better way to think about HRV is

00:24:31.880 just a framework to assess variability, because there are multiple ways to calculate that. There's

00:24:37.280 one category called time domain, where we literally just do some math. The most common one is RMSSD.

00:24:43.540 Root means successive squared differences, where they just do some basic math, and they get that

00:24:47.920 number of milliseconds of RMSSD. There's SDNN, there's PNN50, there's all these different

00:24:54.200 column time domain, where they just are taking that time series, doing some math on it, and giving you

00:24:59.600 a number that represents the average variability. So let's talk about the RMSSD, because it appears

00:25:04.380 to be the most common one. It is most common for multiple reasons. What we're, again, measuring is

00:25:09.600 that average variability across that time span. And what that represents is the input of the vagus

00:25:16.260 nerve, the parasympathetic system, and its input into that sinoatrial node of the heart. Because

00:25:21.720 fundamentally, the autonomic nervous system is governing that heart rhythm, and primarily what

00:25:26.200 happens at rest is it's that parasympathetic system via the vagus nerve. And the way that it

00:25:30.860 works is it's innervating that sinoatrial node in the heart, and it's pulsing in beat with respiratory

00:25:36.380 processes. So as we inhale, that vagus is inhibited, and you get kind of this acceleration

00:25:43.140 of heart rate. Actually, I should back up. If you were to cut out the autonomic nervous system,

00:25:47.940 you'd have roughly an intrinsic heart rate of about 100 beats per minute, somewhere in that range.

00:25:52.440 Let's back up even a little further, Joel. I think there's a lot that you and I would take

00:25:55.120 for granted here. So sort of nervous system 101, we have two nervous systems, broadly speaking. We have

00:26:01.600 one that's under our control, and one that is not. Most of what you and I are doing that people can

00:26:07.380 watch, the movement, speaking, all of these things, that's voluntarily under our control.

00:26:14.220 But what most people can't see when they're looking at themselves is how many things are happening

00:26:18.700 without any input. And thank God for that system. That system happens to be called the autonomic

00:26:24.540 nervous system. Without it, we would forget to breathe, and we would die. Our heart would stop

00:26:29.600 beating. So all of these vital functions from respiration to heart beating to regulating blood

00:26:35.600 pressure to digesting have to happen via a nervous system that we never think about. That system's

00:26:43.160 further subdivided into the two terms you've already brought up, a sympathetic system and

00:26:48.840 a parasympathetic system. And you've already alluded to one of the most important nerves in

00:26:53.480 that parasympathetic system called the vagus nerve, which is a cranial nerve, so originates from a

00:26:58.780 very primal part of the brain. And we won't necessarily get into all the neurotransmitters

00:27:03.720 involved in these things, but what you're basically describing is that the heart is under the

00:27:10.740 influence of both of these.

00:27:12.840 An example that gets to your point is after a patient has undergone a heart transplant,

00:27:18.260 as an extreme example, that vagus nerve is transected. Their heart is no longer under that control,

00:27:25.080 and therefore it's just going to have...

00:27:27.840 And you would see a heart rate variability of basically zero in that scenario,

00:27:31.020 because it would just be like a metronome.

00:27:32.400 Yeah. Okay. So didn't mean to interrupt, but I think that might be just helpful context for

00:27:36.400 people to sort of understand what we're talking about, which is you're talking about, even though

00:27:40.120 that person's heart is beating at 60 beats per minute, there's still a very fine interplay

00:27:45.840 between what the sympathetic nervous system is doing and what the parasympathetic nervous system is

00:27:51.740 doing.

00:27:51.980 Yeah. And we should probably even back up a little bit more. The whole reason that we need

00:27:55.960 this autonomic nervous system is to keep us physiologically within these normal ranges that

00:28:01.060 we have to be in to be able to produce energy and stay alive, right? So if our blood pressure

00:28:05.320 goes too high or too low, if our blood glucose gets too high or too low, if our body temperature gets...

00:28:10.420 All of these things have to be within physiological norms, and we'd call that homeostasis, that the

00:28:15.060 internal environment has to be controlled at all times, regardless of the external environment.

00:28:19.920 So whatever temperatures we're in, whatever we're eating, whatever we're doing, we have

00:28:25.120 to be able to regulate internally and stay within these physiological norms that are necessary

00:28:29.420 for survival. And that fundamentally is what the autonomic nervous system is doing. It's

00:28:33.900 keeping us alive, and it's trying to match the internal demands with whatever we're trying

00:28:37.760 to do, given the external environment. So like you said, people have heard of probably

00:28:42.460 these two branches, the sympathetic, the fight or flight, or the parasympathetic, which

00:28:46.880 people call rest and digest. Now, that's a good terminology to understand, but it makes

00:28:51.120 us only think of the sympathetic.

00:28:52.080 Yeah, it's not nuanced enough.

00:28:53.560 It's not nuanced enough. It also makes us think the sympathetic isn't doing anything

00:28:56.220 unless...

00:28:56.660 Unless you're under stress.

00:28:57.820 Right? It's not really like that. These things aren't binary. They're not switches that

00:29:01.560 turn on or off. A better way to think about these is dials that the brain is constantly

00:29:05.680 manipulating. And fundamentally, what the autonomic nervous system is doing is twofold. One is

00:29:11.300 it's sensory. A lot of information has to go up to the brain to process what the internal

00:29:16.440 environment status is. And then the brain has to make decisions and push motor action down

00:29:22.100 to the different organs to make sure that they're doing what they need to do, given the

00:29:25.700 state of the body, given its external relationship with the world. So fundamentally, the more we can

00:29:31.460 regulate our internal environment and match the demands of our external environment, like

00:29:34.980 the healthier we're going to be. We're going to be more adaptable. We're going to have better

00:29:39.500 overall function. We probably would just say it's broadly better health. And so the interplay

00:29:44.400 between that sympathetic and that parasympathetic and making sure they can do their jobs

00:29:48.880 appropriately is a really big piece of making sure that our bodies are going to stay healthy

00:29:54.240 as we age. Because I would say, fundamentally, if we look at aging as a whole, we lose adaptability.

00:29:59.960 We lose the ability to respond to workouts as quickly. We become more likely to become injured.

00:30:05.140 When we get sick, it takes longer to get over that. And that's just the body's ability to regulate

00:30:09.520 itself, decline with age. So anyway, with that said, at rest, we should have very little sympathetic

00:30:16.580 activity going on. And we can talk about this in terms of waking versus sleeping. Those are

00:30:21.320 different things. Yes, I want to talk about that. We have a pretty low level of sympathetic just

00:30:24.980 sitting down or laying down. And at rest, that parasympathetic dial is going to be higher because

00:30:30.280 we don't need this additional energy that the sympathetic system can drive.

00:30:34.120 So at rest, we're primarily measuring that parasympathetic input into the heart. And as I

00:30:41.100 mentioned, it turns on and off with our respiration. It's called respiratory sinus arrhythmia. And as we

00:30:46.500 exhale, that dial turns up just slightly. And as we exhale, it turns down just slightly. But mostly

00:30:52.880 what's happening is we are inhibiting that bagel input as we breathe in, and we're letting it function

00:30:59.400 correctly or not correctly, but we're disinhibiting it as we breathe out and as we exhale. And so you're

00:31:04.560 seeing just this pulsation type effect of that vagus nerve on the heart rate accelerating and then

00:31:11.680 slowing down and then accelerating and then slowing down. So you're seeing that input pulsing with our

00:31:17.460 respiratory cycles. And so when we measure HRV, regardless of how we do it, we're ultimately trying

00:31:23.480 to understand that tone, we call it vagal tone, that input of that vagus nerve into the heart rhythm.

00:31:29.540 And that's what we're trying to then gauge as a functional marker of what our autonomic nervous

00:31:35.280 system, specifically the parasympathetic nervous system is doing. How is it responding to the world

00:31:40.680 around us? How is it responding to what we've done in the last 24, 48 hours? What is its resting tone?

00:31:46.420 How much input does it actually have? And from that, we then try to gain all the other insights we can

00:31:51.040 talk about. And then just to close the loop on the measurement thing, most people are using devices

00:31:56.220 that are probably calculating the HRV on the RMSSD algorithm. Yeah. Most of the commercial ones.

00:32:04.320 Yeah. Yeah. It's a transformation that's basically run on the data. And if my memory serves me correctly,

00:32:08.660 I mean, we could figure it out, right? Root, mean, square, or the standard deviation.

00:32:11.840 Successive squared. Successive squares of standard deviation. So you're basically

00:32:14.960 going to say average or mean value is X, standard deviation is this, and then you probably do a

00:32:20.920 sum, square, square root of. Exactly. Yeah. You get your number. Now, the only one I would say

00:32:27.460 that's different is Apple Watch, actually. They use what's called SDNN, which is just the standard

00:32:32.320 deviation of the B2B intervals. Why they do that, I'm not sure. That's one that's historically been 0.97

00:32:37.620 used medically, and they'll usually measure it for 24 hours. And they'll just kind of look at like,

00:32:41.720 do you have any autonomic variation? Does the parasympathetic system function well

00:32:45.740 at all? And it's kind of a gross measure. It's not nearly as nuanced because we're not measuring

00:32:50.300 vagal input at a particular time. We're just measuring across longer periods of time. And

00:32:54.600 maybe that's why they did that. It seems to me that that would also introduce a bit of noise

00:32:58.560 because you're combining being at rest with being active. Yeah, exactly. And you would,

00:33:03.420 I don't know if it's the right word, but you'd be penalizing people for being more active because

00:33:07.220 the more you exercise, the more sympathetic tone you have during exercise, the more you're crushing the

00:33:12.680 variability. Yeah. And what's interesting is Apple is just kind of measuring randomly for the most

00:33:17.280 part. It just kind of measures when you don't know. And you can do a manual measurement, which

00:33:21.300 we can talk about, and you can actually check it, which is a better way to do it. But for whatever

00:33:25.120 reason, they've just used this metric that nobody else uses. And then they kind of measure it

00:33:29.040 periodically when you don't know what's happening. And so the number you're just kind of getting in

00:33:32.360 there, if you're not actively measuring it, is just kind of like, I don't know where it comes from.

00:33:37.080 So a couple of things that I remember from a AMA that I did on heart rate variability a couple of years ago,

00:33:42.420 and we'll link to it here in the show notes so that people can go back if they want a real primer

00:33:46.500 on HRV. The reason we did sort of an AMA on that was a lot of people had questions about it. Frankly,

00:33:52.040 I don't think we went into nearly this level of detail about it. We talked much more about the

00:33:57.040 mortality data and things of that nature, but that was one of the first things that stood out.

00:34:01.560 Two things I remember more than anything, Joel. The first is there was a relationship between,

00:34:06.460 in the research literature, what was measured as HRV, and we should talk about what that means,

00:34:12.120 and all-cause mortality and even disease-specific mortality. And the second thing that really

00:34:17.500 stands out is a graph that I'll never forget that shows on the x-axis age, on the y-axis HRV,

00:34:25.840 and what the curve looked like. And I couldn't believe how steeply it declined, right? And I think

00:34:33.100 what it was plotting, if I'm not mistaken, was kind of mean or median HRV with a band of,

00:34:40.000 call it the 80% or interquartile range or something like that. But it was an unmistakable

00:34:45.380 trend, which is like a 50-year-old's HRV is less than half of a 15-year-old's. And it just keeps

00:34:52.280 getting further and further crushed as we go down. I suppose that speaks to what you said earlier,

00:34:56.920 which is one of the hallmarks of aging is this sort of lack of resilience. And we see it on every

00:35:02.140 level, but this is just a very notable example, which is even at the level of the autonomic nervous,

00:35:09.900 system. We lose the ability to recover from insult. And life is an insult. Everything in life is an

00:35:16.520 insult.

00:35:17.060 The world around is this insult to insult time. It's just, we can respond much better to it as

00:35:21.020 we're younger.

00:35:22.100 Yeah. Do you have a sense, by the way, of what it is physiologically at the cellular level that

00:35:27.480 is resulting in this profound reduction in HRV as we age?

00:35:32.720 They've looked at this and I don't know that we have a great answer for the exact physiological

00:35:36.640 mechanism. We know it's tied to mitochondrial density, mitochondrial function. We know it's

00:35:41.480 tied to elements of the immune system. We know it's tied to hormonal status. And we obviously see

00:35:45.400 decreases in all those things as we age.

00:35:48.280 You just have to wonder which ones are causal and which ones are the response.

00:35:51.900 I don't think we know that, but we definitely know that we can increase our age of year. We can

00:35:55.580 at least prevent the decline most effectively through cardiovascular fitness. We see people with

00:36:00.060 higher VO2s have higher mitochondrial function. They have higher VO2 that leads to,

00:36:04.480 or at least correlates with greater HRV. So we know that cardiovascular fitness in general

00:36:09.040 seems to be the most closely tied to average HRV. There's also a pretty strong genetic component,

00:36:14.420 which we can't ignore. We can talk about that. But yeah, if you look at the Hallmarks of Aging 0.99

00:36:18.760 paper, which I'm sure you've seen, they kind of take these buckets of things, right? They say,

00:36:22.340 oh, as you age, you get dysbiosis, you get deregulated nutrient sensing, you get senescent cells,

00:36:27.880 you get stem cell exhaustion, you get mitochondria. They list all these things that happen as we age,

00:36:32.020 and they kind of look at this prism of what's the output. And if you read that, they say, okay,

00:36:35.900 the output to the major things are loss of resilience to homeostasis and lack of a stress

00:36:41.500 response that's appropriate given the world around us. So yeah, where that cause and effect and which

00:36:45.880 one's causing the other is tricky to say, but I think fundamentally aging is this progressive

00:36:50.640 loss of adaptability. And there's multiple pieces to that obviously, but we're measuring that,

00:36:55.960 as you've mentioned, is one of the things that we want to gauge of HRV is how much of that

00:37:00.340 resilience of that adaptability are we losing as we age. And that's something we can influence through

00:37:04.640 lifestyle and training and everything else that we're trying to do here to prevent that slowdown.

00:37:09.780 One of the things in that AMA that I didn't get a great answer to was how much genetics played a

00:37:17.600 role on this. But from our patient population, because even though that's not a huge N, we've got 0.93

00:37:25.440 years and years of seeing these data in patients where every single one of them is using some sort

00:37:31.860 of device. By the way, it's even devices we haven't talked about. Like if you look at really high-end

00:37:36.320 things like mattress covers and things like that, like the Eight Sleep will now measure that quite

00:37:40.000 accurately. From every form of wearable and out, you've got endless streams of data and there's an

00:37:46.540 unmistakable difference between people. There are some people who, and let's just talk about this in

00:37:53.120 RMSSD, we should talk about the other numbers because you have to do this apples to apples.

00:37:57.640 But if we just talk RMSSD, I've got patients who live at 100 and a good day for them, a good day for

00:38:04.140 them, quote unquote, they're at 120 and a bad day for them, they're at 85. But if you follow them for

00:38:11.080 five years, their average HRV is going to be 100 milliseconds. I've got other patients whose average

00:38:17.280 HRV is 15 milliseconds and a good day for them is 25 to 30 and a bad day for them is 10. How could that

00:38:25.280 be explained by something other than genes?

00:38:27.720 It isn't. I mean, I've looked at a bunch of this research just to understand, and it's all over the

00:38:31.440 map, depending on what paper you're looking at, depending on which metrics they calculated. They say

00:38:35.420 genetics is somewhere between like 15 and 70-something percent of HRV. There's just such a wide range in the

00:38:40.160 research of what you see. Where the exact number falls, I'm not sure, but you definitely see a very strong

00:38:45.580 genetic component to it. Why? I don't think we truly understand that. But as you mentioned, I see

00:38:49.920 people who don't work out at all, and they come into the gym or they, whatever, and you look at

00:38:54.040 their numbers and you're like, you have a very high HRV that you would not expect because you clearly

00:38:58.200 don't have a very high level of cardiovascular fitness. But I'll say kind of as a whole, if you

00:39:02.500 start talking to those people, they tend to have a healthier family history. They tend to have

00:39:06.140 better health markers. I think there's something to that, and that higher HRV probably still

00:39:11.100 correlates to a health benefit, even if it doesn't necessarily come from exercise-derived

00:39:15.460 means. It's just a genetic thing that they have that probably confers in benefit.

00:39:20.020 Would you put HRV in a comparable bucket to VO2 max in terms of the following? Amount of it that

00:39:29.720 is genetically determined, amount of it that is modifiable, and the role it might play in

00:39:36.760 understanding overall health status. So for VO2 max, I know the answers to all those questions,

00:39:41.580 right? There is a genetic component. It's not huge. It's probably closer to that 15%

00:39:46.800 than 70%. It's highly modifiable, but difficult. And the fact that it's highly modifiable,

00:39:55.180 but difficult to modify is why I believe it is, and you've probably heard me say this,

00:40:00.660 the single greatest predictor we have of mortality. And if there's a better one out there, I'd like to

00:40:05.800 hear it, but I haven't seen one. And I think that's because I always talk about VO2 max as like

00:40:10.660 the integrator of so much hard work. You can't cram for the test. If your VO2 max is in the top 1%,

00:40:17.460 you weren't born there, you blood, sweat, and tears your way to that, and all that work does so much

00:40:24.600 good for you. Okay. So let's use that framework to evaluate HRV. How genetic is it? How modifiable is it?

00:40:33.420 And are the modifications you have to put into it then speak to, hey, if somebody improves their

00:40:39.480 HRV by 50%, how confident are we that that moves the needle in terms of what actually matters,

00:40:45.340 which is not the silly number, but actually the outcome of their life?

00:40:50.260 No, I think it's more genetically based just from what I've seen and probably a bit less

00:40:54.180 modifiable. I would say it's less predictive in the sense that if I have somebody who's got an HRV of

00:40:58.860 110 to use your example or whatever, and then I have someone who's got a VO2 max, I know is,

00:41:04.200 I don't know, 70, I can pretty well know that person with a VO2 max of 70 is pretty aerobically

00:41:09.380 fit. And they've lived a pretty solid lifestyle and had to done the work and the training to get

00:41:13.600 that level. I'm more confident in that that person's longevity be fitted and affected from

00:41:18.440 that, that I'm confident in somebody who has no workout history that just has a high HRV.

00:41:23.300 I don't know that I can say the same confidence at all, just because they have a higher HRV,

00:41:27.200 that they would have the same prognostic value in all-cause mortality. So it's a metric that we

00:41:33.280 aren't gauging output from. We're just measuring this internal physiological state, and I think

00:41:38.000 that confers benefits to someone who has higher HRV, but I can't necessarily just look in that

00:41:42.900 number and say, oh, this person's really healthy or this person's really fit necessarily, because

00:41:47.180 you do see that much stronger genetic component. Now, if I see a coupling, if I see someone with a

00:41:51.560 higher VO2 and I see higher HRV, chances are that's a reflective of all the things we just talked

00:41:57.040 about a healthy lifestyle and a lot of hard training and the physiological changes that

00:42:01.120 come as a result of that, and we're more confident that those numbers are going to line up with

00:42:05.780 all-cause mortality.

00:42:07.460 So a nice way to interpret that, Joel, would be the following. As much as people are worrying

00:42:11.580 about their HRV and people really do worry about it, you should worry more about your VO2 max,

00:42:16.860 because you have more control over it, and it's a better predictor of all-cause mortality.

00:42:21.800 I think that what gets measured gets managed, right? And because HRV is so ubiquitous and

00:42:26.560 it's so spit out, and basically you're at the point now where if you go get a Starbucks,

00:42:29.820 they'll tell you your HRV that they've somehow inferred from the pressure your lips put on

00:42:33.820 the cup. I'm being facetious. Everybody is inundated with these data, and it is creating

00:42:38.480 a lot of stress.

00:42:39.760 Yeah. I think we want to look at output measures. VO2 max is the best output measure. We can look

00:42:43.960 at something like heart rate and zone two.

00:42:46.160 Heart rate recovery.

00:42:47.160 Heart rate recovery. We can look at actual output measures, because at the end of the day, I fundamentally think,

00:42:51.800 that's what matters. As we age, we need to be able to continue to move, and we need to be able to

00:42:56.340 continue to be able to respond to our environment around us, and output is where we can see those

00:43:00.900 metrics. What's the metabolic cost for us to move around? If we can maintain movement as we age,

00:43:06.360 we can be highly active. If you look at people that you know around you, they're healthy and

00:43:09.860 older. A lot of them, they're very active. They move around. They have hobbies. They have friends.

00:43:13.080 They're social. They do things they love, and that's a big part of keeping them healthy and resilient.

00:43:17.540 If we don't have the metabolic capacity to move, we decline a lot faster. Again, VO2 max and heart

00:43:23.340 rate at different speeds, that correlates the ability to move and maintain that as we age. That's

00:43:27.800 far more predictive, I think, than just an internal metric that is important, but does not have the

00:43:33.580 same predictive power.

00:43:35.120 Do you have kids?

00:43:35.800 I don't.

00:43:37.020 Anybody who's got kids, especially young kids, will appreciate this comment, but I've become so much

00:43:42.240 more cognizant of a metric I would love to introduce to the world that ties into what

00:43:48.900 you're describing called spontaneous movement. I'm an old guy, and even though I'm fit for my age,

00:43:55.480 I don't waste a lot of movement. I'm already at that stage in my life where I actually think of

00:44:00.880 myself as quite lazy. I love to exercise, and obviously, I'm not lazy when I'm doing that,

00:44:05.140 but if I'm walking through the airport, I'm just walking. I will use the stairs and not the

00:44:12.400 escalator and all that stuff. I get it all, but if you're with my kids and my two boys are six and

00:44:19.480 nine, the amount of spontaneous explosive movement is something I don't remember doing as a kid,

00:44:27.940 although I'm sure I did it too, but it's really a remarkable thing. We also have a puppy. We have this

00:44:33.900 puppy that also is bouncing off walls, but when you see older dogs, that's done. A 14-year-old dog

00:44:42.120 is not, even if it's in good health for its age, it's not bouncing off the walls, whereas that puppy

00:44:47.720 can't stop moving. The same thing, when I look at my boys, everywhere we go, they have to race.

00:44:54.260 Everything is a race. They're sprinting there. If we're walking somewhere, they're doing sprints to

00:45:00.800 and from us the whole time. I just think, A, there's something beautiful about that, but I think

00:45:06.540 it speaks to this idea of youth. Youth is about movement. It is converting the chemical energy of

00:45:14.400 our food into the electrical energy that powers muscles, and spontaneous locomotion seems to be

00:45:20.900 this. I don't know. At some point, I would love to know, is there a way to take that as another output

00:45:26.800 metric? Take a GPS on somebody. Exactly. What is the drive to spontaneously move for no apparent

00:45:32.660 good reason? We talk about we lose HRV, but we also lose sympathetic drive too. We lose some of that

00:45:37.860 ability to turn that sympathetic dial up as we get older too. Probably it's as we've lost both of

00:45:43.260 those capabilities, the ability to turn that sympathetic dial up and crank out more energy

00:45:46.760 and produce adrenaline and cortisol. Is that what you think explains the fall in maximum heart rate?

00:45:52.220 Part of it is loss of contractility of the heart, loss of contractility of the musculoskeletal system,

00:45:56.380 loss of hormonal release as a result of the sympathetics. I mean, you're just losing,

00:46:00.900 again, this adaptability, this ability to turn those two dials as necessary to meet whatever

00:46:06.160 demand you're placing in the body. We can't turn that sympathetic dial up as much. We don't have

00:46:11.400 that spontaneous energy that you just described to get up and sprint because that dial goes way slower

00:46:16.620 and it probably can't go up as high. Yeah. So it's like we were born with a zero to 10

00:46:20.980 rheostat or dial on both of them. And as you age, that 10 goes to a nine, eight, seven, six,

00:46:27.540 five, and you can still move them, but you just can't move them as much.

00:46:31.260 Yeah, 100%. I would call it autonomic range. And that really kind of represents what is our body

00:46:35.800 capable of from an energetic standpoint? How quickly can we turn that dial up? And then conversely,

00:46:40.840 how quickly can we turn that dial back down and crank up that parasympathetic side to restore

00:46:45.380 homeostasis and get our bodies back to normal? And if you look at, I've looked at a paper where

00:46:49.920 they looked at different Navy divers that were going through this qualification school, which

00:46:54.300 is kind of their equivalent of SEAL training and Hell Week and all that. And they tried to pick out

00:46:59.340 what are the variables that separate the people that are really good at this and succeed and make

00:47:03.300 it through versus the ones that don't. And they measured HRV throughout the process.

00:47:07.520 And essentially, they find what I just talked about is this autonomic range where they could really

00:47:10.980 crank up the sympathetic system when they needed to, and then turn it off as soon as the stressor

00:47:14.800 over and respond in the other direction with a much higher parasympathetic response. That ability

00:47:19.840 to use those dials quickly and in the right combination seems to be a really key thing to,

00:47:25.200 again, adaptability. And as we said, if you age, if both of those dials lose their range and they lose

00:47:30.700 their coordination to some respect, then we have much less resilience, much less adaptability.

00:47:35.920 And we should mention too, as you pointed out, the vagus and the sympathetics, they do influence

00:47:40.020 behavior in a lot of ways. There's a whole thing that's outside my lane in the psychosocial

00:47:44.480 aspect of this. And you can look up the polyvagal theory by Stephen Porges. But fundamentally,

00:47:49.520 the brain regulates emotion through autonomic function in some capacity. And the vagus is

00:47:55.040 related to social behaviors. It's related to cognitive control in different scenarios. It's

00:48:00.660 related to all sorts of stuff. They call it fight, flight, freeze, all these things related to how our

00:48:06.000 autonomic nervous system is influencing our emotions. And if we don't have the autonomic range,

00:48:10.480 we probably have less drive to get up and move around as a result of that, as those nerves in

00:48:15.240 the autonomic system changes what it can and can't do.

00:48:18.520 I don't think that should ever be underestimated or understated. I think, again, just even looking

00:48:24.280 at a sample size as small as our patient population, I think we see that a lot. I think there's a very

00:48:30.140 clear association between an individual that, if you just look at them from a movement and exercise

00:48:37.040 perspective, has a very difficult time relaxing. And relaxing sounds like such a silly word, but I mean

00:48:44.620 that in a sort of a clinical sense. If they can't let their rib cage down, if they can't properly

00:48:49.260 generate intra-abdominal pressure, if they can't go through a sequence of movements that generates some

00:48:53.840 amount of motor control and compensatory relaxation contraction, I think there seems to be very high

00:48:59.500 association between that and emotional stress and psychological stress.

00:49:03.340 A hundred percent.

00:49:04.260 And actually pain as well. So I think we see chronic pain. And again, you could argue, well,

00:49:09.820 where's the chicken? Where's the egg? If you're in pain, does that lead to more emotional stress?

00:49:14.540 Does that lead to an inability to regulate relaxation within the body, which further exacerbates pain?

00:49:20.220 It's a very vicious cycle.

00:49:22.120 And you see lack of sleep. Older people need just as much sleep, but they have a harder time getting as

00:49:26.220 much sleep. And sleep is very much tied to that vagus nerve and the peristumatic.

00:49:29.500 Nervous system. So again, if we get worse quality sleep, we get less adaptability. It is chicken

00:49:34.200 and egg. But fundamentally, that's why we want to regulate ourselves correctly. And that's where

00:49:38.560 exercise, I think the biggest thing exercise does is improves our body's ability to regulate itself.

00:49:44.020 It improves the use of those dials because we are exposing the body to the aerobic training that

00:49:49.320 we know has some broad correlation to that. And we're giving the body a stress it can adapt to in

00:49:54.100 a positive way if we do it in the right amounts. And that's the caveat there.

00:49:56.880 Before we leave the measurement thing, I want to go on to another one or two of those measurement.

00:50:02.640 At Morpheus, what do you guys use to measure?

00:50:05.180 So we use RMSSD and then we use log natural transform and a multiplier. That sounds like

00:50:09.640 a lot of math. But essentially, if you look at the data of RMSSD and you look at like a normal

00:50:14.880 bell curve, it's skewed. It's non-normal. It's all the way to the left. So you get these normal ranges

00:50:19.080 of 20 to 80 or 100, like you said. But then you can get elite athletes, 180, 200, and see this big

00:50:25.480 bunch of data on the far left-hand side. It's kind of hard to enter for it. So again, I'm not a

00:50:30.060 statistician. I didn't create the formula. But essentially, to normalize this data and make

00:50:34.000 it look more naturally distributed, more like a normal bell curve, you do this log natural transform,

00:50:38.680 you use a multiplier, and Morpheus ends up on a scale that looks more like a 100-point scale.

00:50:43.000 So people with lower HRV are going to be more like the 50-60. People with more moderate, 60-70.

00:50:49.300 Higher, 70-80. Elite athletes are going to be 90 to 100. It's more of a scale that we have

00:50:54.180 a more familiar relationship with, I would say. And the data is more normalized from a

00:50:58.760 standpoint of a bell curve. So it's just an easier to interpret. But that's the reason for it.

00:51:04.360 So do most people, when they start using Morpheus and they're also using whatever other device

00:51:12.100 they're using that's just a purely RMSSD device, how much discordance are they typically seeing

00:51:17.320 between them? It really depends on what they're using. It can be a big difference between the

00:51:21.620 device that they're using and the numbers they're getting. But the trends should generally line up.

00:51:25.760 If you're seeing Morpheus increase as a whole, you should see the other one increase as a whole.

00:51:29.920 You should see the directional change matching, but the actual numbers will be somewhat different

00:51:35.180 depending on where you're at in that spectrum. Let's now talk a little bit about Morpheus.

00:51:39.040 I've alluded to it a couple of times. So it's a product I've been using for about a year and a

00:51:44.720 half, maybe close to two years. But before I talk about my experience with it and why I use it,

00:51:52.100 everything I do, I do for a reason. I'm a very deliberate human being. So there's a very particular

00:51:56.500 use case that is pretty narrow for how I use it. I know that I'm not using it to its full potential,

00:52:02.420 but tell folks what this is about, your involvement in this, and that's obviously how we got to know each

00:52:06.520 other. Yeah. I mean, when I started using the old system, I had to wait for people to come in the

00:52:11.360 gym to measure them. And so I realized I was getting a pretty small snapshot of what their

00:52:15.360 life story was, because I might measure them two or three times a week. Sometimes they come in the

00:52:19.540 morning, sometimes they come in the afternoon. And I realized the limitations of that. This is back in

00:52:23.740 2007 or 8. And so I wanted to create something that people could use their phones and that we could

00:52:28.540 get way more data from. And that was my first system, Bioforest HRV 2011. And that was really one of the

00:52:33.740 earliest HRV apps out there where we could take your phone, you could do a recording, and you could

00:52:38.040 get your HRV on your own. You didn't need to come into the gym. And one of the limitations of that was

00:52:43.560 all I could look at was your HRV. I didn't necessarily have any idea what else was being

00:52:48.100 tied to that. And so I could look at the change and ask you a bunch of questions and maybe try to

00:52:52.420 figure out what those changes were coming from. But I wanted to create something that also tied in

00:52:56.580 training and sleep and subjective markers and other metrics so that as a coach, I just had a more

00:53:02.420 complete story of what was happening. And so that was really the genesis of Morpheus. And I started

00:53:07.260 that in 2016, 2017. So quite a few years ago. And basically what we're doing is we're measuring HRV

00:53:13.760 and then we're tracking activity, sleep, workouts, all that sort of stuff. You can use the Morpheus

00:53:19.100 device for a lot of that, or you can use other devices. So if you're using an Apple watch to track

00:53:23.800 your activity, your sleep, or Garmin, we'll pull that data in. But what we're trying to do is take

00:53:28.600 that HRV, this is something we should definitely talk about. There's a lot of apps that give you

00:53:32.080 HRV. And then you can either say, okay, I'm going to interpret what this means myself. And I can try

00:53:37.720 to figure out what these changes are. It's a physiological metric. Or you can say the apps

00:53:42.380 are giving me a recovery or a readiness or some gauge that's based on that. Now, this comes back to

00:53:50.040 every app's doing this completely differently. We have a lack of standardization across not just the

00:53:54.940 HRV measurement, but then how that information is interpreted to generate recovery or readiness or

00:54:00.900 some sort of number that the person in the app is saying, oh, my recovery is 80% or my readiness is...

00:54:07.120 These are just numbers that we are creating as a way to try to interpret the data. And then some of

00:54:11.620 those apps, I think, do a reasonable job with it. Some of them don't. But I created Morpheus and this

00:54:17.120 idea of the recovery score based on what I had seen using HRV for 20 years or maybe 15 years at that

00:54:22.980 point. It's just a metric of what's appropriate for you on a given day. What is your body more

00:54:27.940 likely to benefit from? And so we should probably talk about, again, what is recovery? What is

00:54:32.800 readiness? Because those are metrics that Morpheus gives you and Ura gives you and Whoop gives you and

00:54:37.680 Garmin gives you like a body battery or every kind of app has their own gauge of that. But then the

00:54:43.060 question is, how accurate is it? What does it really mean? And I think that's where a lot of confusion

00:54:47.560 also comes in because we're trying to take metrics and turn them into something that's not a

00:54:51.680 physiological measure, but something we try to create. Yeah. I mean, look, I used an aura ring

00:54:56.620 for many years, probably haven't used it in a year, year and a half. There's better devices I can use

00:55:02.480 to track my sleep now. And the recovery score, as you said, even if you believe the score is accurate

00:55:08.780 and there's no reason to believe it necessarily is, more importantly, it's not something you can act

00:55:14.380 on. Let's just say you believe the number and you say, okay, my recovery score is 80 today. I

00:55:20.420 certainly believe that if it spits out a 90 versus a 70, your Whoop or Aura, there's probably a

00:55:26.260 difference there. You're probably better off on the day you're 90 than the day you're 60. But how do

00:55:32.120 you operationalize that information? And so when I was introduced to Morpheus, it was actually someone

00:55:38.540 on my clinical team that said, we're having a hard time giving people real instruction around zone two,

00:55:46.840 because most people don't want to do what you do, Peter. Nobody wants to check their lactate levels

00:55:53.060 and go through this. And I get it. I'd never fault somebody for not wanting to do a finger stick every

00:55:58.220 time they do a workout. And truthfully, for some people, just relying on RPE can be challenging.

00:56:05.140 So this person said, hey, look, the Morpheus app, and we'll talk about how it works. I also appreciate

00:56:11.240 how the measurement is taken. The Morpheus app gives you target zones for heart rate every day.

00:56:18.880 And if you use the cutoff between what it's calling zone one and zone two, that's a pretty good proxy

00:56:26.280 for what your zone two is on that day. So I bought the system and started using it. And I should show

00:56:33.900 you the data because I have recorded every single workout I have ever done for the last year and a

00:56:39.680 half. And I record the following. I record the heart rate predicted by Morpheus for what my zone two

00:56:48.080 is. The heart rate I largely end up at by RPE. And sometimes they're close. Usually they're quite

00:56:55.100 close. Sometimes they're quite far. So sometimes Morpheus tells me to be at 138, but I'm kind of

00:57:02.020 gassed out at 131. And sometimes it's the reverse. Sometimes Morpheus says you should be at 133,

00:57:07.400 but I feel fantastic. And I go to 140, what the power was for the interval and what the lactate

00:57:13.140 is. And I have to tell you, Joel, I cannot put in words how impressed I am with that system

00:57:18.600 and how remarkably accurate it is at predicting something that is very difficult to predict.

00:57:26.000 So kudos to you for doing that. What I find amazing are the days when, and I had one of these

00:57:32.240 days a week ago, Morpheus said I should have been at 140 or 139 for zone two. I got on the bike and I

00:57:40.880 did not feel great. And I sort of said, I think Morpheus got it wrong. I'm going to ride to this

00:57:48.920 wattage. And my heart rate was about 132. And I checked my lactate and it was 1.1 millimole.

00:57:54.980 I was nowhere near my limit of where I could have been that day. Now, again, we could get in the weeds

00:58:00.040 on maybe that's fine. Maybe that's all I needed that day. And maybe I should have just been

00:58:03.340 following how I felt. But if I'm really trying to get the right training effect, I was under

00:58:08.900 training a little bit on that day. So I'll kind of pause there because I want to kind of let you

00:58:13.980 sort of interpret what I'm saying. And why is it about that? Because you spit out three zones,

00:58:19.320 zone one, zone two, zone three. And I think you call them recovery and conditioning.

00:58:23.460 Yeah, it's just terminology, right? It's just a way to gauge low,

00:58:25.460 moderate, high intensity, more or less.

00:58:27.120 Does it surprise you? Because I haven't told you this story before in all of our communications.

00:58:31.080 We've never talked about this particular issue. Does it surprise you that the heart rate that is

00:58:35.440 on the cusp between your first and second training zones happens to correspond to this lactate of two

00:58:41.880 sweet spot? I mean, that was really the intent when I designed it. And I think probably that's

00:58:46.400 difference in Morpheus, I would say, is when I designed it, it was as my experience as a coach for

00:58:51.120 many years of testing lactate, of VO2 max testing, of HIV testing. I synthesized a lot of information

00:58:56.980 that I gained and knowledge and insight I gained to create this structured system of low, moderate,

00:59:02.500 high intensity. You can call them whatever you want to call them. That's basically what

00:59:05.440 those zones are. And the biggest thing I realized is as people's autonomic nervous system changes,

00:59:10.940 intensity and heart rate changes. And you learn this over time when you see today 140 is doing this,

00:59:17.720 and tomorrow 140 might be doing that based on changes in the autonomic nervous system.

00:59:22.280 And so when I created the algorithm, it was just based on a lot of data collected over

00:59:26.360 a number of years of what I'd seen in the gym, what I'd measured, what I didn't look back at the

00:59:30.240 data. And so that was the end result of Morpheus was trying to translate changes internally with

00:59:36.120 how we can then turn that into smarter, more accurate, precise training. And that's what you're

00:59:41.120 seeing with Morpheus. So it's awesome to hear that it's accurate for you. And some people,

00:59:44.880 it's always going to be much more accurate than others. But as a whole, Morpheus is by far

00:59:48.860 the best way to translate, again, changes internally with changes we should be doing

00:59:53.760 in the gym. Yeah. I mean, basically for our patients now, we almost never bother with them

00:59:59.400 checking lactate. It's basically RPE, if you can manage it, if you really have a sense of what

01:00:05.520 zone two feels like, great. But if you want some guidance, look, take the Maffetone formula 180 minus

01:00:11.540 your age, great place to start. Once you're getting a little more nuanced, if you want more guidance,

01:00:16.000 use Morpheus and go to that heart rate. Again, tell folks how I get that number every morning.

01:00:21.700 How is it spitting out that number for me? I have to do a measurement, right?

01:00:24.780 Yeah. I mean, you have to measure your HRV and then we're looking at other things that you've

01:00:28.240 done in the past 24 hours if you're recording it. And then based on your fitness level,

01:00:31.840 so we ascertain your fitness level by looking at your resting heart rates, by looking at your

01:00:35.660 average HRV, by looking at some of your heart rate trends when you're training. And we say,

01:00:39.180 okay, Peter's roughly at this level of fitness. Peter's autonomic nervous system is responding in this

01:00:44.400 way, which again, correlates to how it's going to react today. If I'm fatigued, then it's going to

01:00:49.980 take more energy to produce the same level of power output. And then we estimate, like you said,

01:00:54.360 what for you is this cutoff between low intensity, moderate intensity, and high intensity. And a way

01:01:00.340 that people can think about this, I would say, is muscle fiber recruitment. So low intensity is

01:01:05.500 primarily slow twitch muscle fibers doing the majority of the work. And specifically,

01:01:09.800 zone two, we're talking about where they're mostly oxidizing fat as much as possible,

01:01:13.480 kind of this moderate intensity, we're starting to recruit some of those moderate threshold fibers.

01:01:17.440 And we can talk about what that means. But we're starting to recruit some of those faster twitch, 0.74

01:01:21.180 higher threshold muscle fibers, and then higher intensity, we're starting to really recruit

01:01:24.920 all the muscle fibers and the highest intensity muscle fibers. And so as Morpheus is looking at this,

01:01:30.500 it's basically saying, okay, roughly this heart rate, we think this is a level of intensity that is

01:01:35.080 going to correspond to these low, medium, and high. And then it's giving the ability to just

01:01:39.380 plug into Morpheus and say, I want to do zone two. Morpheus says, okay, based on what we've seen,

01:01:44.040 this is where we think your zone two is for you today. And so on the same thing.

01:01:48.200 Yeah. And again, I just want to reiterate, this is why it's very valuable.

01:01:51.620 It's actually giving me the prescription every single day. When I got it, I was a little surprised.

01:01:57.360 You know, I sort of bite at sight unseen. I was just kind of told it's valuable,

01:02:00.820 but I didn't understand the nuance. So the thing shows up and I realize, oh, every morning I need

01:02:05.780 to do a two and a half minute lay down in bed before I get up, still test, measuring my heart

01:02:12.700 rate. And I guess you can do that. Morpheus at the time came with a chest strap and an armband. I

01:02:17.060 think it now is just a chest strap. So you put the chest strap on in bed, you lay there,

01:02:22.140 you answer a couple of questions. So it wants to know how many hours did I sleep the night before?

01:02:26.160 So I pull that data off my sleep tracker. I ate sleep.

01:02:30.820 What's the quality of my sleep? I'm pulling off that as well. And then I think it's saying,

01:02:35.740 how sore am I subjectively and how good do I feel?

01:02:39.300 Yep. The subjective markers.

01:02:40.440 Those are actually reasonably validated markers for training performance.

01:02:44.380 They are.

01:02:44.800 And then I measure, I lay there and just kind of do nothing, meditate usually. And it measures my

01:02:50.000 HRV and heart rate. And then it spits out, here's your HRV, here's your heart rate,

01:02:56.520 here's your, basically your recovery score as a percent. And then here are your target training

01:03:02.740 zones, which then come up again. When you train that day, it's already loaded into the app.

01:03:08.240 So your training zones change with every day. So one of the things that surprised me, Joel was like,

01:03:12.440 wait a minute, I'm used to having to measure HRV over the course of a night.

01:03:17.340 What is the difference between what my aura ring used to tell me by measuring over eight hours of

01:03:24.700 sleep versus this thing that's telling me in two and a half minutes in the morning before I've gotten

01:03:31.160 up? This is probably the most important part of HRV because there is so much confusion in this.

01:03:35.840 We look back historically at all the data that's been used, these all-cause mortality studies and all

01:03:40.360 the different pieces of literature out there, 95% of them are from spot HRV measurements that we are

01:03:47.340 measuring at a specific time. And you're doing this in standardized conditions as much as you can to

01:03:52.540 get a baseline because we want to know where is your number system? Same time every time. Same time

01:03:57.440 every day, the same conditions. Because what I want to know is last 24 hours, you did something

01:04:02.560 yesterday, you did lots of things, I assume. You ate food, you maybe worked out, you maybe had alcohol

01:04:07.600 or maybe you didn't, you did mental stress or maybe you didn't. You put your body in a situation where it

01:04:11.980 had to respond for the majority of the day to do something and then you went to bed. And we want to see the

01:04:16.600 result of that. We want to see this stress and recovery cycle that you went through yesterday

01:04:20.980 because that tells us where your body's at right now. How is it responding? Because we'll look at

01:04:26.000 changes over time and understand how your body is adapting to the world around you. And that's what

01:04:31.100 most HRV has been built on is we measure in standard conditions, we see where you're at today,

01:04:35.940 and that informs us about what happened over the last 24 hours and maybe slightly beyond. And the

01:04:42.040 analogy is if I was going to weigh myself, I'd want to weigh myself first thing in the morning

01:04:45.940 in standard conditions. I wouldn't want to have a meal and then go weigh myself. I'd want to have

01:04:50.400 very standard ways of measuring so I can see the changes because ultimately it's

01:04:54.060 you changing against yourself that's the most informative. So we wake up, we measure HRV,

01:04:59.040 we see where you are, and we see where you were, what your averages have been, what your variations

01:05:02.840 have been. And that tells us where you are today. And that helps us make a decision about

01:05:06.800 what are you ready to do right now? What's the most appropriate for you to do right now?

01:05:10.840 One thing I'll say is if your HRV is high or low, we can talk about what those mean.

01:05:14.320 It doesn't mean that you can't train hard. It just might mean that that might not be the

01:05:18.940 most beneficial thing for you. And there might be a cost associated with that. If I wake up,

01:05:22.480 my HRV is way outside of normal. You can't say, oh, I can't work out today. You can. It's just a

01:05:27.700 question of, is that what your body needs? Yeah. To be clear, I don't think I've, I mean,

01:05:31.620 I don't think I have never once not exercised as a result of what that said. And there have been

01:05:36.640 days when I've had abysmal scores and it's told me like my heart rate range on what it has told

01:05:43.280 me is never above about 141. There've been days it's been as low as 121, which for me,

01:05:48.740 it means like my recovery was 35 or 40%. That's a night when I didn't sleep and something was

01:05:53.940 dramatically off. You'd still do the workout. You just are aware of what the cost of that

01:05:58.600 workout would be. You might make adjustments tomorrow or through your plan. That's what

01:06:02.080 we're getting when we're measuring at the end of sleep. The morning time we're seeing what was the

01:06:05.720 result of our sleep? What was the result of our workout? So yesterday, everything else we did,

01:06:08.960 if we're measuring HRV overnight, HRV number one is always higher at night because the

01:06:14.520 parasympathetic system is that dial is already turned up quite a bit because you're sleeping

01:06:19.000 where it's the highest. Although for most people, Joel, they will see the reverse. Like my RMSSD

01:06:25.420 HRV overnight is lower than the log normal transform I get out of Morpheus. Yeah. Cause Morpheus,

01:06:32.960 if you look at the actual RMSSD data, you would see that you'd be higher. And I know Morpheus is

01:06:37.700 different obviously. And can we see that in the app? You can't right now. We've honestly,

01:06:41.580 when we first came out with BioForce and Morpheus, there weren't so many other apps to compare again.

01:06:45.540 So it wasn't as big a deal to not show the raw number. So we chose not to, but now I think we

01:06:48.920 probably will just because people do want to compare. But anyway, overnight that dial of that

01:06:54.040 parasympathetic system was already higher. So we're getting less of a responsiveness to see what

01:06:57.980 actually is changing at rest. We're not measuring at rest. The second thing is if you have arrhythmias,

01:07:03.980 if you're an athlete who has very high HRV, you don't have as much variability. We're not

01:07:07.700 really gauging the true responsiveness of the 24 hours before we're measuring more of what's it

01:07:12.820 doing during the recovery period, which has some correlation obviously, but we're not really seeing

01:07:18.600 where are you at the end of that recovery period? Where are you ready to go today for this next

01:07:24.060 period of stress? And here's the biggest thing is if you do something, if you do a workout in the

01:07:28.980 evening, if you have a few glasses of wine, if you're doing something very mentally challenging,

01:07:33.060 the first part of your sleep, you're just responding to that. And so your HRV is not

01:07:37.860 reflective of this whole process. It's just reflective of, hey, you just did an evening

01:07:41.480 workout and your HRV is still suppressed for the first half of your sleep responding to that workout.

01:07:45.720 So we don't get a true picture of where am I at right now and how does that correlate to what I

01:07:52.180 should be doing for the next 12 to 18 hours as I'm awake across the next day. So I think we're just

01:07:56.960 probably getting a much better gauge of sleep and how our body is responding during sleep,

01:08:01.900 but we're not really getting this true picture of how did our body go through the whole process of

01:08:06.460 life, sleep, recover, next. We don't see that picture as well because we're not measuring at

01:08:12.200 the end of sleep. We're getting this average across it. So I don't think it's telling us really the same

01:08:16.400 thing and it doesn't have the same utility for telling us from a workout perspective, what's the

01:08:20.540 most appropriate thing for us to do. It might be a silly question, but it occurred to me now as we

01:08:24.840 were talking about this, the one fundamental difference from one day to another in that morning check for me,

01:08:31.300 is there are some mornings I wake up and I have to pee so badly and there are some mornings when I

01:08:36.080 don't. And there's a part of me that's wondering as I'm laying there doing my test while needing to

01:08:41.000 pee, is that putting a little more sympathetic tone into this? Am I getting a skewed measurement?

01:08:47.260 Would I be better off going, peeing, coming back, waiting a few minutes?

01:08:50.700 I go to the bathroom and get up. It's not that big of an issue if you just go up, go to the bathroom,

01:08:53.700 you come back and you reestablish. And part of that's actually measuring how well can you reestablish

01:08:57.120 that? If that significantly impacted your HRV, it probably was on the lower end to begin with.

01:09:01.260 But it does bring up a point. I should mention, people with really high HRVs, I don't think this

01:09:05.780 becomes an issue, at least in Morpheus until you're in the 90s or resting heart rates in the low to mid

01:09:10.460 40s. Your HRV is already very, very high that laying down, you're taking some of the responsiveness

01:09:15.780 away. If you start getting to those categories, you probably want to take it seated. The challenge for

01:09:19.860 most people seated, they get antsy, they move around, you introduce more motion artifacts and

01:09:23.820 it becomes more difficult. But we really want to have as much range that dial available as possible

01:09:28.640 because we want to see how the nervous system is turning that dial. So if you're very high,

01:09:32.900 like I said, I would say someone who's resting heart rates. Yeah, I'm never above the low 80s is

01:09:38.080 as high as I get. Yeah, I think at that point, laying down is still a good way to take it. But if you get

01:09:41.560 up in the 90s, mid 90s, you really want to maximize that potential responsiveness. And that's where a

01:09:46.900 seated measurement makes more sense. But that's a fairly small percentage of the population that's going to

01:09:50.900 be up in the ranges. RMSD, you're talking 120s, 130s, 150. When you get up in those ranges, you probably

01:09:57.520 want to take it seated. Okay. When I bought the Morpheus a couple years ago, you had an armband and a

01:10:04.360 chest strap that came with it. So I still use the armband as my morning check. That way I don't have to move

01:10:09.960 them back and forth. It always sits right there. And then I use my chest strap when I'm exercising. You've gone to

01:10:15.760 just a chest strap. Is that because you think you're going to get better data and it's just better to have

01:10:19.400 people using the chest strap for both? And should I do the same? I mean, it's two things. I don't

01:10:23.400 think the data, as long as you're measuring consistently and you have the chest or you

01:10:27.020 have the armband placed correctly, that's not any less accurate, I don't think. The problem we ran

01:10:30.960 into is people, as you know, are trying to use our armband to train because it's more convenient.

01:10:35.680 Like, oh, the armband goes to my wrist and their workout data was just not as accurate. The second

01:10:40.780 thing is it looked like a watch, but it wasn't a watch. And so we had a bunch of confusion with

01:10:45.600 people putting it on their wrist and not sure what to do with it. It just created a

01:10:49.220 lot of confusion. So at the end of the day, I said, look, the chest strap is giving us the

01:10:52.740 best data. It's unambiguous. If people really want to wear an arm device because they just

01:10:58.620 don't want to put the chest strap on, we work with the Scosche Rhythm 24 because it uses the

01:11:02.480 exact same sensor that Morpheus used in our original armband that you have. And so if they

01:11:06.720 want to do that, they can measure it that way. But from an accuracy perspective, from both

01:11:10.960 the HRV and the workout, it just made sense for us to standardize that, use the chest strap

01:11:16.080 and make it as universal as possible rather than try and sell to devices, which people

01:11:19.680 were ultimately somewhat unsure of.

01:11:22.040 Okay. So let's talk a little bit about the question that I'm sure is on everybody's mind,

01:11:25.300 which is I've been doing this for a while and I get that at the individual for my data,

01:11:31.120 I see my up and down level and I know that, hey, when it's higher, I'm generally going to

01:11:36.700 perform better and I can push a little harder when it's lower. I'm probably not going to perform

01:11:40.320 as well and maybe need to make that a little bit more of a less hard day. But then you get this

01:11:45.620 question of, hey, what can I be doing to improve the quality of my health in a way that is measured

01:11:53.720 by this output of HRV?

01:11:56.020 Sure. I mean, this is where HRV is driven by genetics, fitness, primarily cardiovascular fitness

01:12:01.720 is the biggest thing we see in quarterly and drive it. And then obviously lifestyle, doing things in

01:12:06.560 your lifestyle that make that sympathetic dial come down when you don't need it.

01:12:10.320 And doing things that turn on that parasympathetic dial when you're not using this is going to put

01:12:15.180 you in your highest level of your particular range from a lifestyle perspective. And that's

01:12:20.260 where I think most people underestimate the lifestyle impact on HRV and train everything

01:12:25.400 else. They don't realize if you're stressed out from work, six, eight, 10 hours a day, you're

01:12:30.620 running around chasing your kids, you're doing all these things and you know, your daily life,

01:12:34.380 that's a pretty significant impact on your HRV because that sympathetic dial will be turned up for

01:12:39.120 hours on end. Maybe not the same degree, of course, as a workout, but hours on end. So a lot of it

01:12:44.160 comes down to just the stuff we know in everyday life that makes us healthier, eating healthier foods,

01:12:50.320 making sure we're getting enough sleep, managing our mental stress effectively, doing things that

01:12:54.340 allow us to relax and turn that parasympathetic dial back up and that sympathetic dial back down

01:12:58.600 and then build VO2 max. So do you think it's more impacted by peak aerobic fitness or by

01:13:06.540 base aerobic fitness? Would you say it's more impacted by a higher zone two or a higher VO2 max?

01:13:12.540 I mean, they both contribute exactly how much, you know, I couldn't say. We tend to measure aerobic

01:13:16.860 fitness from a peak standpoint for the most part. So that's more standardized what we would look at.

01:13:21.600 But I think training frequency matters, which is where you get zone two, right? You can't do VO2 max

01:13:26.420 training five, six days a week. We do a lot more zone two. We do a lot more frequency and volume of

01:13:32.200 that. And I think that translates more than likely into a higher HRV, even if you didn't go out and do

01:13:38.000 a bunch of the zone two or the VO2 max type work. Yeah. And then one of the advantages, I think,

01:13:43.580 of those overnight tests, again, whether it's aura, whoop, eight sleep, any of these things is

01:13:49.000 people have noticed how much of an impact alcohol has on overnight HRV. It's probably one of the most

01:13:56.100 profound changes you see in response to alcohol. And I would argue that a big part of the movement

01:14:01.420 we're seeing around people drinking less can be attributed to those devices, which is giving

01:14:06.700 people visibility into, oh my God, like I didn't realize that alcohol could have such a profound

01:14:13.240 impact on this. I guess that would kind of be out of your system maybe the next day or would that

01:14:17.800 still be there in the morning? It would depend in the sense of if you had alcohol close enough to

01:14:21.820 bedtime, it's going to impact your sleep, which impacts your recovery, which will impact the

01:14:25.380 worry measurements. You'll still see some remnants of it for sure. But yeah, you'll see that more

01:14:28.600 directly in the overnight stuff. I think as a whole, what we see is people become much more aware of

01:14:33.800 things like alcohol, things like excessive stimulants, or God forbid, smoking, or massive

01:14:38.860 amounts of chronic mental stress. Those things impact much more than I think people realize.

01:14:43.700 An example of this, we were measuring a college soccer team across a couple of seasons. And we would

01:14:49.500 see that during finals week, they would look far worse than during tournaments, even competitive

01:14:54.760 playoffs, just because that stress of being in a finals week where you're studying and you're not

01:15:01.620 sleeping and you're stressed out. Give me a sense of the range that you would see. Give me an average

01:15:06.640 athlete where this would be their morning HRV under these circumstances. This is what it looks like

01:15:11.060 when they're overtrained. This is what it looks like when they're in the tournament. This is what it

01:15:14.400 looks like in finals. Yeah. Again, there's a lot of variability there. But from a college standpoint,

01:15:18.480 most soccer athletes that we would see in these are female athletes to be in the 1.00

01:15:21.140 low to mid-80s on a normal basis. And again, this is Morpheus system. You can't compare these to other

01:15:27.440 numbers. But they'd be in the low to mid-80s kind of as a normal range. We should then drop into the

01:15:31.340 70s, sometimes down to the 50s, which is in Morpheus. That's the stress?

01:15:35.240 Yeah. That was the stress of finals week. It's two, three, four days of not getting much sleep and

01:15:39.360 studying a lot and just working out very less or very sporadically, probably compared to normal

01:15:44.260 training sessions. You just see the impacts of life being very, very significant that people don't

01:15:49.640 necessarily expect that because they feel like, oh, the workout's the most impactful thing. Well,

01:15:53.920 it is in a way, but it's also only an hour, maybe two hours. It's the rest of your life that also adds

01:15:59.560 up to a huge amount of stress if you are very, very stressed. And if you're going through your life

01:16:04.460 in that kind of type A, I'm always turned on. I can't turn off my stress. That has a very big

01:16:08.700 impact. I think Sapolsky, who I know you've had on the show, talks a lot about the mechanisms.

01:16:12.860 You see that play out pretty frequently when you look at HIV data.

01:16:16.900 I don't utilize the Morpheus system fully because I only wear it during my zone two workouts. So I

01:16:24.140 don't wear it when I'm doing my VO2 max workouts because I'm already wearing that Polar system

01:16:27.860 because it pairs with the Garmin and it pairs with the other power meters and all the other

01:16:31.800 stuff I'm using. I guess I could double up. Can you wear two chest straps?

01:16:35.860 I mean, you could. The Polar should be able to connect to the Morpheus app directly while you're

01:16:40.380 training.

01:16:40.920 But then it would have to pair to two apps.

01:16:42.780 It should be able to if it's got two Bluetooth radios, which it probably does.

01:16:45.960 Okay. So I can do that. That would be good to know. But I don't wear it when I'm strength

01:16:50.400 training. So I realized that I'm failing to give it all of the data because that's another,

01:16:57.260 I don't wear it when I'm rucking. There's a lot of time I'm active, but I'm not wearing it.

01:17:01.780 So how much am I missing out on in terms of the fidelity of what it might be telling me? And I want

01:17:08.580 to, cause I want to talk about the algorithm. One of the things you and I spend so much time on is I

01:17:13.080 can't make sense of how it's coming up with the numbers, even though they end up being right

01:17:16.860 most of the time.

01:17:17.520 The more data you give it, obviously the better it's going to be. Exactly how much you're losing

01:17:22.860 and it's hard to say, but we're measuring the output with that HRV change and with the numbers

01:17:27.580 that you're putting into it. We know the output of where you are. We can't always ascertain how

01:17:32.080 you got there if we don't have all that data of the workout sort of things. But as long as we have

01:17:37.760 that consistent HRV measurement every morning in standardized conditions, you know, we're still

01:17:42.320 able to get the vast majority of what we're trying to get, which is what are you most likely to do

01:17:47.920 when you work out right now? How much is that going to impact you?

01:17:50.140 So that is the most important thing.

01:17:51.400 That's by far the most important thing.

01:17:53.040 To make sure that every day you see my heart rate, my heart rate variability, how long I slept,

01:17:57.860 how sore I am and my desire to train.

01:18:00.320 Yeah. That's the vast majority of it because again, it's telling us where you are right now.

01:18:03.540 The readout state.

01:18:04.340 The readout. That's the output. We know the output. This is where your body is at right now.

01:18:08.060 If we can reverse engineer that from the input, we can have some more insight into that.

01:18:12.520 But you're not losing a bunch of accuracy because you didn't get that. We want the output. We want

01:18:17.120 that as standardized and accurate as possible. So I would say as long as you're measuring consistently

01:18:22.020 every morning, the same context, same conditions, it's going to be more than accurate enough for

01:18:26.600 what you want to do. Okay, great. And again, my use case is quite simple, which is mostly just

01:18:31.840 predicting that zone too. But I have a feeling a lot of people listening to us will immediately

01:18:36.180 resonate with that use case because I think for many people, there's still a little bit of ambiguity

01:18:39.880 on not the concept of what zone two feels like, but the day-to-day variation, which again,

01:18:45.180 is really significant.

01:18:46.540 And that speaks to the body's dynamic. We can do a zone two test and look at lactate and all

01:18:51.460 these things. If we just take one test, we don't realize how much the body changes on a daily basis.

01:18:56.280 And so if you just, okay, I took my lactate test six months ago. I'm just using the static zone two.

01:19:00.840 You're missing that dynamic change that the body is going through on a daily basis. The body is not

01:19:04.800 static. I measure blood pressure. I measure testosterone. Like all these numbers change

01:19:08.860 constantly. Like the body doesn't sit still. And I want to make another point, which is people

01:19:13.200 again, might be listening and saying, you know, Peter, man, it's too much data, dude. Just go out

01:19:17.400 there and do it. I'll give you the counter argument to that. When I was a competitive athlete

01:19:22.580 and I was training 28 hours a week, I had the luxury of junk miles. I didn't always have to be

01:19:30.340 perfect, but I'm not a competitive athlete. I'm a competitive father. I'm running three businesses.

01:19:35.860 I don't have the time for nonsense. Every minute I'm training, I have to get the training stimulus

01:19:43.700 right, or at least as close to right as possible. So when I'm setting out to do zone two, I got to do

01:19:50.900 it. And if I'm out there trucking along and my lactates 1.1 or 2.9, I'm missing the training effect

01:19:59.020 I want. So the more insight I can get to narrow that down, the better. I'm only going to give three

01:20:04.700 hours a week or four hours a week to that training. I'm not going to put 12 hours into it where if I

01:20:10.120 do 12 hours and I screw up three of them, who cares? I still got nine. So that's why I know that

01:20:15.760 there are people watching this saying, dude, you're a psycho. And it's like, no, I'm just 0.97

01:20:19.500 efficient with my time and I don't want to waste my time. I think what it comes down to is each person

01:20:24.080 has a certain amount of training in a week. I think it's a weekly basis because that's kind of

01:20:27.600 the cycle we live in. There's an amount and a type and an intensity of training that you are

01:20:32.420 individually going to respond the best to. And if you go above it, not so great things are going

01:20:37.820 to happen. If you go below it, you're not going to get the outputs that you want. And that amount

01:20:42.200 is not easy to find. And it changes on a weekly basis. It's not the same week in, week out. And

01:20:47.460 so the more that you can hone in on that core, how much volume do I need and how much intensity do I

01:20:52.740 need? If you can get that right week in and week out, you will see continual improvements and

01:20:57.740 that work will turn into result. If you don't answer that correctly, bad things happen. You

01:21:03.180 either waste your time because you're not getting as much out of it as you could, or you do end up

01:21:06.980 overtrained, in which case we see injuries and we see burnout and lots of things that are going to

01:21:11.180 have negative health consequences. So I think ultimately what you said is I want to use information

01:21:15.920 to make sure that I'm getting the most bang for my buck, the best result for the amount of work I'm

01:21:20.860 putting in. And that's where I think data can play a really powerful role is it's information that you

01:21:25.400 can use to make much more granular decisions around rather than just guessing, oh, I should

01:21:30.360 go do 40 minutes today. Should you? Or I should do 200 minutes this week. Is there anything to support

01:21:36.060 that other than just throwing stuff at the wall and seeing what sticks? And I think that's what

01:21:39.720 intrigued me so much about HRV is it felt like I could open the black box and get some real answers

01:21:45.260 other than like test, train for eight weeks, remeasure, see what happened. I don't want to wait eight

01:21:50.620 weeks. I don't want to potentially lose the gains I could have made across that time period. So

01:21:54.640 for someone like yourself, that's again, wanting to get as much out of their time as they possibly

01:21:58.700 can. Yeah. Data can play a really strong role in that because it can answer questions that can't

01:22:02.300 be answered otherwise. And I think it only gets more significant as we age. 100%. When I was 40,

01:22:08.540 my recovery capacity was so much greater than now. Nevermind 30, 20, that's obvious. But even the

01:22:14.040 difference between 40 and 50 is significant. And I know there are a lot of people listening to us

01:22:18.600 right now who can relate. They're sort of like, hey, I'm getting a little bit older. I don't feel as

01:22:23.380 great as I did. It's a question of time and age. And I think the further we get along that, the

01:22:28.440 more this type of system. For me, I can't say enough about it. Age reduces your margin of error.

01:22:33.880 Yeah, exactly. I mean, I'm 44 and you can do a lot of things wrong in your 20s and maybe in 30s

01:22:38.100 and you can still get a lot of benefit out of it because you're so resilient. Your metabolism will

01:22:41.500 adapt. But like you said, the older you get, the less you can do that. And so you have to be much

01:22:46.180 more acutely aware of what your body can and can't do. And that's part of what HRV can help you

01:22:50.680 understand. You said something maybe 20, 30 minutes ago, and I jotted a note down because

01:22:55.680 I didn't want to derail us at the time. I'll come back to it now. So when I'm doing my VO2 max sets,

01:23:01.140 the thing I monitor for every set is heart rate recovery. So as soon as I get to the top,

01:23:06.400 because I do those on a hill. So as soon as I go to the top of the hill, I hit the lap timer

01:23:09.260 and I count how many beats does my heart rate go down in the first 60 seconds at the completion of the

01:23:16.160 interval. And that's a great proxy for how I'm doing. So I've got kind of my normal range should

01:23:22.640 be 30 to 35 on a really good day. I'm 40 to 45 a week ago or so. I had like one of the worst days

01:23:29.540 I've ever had where I was like 19 to 21 in a minute. That's all I could recover. I was smoked.

01:23:37.880 Now I didn't sleep the night before. I made an Instagram post about it because I thought it was

01:23:42.040 just a great sort of illustrative teaching point. You said something that made me wonder about

01:23:46.460 another test. Would there be any utility in right after a VO2 max interval doing an heart rate

01:23:54.040 variability test to see how much sympathetic tone can I dial down and how much parasympathetic tone

01:24:02.120 can I dial up after what's probably a peak, you know, a very high sympathetic, low parasympathetic

01:24:07.500 exercise? You can do that actually. I mean, it's very specific. Like you can't move around if you

01:24:13.140 want to get some standardization to it. So yeah, you could do a 10, 15 minute recovery period.

01:24:16.700 What if I just did it right at the top? Would that be too much?

01:24:19.620 Probably. I think heart rate recovery is far better. It's illustrated of what you wanted to get

01:24:24.080 to because what we see is heart rate recovery is driven by what? The reactivation. The balance of those

01:24:29.780 two. It's turning the sympathetic down as quickly as possible and turning the vagus,

01:24:33.420 the parasympathetic up as fast as possible. So heart rate recovery is already

01:24:36.260 giving you that information without having to get as granular as pulling out the B2B. Like you don't

01:24:41.160 necessarily have to get that level of detail. You can just see the heart rate drop and that drop is

01:24:45.920 being caused by the increase in heart rate variability. And what's really fascinating is

01:24:50.440 they used to think, and this is pretty new research I was going through. We used to have this idea that

01:24:55.120 when your heart rate increased above a hundred beats per minute, that there was really no vagus input.

01:25:00.140 There was almost no parasympathetic. And they figured that because they would look at acetylcholine

01:25:07.000 as essentially they could block it into the heart. And they'd say, oh, we block acetylcholine from

01:25:10.540 vagus and heart still pumps just fine during their exercise. So exercise must not be vagusly driven at

01:25:15.560 all. But what they found is that the vagus might actually be turned up a bit during exercise in some

01:25:22.140 sense because it can increase coronary blood flow by increasing vasodilation in the coronary artery.

01:25:26.740 And so HRV is probably not, the vagus, I should say, is probably not completely inactive,

01:25:32.960 probably not completely off. It's this ratio. It could be turned up, but the sympathetic system

01:25:36.580 is turned up so much more and it's using a different mechanism. We don't see the heart rate

01:25:40.800 variability high, obviously, but it's probably more ready to turn that back up even further as soon as

01:25:46.380 that sympathetic system starts turning down. And we're dropping our adrenergic hormone levels or

01:25:51.480 catecholamines, all these things are dropping pretty quickly. And the faster we can turn that

01:25:55.440 parasympathetic up, the faster our heart rate comes down. The other thing that's interesting

01:25:59.360 is they've looked at heart recovery in terms of it represents to some extent the balance

01:26:03.400 of the aerobic and anaerobic systems that contributed to that exercise. So the more

01:26:07.400 aerobically driven something was, the faster our heart rate drops. Because in a lot of ways,

01:26:12.720 higher heart rates are driven by that sympathetic and by the anaerobic pieces of metabolism.

01:26:17.420 So what do you see in young exceptional athletes? I mean, I wish I had tracked this metric when I was

01:26:24.620 a teenager. Like I wish I know how much my heart rate recovered then back when my peak heart rate

01:26:30.080 was 205 to 210. What do you see in these young collegiate athletes?

01:26:34.620 I mean, a really good number is 50, 60 from a near max. So I trained combat athletes, UFC fighters,

01:26:40.520 pride fighters for many, many years, and they have to go into the octagon or cage ring,

01:26:44.400 depending on what they're doing. And they have to fight for three, five minute rounds to five,

01:26:48.200 five minute rounds, probably just a 10 minute round, which is crazy. So we would use heart rate

01:26:52.180 recovery between rounds as a really good gauge. Because it's one minute between rounds, same as

01:26:56.440 boxing. It's one minute between rounds. Yes, exactly. And they would sit down between rounds

01:26:59.080 so we could standardize that. And so I would use that drop as a very good gauge of how well

01:27:04.400 conditioned is this athlete, how ready to go out there. Give me an example of what you would see on

01:27:08.520 a fighter. They would come out of the previous round at what? They come out 160, 180,

01:27:12.740 depending on what the round was like. And the round was slower, it's going to be much lower.

01:27:17.020 But most of the rounds are between like 160, 180, depending on their age or anything else.

01:27:21.340 You might see them spike up. And they could drop by 50 to 60.

01:27:23.840 I would want them to get to 130s between each round. We would simulate this in sparring rounds

01:27:28.500 leading up to the fight. We wouldn't measure during a fight, obviously. But if they're doing

01:27:31.700 a simulated fight round where it's three fives or five fives, I'd want them to be able to drop in 0.52

01:27:36.100 the 130s between every round. If we started seeing their first round, they weren't coming down below

01:27:41.080 150, 160. They're not in good enough shape.

01:27:43.460 They're not good enough shape. They were going to fatigue every time. And that just told us

01:27:46.220 they were having to rely so much on the anaerobic piece. They were going to fatigue because at a

01:27:51.040 sport like this, you have to be really explosive, but you have to also have the endurance. It's about

01:27:55.840 the ratio of energy utilization that matters so much. If you don't have enough anaerobic, you're

01:28:00.140 going to lack power and speed and ability to finish. But if you have not enough aerobic, you're not

01:28:05.800 going to sustain that explosive power for very long. And so it's really tricky to get that ratio

01:28:09.980 right. And you see that fastest, most explosive, hardest hitting athletes often fatigue the fastest

01:28:15.080 because they're generating that from the anaerobic side and they're relying on that. And that's

01:28:20.160 great if they can win and they can knock the person out or submit them. But if they can't,

01:28:24.200 and you get in the later rounds, that's where they're going to really struggle versus somebody

01:28:27.020 else who's more aerobically dominant. So that's a really hard part about that sport is getting that

01:28:32.020 ratio correct and training the right side of it. But the heart rate recovery was such a great way

01:28:36.760 to see that. And so like I said, I would want to see, in sparring at least, there are going to be

01:28:41.340 higher heart rates during the competition from the psychological stress, but we'd want to see them

01:28:44.980 drop in the 130s. Again, this is seated, going from standing to seated, but they should be able

01:28:48.880 to get the heart rate back in the 130s between each round, ideally, before they go for a fight.

01:28:53.200 If they could do that and they were fighting at a high level pace, you knew they were in pretty good

01:28:57.380 condition and they'd be ready to go out and go. If they weren't doing that, especially if the early

01:29:00.420 rounds, if they're, like I said, round one, they're at 150, 160, you know they better finish the fight

01:29:04.920 quickly, they're going to be in trouble. So interesting. I want to go back to kind of what

01:29:08.820 we were talking about on the trailing average of HRV data. So I'm glad to hear that my failure to

01:29:16.040 utilize Morpheus in every moment of the day isn't really impeding its fidelity. I kind of know that

01:29:22.720 because again, as I said, it just, it's so damn accurate in what it predicts. It's like a shaman.

01:29:27.640 But what amazes me, you have the patience of a saint, Joel, because I email you so much,

01:29:33.180 it must be infuriating. But what amazes me is how sometimes the data I see, I don't believe it.

01:29:40.200 You're always like, yep, but you got to look back at what it did seven days ago and six days ago and

01:29:44.640 five days ago. And so walk me through the arc of HRV over time and how the undulation of the HRV

01:29:55.500 matters as much as the HRV on a given day. This is the part that is totally news to me. And again,

01:30:02.360 I think the proof is in the pudding because at least for me, this seems to really work.

01:30:08.300 Sure. Yeah. I mean, again, we're looking at not just what is your number, but what is your number

01:30:13.280 in relation to where you're usually at? And we're looking at how much does our number move today

01:30:17.860 versus what's like a normal movement for you. And again, we're looking at just this manipulation

01:30:22.400 of dials by the autonomic nervous system. So what Morpheus is doing is looking at your seven days

01:30:27.100 and seeing, okay, what's your normal baseline number itself, but also what's your normal level

01:30:32.580 of variability across those different time periods. And then we look at standard deviation.

01:30:38.080 And so if we see big changes greater than one standard deviation away from your average,

01:30:42.700 away from your baseline, we know that that's the autonomic nervous system responding to a greater

01:30:46.980 input. It's responding to more stress and it's having to make bigger dial adjustments.

01:30:51.440 And that tells us that you've been paying a higher cost because you put your body under more stress.

01:30:56.740 If we were to probably illustrate to just isolate, our body's in kind of a normal rested state,

01:31:01.460 let's say. And then we do a single workout and we allow it whatever necessary time needs to recover.

01:31:07.360 During the workout, obviously the body turns up that sympathetic dial. It turns down that

01:31:11.800 parasympathetic dial. We produce more force, more energy, more power, blah, blah, blah.

01:31:15.900 And then after the workout, like we talked about, the sympathetic system starts to come down,

01:31:21.300 the parasympathetic starts to come back up. Now, depending on the workout you did,

01:31:26.420 that could happen in very different timescales. And Steven Seiler, again, a really popular

01:31:31.020 polar training, he was just showing some stuff. If you do like zone two and below aerobic work,

01:31:36.320 that recovery happens pretty quickly. Within a few hours, we'll see that parasympathetic

01:31:39.960 system turn on and it'll come up above your baseline to maybe 110, 120% of what your HIV was

01:31:47.800 at rest. So Morpheus, you know, if you were, I don't know, a 70, if you were to theoretically

01:31:52.040 remeasure it, you might be a 74, 75, you would see a noticeable uptick above baseline because our

01:31:58.840 body has been able to shift very quickly into that restorative phase by that vagus nerve firing

01:32:05.220 more forcefully and turning up the parasympathetic. So we see, again, suppression during

01:32:09.700 the workout and the curve starting to come back up during lower intensity workouts, we'll see it

01:32:14.260 pretty quickly come up above baseline. And then kind of as your body went through that complete

01:32:18.100 cycle and restored homeostasis and did what it wanted to do, it was going to just kind of settle

01:32:21.900 back down to where it started. And so you'd see this curve, this very clear curve now in a much

01:32:26.460 higher intensity. But the amplitudes aren't as big. The amplitudes aren't as big, exactly. In a much

01:32:30.000 higher intensity, higher volume, even lower intensity, it's easier to don't do me three hours of it in

01:32:34.440 heat. That's a much bigger stress. You see the same curve, but it'd play out over a much longer

01:32:39.300 timescale because it would take much longer for the body to restore back to where it was and then

01:32:44.620 to get everything where you started with from an HRV standpoint. So we'd see a much bigger depression,

01:32:49.520 a much lower drop, and then we'd see a much longer timescale for it to come back up to normal.

01:32:54.600 And then we might see it come up above normal for some period and then restore.

01:32:58.100 But I'll say there's a big difference in the individual as well. People with higher HRV and people

01:33:03.740 that are more fit, they're more likely to see this increase above baseline. Somebody who has less

01:33:08.820 autonomic range, they can't quite turn that peristeme thick dial up as much. They might never

01:33:13.200 get up above baseline. They might just kind of spend time getting back to that baseline HRV and

01:33:17.340 you never really see the peak above baseline and then they're restored afterwards. So it's an

01:33:21.680 individual thing based on fitness, but we should generally see suppression of HRV, an increase of

01:33:26.920 HRV, and then kind of a back to normal HRV. It's that process. Now the tricky thing comes is we have so

01:33:33.080 many other things that influence it. Alcohol influences it, mental stress influences it, sleep influences it.

01:33:38.260 So it's overlapping influences that will get to where you're seeing, but that's the core thing

01:33:43.540 of our body response to stress by dropping HRV and then by recovering HRV and then by coming back to

01:33:48.820 whatever that normal range is. Yeah. And that's why, again, I think it's not intuitive enough to

01:33:53.960 just look at the HRV that given day. You can't just say my HRV was X today and there's a one-to-one

01:34:00.500 map between what my HRV is on a given day and what my training output should be that day,

01:34:05.480 because you need to know the first derivative, the second derivative, and frankly, even the arc

01:34:11.900 that it's on. It's an all-cause metric that's not just looking at one input, it's looking at

01:34:16.440 everything. And so interpreting that output is not always the easiest thing in the world. And again,

01:34:21.240 if you kind of look at this, like, oh, up is always one thing and down is always one thing.

01:34:24.980 It's a little more nuanced than that because it's this responsiveness of, okay, it's going to drop

01:34:29.300 after peer stress, then it's going to come back up. It might come up way above baseline and then kind

01:34:33.060 drop back down to normal. This curve is not always the most intuitive thing to understand.

01:34:37.320 And that's part of the challenge, I think, with people just kind of looking at it and say,

01:34:40.520 oh, it's this, I'm going to go do that. Yeah. I mean, it's certainly not intuitive to me,

01:34:44.280 which is why I annoy the hell out of you and pepper you with all these questions.

01:34:47.900 I want to change gears to talk about something I've alluded to a couple of times in the past,

01:34:52.400 which is a phenomenon that we see in patients taking this new class of drug. I mean,

01:34:57.940 it's not really a new class of drug. It's been around for a decade, but a class of drug that has

01:35:01.560 gained a lot of popularity in the past probably two to three years, which is the GLP-1 agonists

01:35:06.440 and the dual agonists with GIP as well. And again, we don't have an enormous and overwhelming body of

01:35:12.620 evidence on this. We don't have that many patients in our practice, and we frankly don't have that

01:35:16.400 many that take it. But certainly over the past three years, have to think we've seen two dozen

01:35:22.160 patients on these drugs. And again, in all cases, we have overnight information on heart rate and

01:35:28.940 heart rate variability. The unambiguous sign that we see is that resting heart rate is going up and

01:35:36.160 it's going up an average of 10 beats per minute with a range of about eight to 12. This is unmistakable.

01:35:44.500 It's not subtle. And when they come off the drug, usually within a month, it goes back to normal

01:35:51.580 for patients who do indeed go off the drug. We're also seeing a compression of heart rate variability.

01:35:57.820 So we see heart rate variability come down, although that's less predictive. But I now realize we're

01:36:03.160 kind of using the standard, we're not using like a Morpheus, we're using kind of the aura ring or

01:36:07.580 whatever, and maybe those data just aren't as accurate. My first question for you is, if there's

01:36:13.080 something going on in a drug that is predictably driving heart rate up, would you expect it is also

01:36:18.920 driving heart rate variability down? Do those tend to move in that opposite direction?

01:36:22.780 Yeah. Generally speaking, you'd see that. Now, just as you mentioned that, my thought would be

01:36:27.580 it's a strong appetite suppressant. Correct.

01:36:31.000 The vagus is very tied to appetite. So if you think about this, if we were to turn up that

01:36:35.920 sympathetic dial, our hunger gets turned down. We don't really want to be hungry while we're in the

01:36:39.980 middle of some stressor, right? But after that period of stressor is over, theoretically we've

01:36:44.060 burned energy, we need to restore and eat. And so the vagus is very connected to the gut and to hunger

01:36:50.160 centers. And it feeds up into the medulla and it's controlled hypothalamus. There's a very strong

01:36:55.300 vagal relationship to hunger and a desire to eat. So I almost wonder if suppressing the vagus and

01:37:02.140 decreasing HRV, increasing heart rate is a byproduct of how this is inhibiting appetite. That would make

01:37:08.200 sense.

01:37:08.900 And so it's an interesting question because it then leads to another question, which is,

01:37:13.220 if I told you that I'm going to change you in a way that your heart rate is 10 beats higher and

01:37:21.440 your HRV is 10 milliseconds lower, you would say, well, whatever you did was negative.

01:37:27.540 Sure. There's a cost to that, right?

01:37:29.360 Do we think that that could be the case here? I mean, again, we're really wildly speculating.

01:37:33.940 You know, I get asked all the time, Peter, are these drugs safe? Are these drugs good? What do you

01:37:37.360 think of them? And I'm always say sort of the same thing, which is look, clearly for some people,

01:37:41.500 whatever unknown exists around these drugs is worth it. If an individual goes from being 250 pounds to

01:37:50.020 200 pounds and having a hemoglobin A1C of 7% to 5.5%, that is so positive for them that I think it

01:37:59.980 justifies whatever unknown exists around these drugs, if there's no other way they're going to

01:38:04.080 achieve that benefit. But I'm really more interested in these marginal cases of people who don't have

01:38:09.180 diabetes and want to lose 20 pounds, which by the way, might not really matter in their overall

01:38:15.840 health. It's cosmetic. And are they taking too big a risk? That's kind of the question I'm interested

01:38:22.040 in. And that's why I keep coming back to if it raises your heart rate that much and lowers your

01:38:27.220 heart rate variability, are we picking up a signal that is just a niche representation of appetite

01:38:32.420 suppression via the vagus nerve? Or is it actually playing a role in the parasympathetic sympathetic

01:38:39.380 dials? I would imagine it's got to be, I mean, to move it that large, 10 beats is a fairly

01:38:45.020 significant amount. I can't imagine that's not having an actual effect on our ability to regulate

01:38:51.620 ourselves effectively. We're probably in a more sympathetic state all the time, which is going to

01:38:56.800 have a cost. Now, if it's a few weeks or a month, you know, maybe that cost is relatively small. I

01:39:01.480 don't know. But if it's weeks or months, they're living on this drug, I would suspect there's a real

01:39:06.280 cost to that. And to your point, if it's, the benefit is they lose a bunch of weight and all

01:39:10.540 their blood markers improve and we see health outcomes, maybe it's certainly worth that cost.

01:39:14.420 But maybe somebody, like you said, who cosmetically wants to lose weight and they have an easier time on

01:39:19.060 the drug doing that, is that a net benefit? I mean, it's hard to say because we don't have long-term

01:39:24.120 studies on those drugs. But I would just kind of say in general that, yeah, if we see these noticeable

01:39:28.680 decreases in HRV and these very noticeable increases in heart rate, that's a real sign that

01:39:33.820 the body is, that autonomic nervous system is being adjusted in an artificial way. And that probably is

01:39:38.780 not a good thing in the long run, specifically if it's for a long time. Yeah. I don't know where

01:39:45.300 the companies are at in terms of their post-surveillance, meaning post-marketing surveillance studies.

01:39:51.460 And if this is a metric that they're tracking or interested in, but again, given the popularity

01:39:56.940 of these drugs, there's no shortage of opportunities to kind of measure these things.

01:40:01.580 Yeah. It'd be interesting to see. Also, I'd be curious to see what happens when they come off.

01:40:04.520 Do we see a big rebound? Does their heart rate stay suppressed? How does it change?

01:40:08.540 Our sample size is so small that I want to be very careful and note that everything I'm saying is,

01:40:13.620 it could be nothing. It could be that it's just a very small N and 25 people is not enough. 0.99

01:40:19.340 But the thing is, in the 25 people, I've never seen an exception. When you don't need statistics

01:40:24.000 to measure things, you kind of need to pay a little closer attention to them.

01:40:27.540 It's pretty hard to modulate appetite that significantly without suppressing the vagus

01:40:31.720 to some extent because it's so closely connected.

01:40:33.200 But we do see everything come back to normal when people are off the drug.

01:40:37.040 And I'm trying to remember, it's been a while since I've looked. I feel like it's within a month,

01:40:41.520 maybe even less. It might be within two weeks. Everything is coming back to normal.

01:40:43.940 The half-life is very long on those, so it would probably just take a few weeks to clear out.

01:40:46.780 Yeah. And I don't think I have enough insight to contrast the two most popular drugs,

01:40:53.320 semaglutide and terzepatide. But again, I don't think these drugs are going anywhere.

01:40:57.620 And I would love to better understand this. My guess is there are a number of people on these

01:41:03.140 drugs that might not be aware of this because either they're just not tracking it or they are,

01:41:06.440 but they haven't noticed it. Some patients will say this after a few months. And I realized they

01:41:12.100 kind of forgot when I told them this in the first place. But anyway, wondering if you had seen

01:41:16.300 anything about that or heard anything about it?

01:41:18.360 ADHD medications, for example, you see a very suppressed HIV and a very elevated

01:41:23.200 sympathetic heart rate. You see it in several types of things like that. Any sort of

01:41:27.620 strong stimulants, obviously those are not stimulants, but people who live on caffeine and Red Bull and

01:41:32.820 are constantly shoving coffee. To constantly turn that sympathetic dial up is, to me, a kind of sign that

01:41:39.700 the sympathetic system isn't working the way that it should by itself, probably because

01:41:42.660 you've overstimulated to begin with through stress and the lack of ability to turn that off. And we

01:41:47.240 see people reach for stimulants and artificial ways of turning that sympathetic dial up once their

01:41:52.880 body's not doing it the way that it should. And so we see people kind of self-medicate with

01:41:57.240 stimulants to get that sympathetic response when if they had been able to manage stress more

01:42:02.040 effectively, they probably would have a normal functioning sympathetic system that wouldn't need

01:42:05.160 that artificial stimulus to turn it up.

01:42:06.760 Let's take a step back and now just talk about where you think HRV belongs in the hierarchy

01:42:12.980 of health metrics. We've talked about so many different types of health metrics and we've

01:42:18.300 talked about how output metrics tend to be preferable. We can sort of think of them as

01:42:23.800 functional metrics. I like an oral glucose tolerance test more than a hemoglobin A1c. One is kind of

01:42:29.100 functional. It actually is a test. It watches how you do something. VO2 max, strength, those are more

01:42:35.440 functional tests than maybe resting heart rate or muscle mass. Muscle mass is great. It's

01:42:41.140 predictive as is resting heart rate. But when you actually put the workout, HRV is obviously a

01:42:45.820 readout state. Where do you think on the dashboard of health longevity, where do you place it?

01:42:52.540 You know, I think it's important to monitor as you're going along, but I would say if all your

01:42:56.080 other metrics are telling you you're healthy and you're going the right direction, nine times out of

01:43:00.240 10, we tell you the same thing. The point where I would be more aware of it is where it's

01:43:05.400 very low and you don't have a reason for that. You'd want to figure that out. We've had people

01:43:09.400 have arrhythmias they had no idea about and they go to the doctor and they're on, have some serious

01:43:13.100 heart condition that they need to be aware of. But I would just say long term, it should line up

01:43:17.340 with VO2. It should line up with your metabolic market. It should line up with all of this stuff.

01:43:22.380 It's easier to track because we can measure it on a daily basis. I think that's probably the

01:43:26.240 advantage I would say it has is you're not getting a VO2 max test every day, obviously, or every week,

01:43:31.740 every month. You're not doing lactate tests. You're not doing these markers that are more

01:43:35.900 output based that are really important frequently enough to get feedback of whether or not you're

01:43:40.400 going the right direction. So I think we can look at HRV in a more granular daily basis, just kind of

01:43:45.620 say, am I going the right direction? And that's probably more of a utility than a great predictor of

01:43:51.040 something. And we can look at those daily changes to help us make more informed decisions. We can't do

01:43:56.060 that with VO2 max or with more invasive tests. So it's a more narrow data point, but it gives us

01:44:02.500 something we can use more actionably than these longer term tests, I think are better actual

01:44:07.080 measures of outcome. And if we see our HRV trending down significantly, that is a warning sign. We're

01:44:11.660 doing something wrong. Our body is not adapting the way that it should be. And we need to make

01:44:15.660 adjustments, whether it's training or lifestyle. So I think it's used differently. We use VO2 and we use

01:44:21.120 those sorts of output measures as are we going the right direction? And we have some

01:44:24.920 prognostic value specifically from them. We use HRV to say, are we more likely to be making

01:44:29.960 improvements on the short term or are we heading the wrong direction? Where if we keep doing that

01:44:34.240 for a long period of time, we're going to either see the benefit or the cost depending on which

01:44:37.860 direction we're going. So I think they're different. Is there a number, Joel, on the log normal

01:44:43.080 transformation you're doing on the RMSSD, a number below which, if it's consistently there,

01:44:49.760 I would want to inquire more. 50s, 40s. I mean, those are pretty low numbers for Morpheus.

01:44:54.920 Again, just for folks listening, that's not the RMSSD number. The RMSSD number might be

01:45:00.960 far, far lower than that. Yeah, exactly. Yeah. So you really do have to be specific

01:45:05.480 of what we're talking about, the systems. This is a NM1, but I was at a guy, there was

01:45:09.220 a guy named Mel Sif who you probably aren't familiar with, but he was a very well-known sports

01:45:13.720 scientist. He wrote a book called Super Train. I was at his house with the original HRV system

01:45:18.040 I was using and had all the metrics. And his RMSSD was like five and he just looked really bad.

01:45:24.920 Then he had a heart attack. And I said, Mel, this looks really concerning. He kind of brushed

01:45:29.880 it off and he died of a heart attack a few months later. And so if you're really seeing

01:45:34.860 this super low RMSSD or Morpheus numbers, it's definitely an indication that that autonomic

01:45:40.620 nervous system is not responding well to the world around it. If it's really, really low,

01:45:44.700 there could be a legitimate medical concern that's driving that. So if you're Morpheus 40s,

01:45:48.920 40s, 30s, and you're just not getting up, it's probably worth looking into. And it's

01:45:53.160 something to definitely be concerned about.

01:45:55.040 And do you see the opposite where really high numbers or a sudden change from low to high

01:45:59.160 can indicate an arrhythmia or something like that?

01:46:01.780 Yeah, absolutely. You definitely do see these weird numbers come out that could be arrhythmia,

01:46:06.340 that could be medications, that could be who knows. You see some of these things jump around from

01:46:10.760 anomalies that you just kind of like, oh, it's an artifact. I don't know what happened. And then

01:46:14.120 sometimes you do see a medication or you see weird stuff with COVID and you definitely see some weird

01:46:19.180 stuff in the data that you want to be aware of that you probably wouldn't pick up otherwise.

01:46:22.860 I would say HIV is more of a leading indicator and kind of how you feel and some of those symptoms

01:46:27.000 can come days later. So you'll oftentimes pick up something that looks weird. And then two days

01:46:32.720 later, you get a cold or you get flu or you get some sort of thing that explained it.

01:46:36.960 Good, good. Joel, this is awesome. I feel like I know much more about this now than even after I did

01:46:41.960 the AMA a couple of years ago, and I learned a lot to sort of prepare for that. So appreciate it.

01:46:46.900 And again, I apologize because on some level, this for some of the listeners probably sounds

01:46:51.240 like a bit of a Morpheus commercial at times, but obviously I have no affiliation with Morpheus

01:46:55.800 at all. But I point to it because again, I'm a big consumer of data. I test every device out there

01:47:04.160 extensively and I view it as the stickiest device I've ever found for this use case.

01:47:10.960 And so I do hope that people check it out. I think I just bought it online. You buy it direct

01:47:16.900 to consumer, right? I don't think you buy it on Amazon or whatever. So I would encourage people

01:47:21.080 to, if they want to take this to another level, that they should check out Morpheus. I appreciate

01:47:26.200 the work you're doing, Joel. The curiosity and inquisition with which you bring to this is

01:47:30.620 fantastic. I benefit from it greatly. And those are my patients who use it also as well. So thank you

01:47:35.220 for both all the work you're doing and obviously coming here to talk about it.

01:47:38.160 I appreciate being on here and being able to speak to your audience. I think the biggest thing I've

01:47:41.680 learned after 20 plus years of coaching is we're all different. And as much as it may be easy or

01:47:46.780 attractive to say, oh, this is what somebody else is doing. Let me just try that. Let me just do the

01:47:50.300 same thing. If you can spend the time just copycatting someone else's workout and really dig into your own

01:47:56.720 physiology, how you respond to training, how you respond to diet, how you respond to different

01:48:01.060 food intakes, really learn how your physiology works, you will reap the reward to that in the

01:48:06.460 long run. I think that's where tools like HRV and sleep monitors, they build awareness. And whether

01:48:10.860 you're using Morpheus or URA or whatever else, using that data to really be your own coach and

01:48:16.640 really understand your own physiology and your own biology, that's the ultimate way to get better

01:48:22.520 results and have the health outcomes you're looking for. So I appreciate the opportunity to talk about

01:48:26.220 that. Thanks, Joel. Thank you. Thank you for listening to this week's episode of The Drive.

01:48:31.660 It's extremely important to me to provide all of this content without relying on paid ads.

01:48:36.640 To do this, our work is made entirely possible by our members. And in return, we offer exclusive

01:48:41.880 member-only content and benefits above and beyond what is available for free. So if you want to take

01:48:47.660 your knowledge of this space to the next level, it's our goal to ensure members get back much more than

01:48:52.600 the price of the subscription. Premium membership includes several benefits. First, comprehensive

01:48:58.440 podcast show notes that detail every topic, paper, person, and thing that we discuss in each episode.

01:49:05.040 And the word on the street is nobody's show notes rival ours. Second, monthly Ask Me Anything or AMA

01:49:12.260 episodes. These episodes are comprised of detailed responses to subscriber questions typically focused

01:49:17.800 on a single topic and are designed to offer a great deal of clarity and detail on topics of special

01:49:23.500 interest to our members. You'll also get access to the show notes for these episodes, of course.

01:49:28.580 Third, delivery of our premium newsletter, which is put together by our dedicated team of research

01:49:33.960 analysts. This newsletter covers a wide range of topics related to longevity and provides much more

01:49:39.900 detail than our free weekly newsletter. Fourth, access to our private podcast feed that provides you with

01:49:47.360 access to every episode, including AMA's sans the spiel you're listening to now and in your regular

01:49:53.500 podcast feed. Fifth, the qualies, an additional member only podcast we put together that serves as

01:50:00.440 a highlight reel featuring the best excerpts from previous episodes of the drive. This is a great way

01:50:05.940 to catch up on previous episodes without having to go back and listen to each one of them. And finally,

01:50:11.140 other benefits that are added along the way. If you want to learn more and access these member

01:50:16.120 benefits, you can head over to peteratiamd.com forward slash subscribe. You can also find me

01:50:22.920 on YouTube, Instagram, and Twitter, all with the handle peteratiamd. You can also leave us

01:50:28.500 review on Apple podcasts or whatever podcast player you use. This podcast is for general informational

01:50:35.060 purposes only and does not constitute the practice of medicine, nursing, or other professional

01:50:39.680 healthcare services, including the giving of medical advice. No doctor patient relationship

01:50:44.580 is formed. The use of this information and the materials linked to this podcast is at the user's

01:50:50.760 own risk. The content on this podcast is not intended to be a substitute for professional

01:50:55.360 medical advice, diagnosis, or treatment. Users should not disregard or delay in obtaining medical

01:51:00.980 advice from any medical condition they have, and they should seek the assistance of their healthcare

01:51:05.860 professionals for any such conditions. Finally, I take all conflicts of interest very seriously.

01:51:11.340 For all of my disclosures and the companies I invest in or advise, please visit peteratiamd.com

01:51:18.340 forward slash about where I keep an up-to-date and active list of all disclosures.

01:51:41.340 Thank you.

01:51:48.340 Thank you.

01:51:50.340 Thank you.

The Peter Attia Drive - June 10, 2024

#305 ‒ Heart rate variability： how to measure, interpret, and utilize HRV for training and health optimization ｜ Joel Jamieson

Episode Stats

Summary

Transcript