00:00
Hey everyone, it's Mikki here, you're listening to Mikkipedia, and this week on the podcast, I speak to Marco Altini.

00:21
Marco is a PhD and expert in heart rate variability. So unsurprisingly, this is what we spend the majority of our time speaking about today. We discuss the definition of HRV, the best conditions for measuring it, and the pros and cons for different ways of measuring it. We discuss the impact that lifestyle factors have on HRV and what control we might have over them, and also whether sex differences exist

00:51
how clinically meaningful or otherwise these might be based on the research. We also discuss HRV biofeedback and the impact that simple actions we can take for improving our stress response has on our overall HRV. Finally, we discuss wearables and there's just such a appetite out there for measuring everything. So, you know, what are some of the limitations or benefits that tracking sleep can have

01:21
on our overall health and wellbeing. So this is a great conversation. Marco is an absolute expert in this field and I was super stoked to have the opportunity to speak to him about it. Marco holds a PhD in applied machine learning, a master's in computer science and a second master's in human movement sciences and high performing coaching. And we discuss all about his background in this conversation.

01:48
He has published more than 50 papers and patents in the intersection between physiology, health, technology, and human performance. He is co-founder of HRV for Training, advisor at Aura. He is a guest lecturer at VU Amsterdam and editor for IEEE, Pervasive Computing Magazine. He loves to run and absolutely follow him on Strava to see what he is up to.

02:17
He also has such an informative blog where all of his previous sort of podcast guest appearances are held in addition to a number of his articles, which his sub stack, which he publishes very frequently. And Marco also, of course, is the creator of the HRV biofeedback app. And we discussed this in our interview too.

02:42
Just before we crack on into the podcast, I'd like to remind you that the best way to support the podcast is to hit subscribe on your favorite podcast listening platform because that increases the visibility of Micopedia in amongst the thousands of other podcasts that people could choose to listen to. And that way, more people get to learn from the guests that I have on the show. All right, team, enjoy the conversation that I have with Marco L.T. Ney.

03:13
Thank you so much for taking the time to speak with me today. I really appreciate it and I'm really looking forward to delving into the science of heart rate, heart rate variability and how we translate that into applicable information. I've been aware of HRV as an athlete for probably, you know, at least, I don't want to say 10 or 15 years or something like that.

03:42
But I've never really quite understood how I might apply it to myself, maybe because I'm not a data person. So I've looked at it and went, oh yeah, that's interesting, but have never really taken it any further. So this will be super interesting for me. Yeah, glad to hear and glad to be here. So let's see if we can clarify some of these aspects related to Hattlinit variability. Yeah. So

04:10
Firstly, a little bit about your background, which I found super interesting, was that it appeared you started your studies in computing and engineering, like quite removed from sports science and then you came to data. Can you just chat to us about how that came to be? Yeah, yeah, for sure. About maybe, when was that? 15 years ago, 20 already. I was studying computer science.

04:41
And yeah, I would say nothing really clicked until I took this course in which we would play with these sensors. We're just early prototypes of what we have on the market today at that point that could measure things from the body. And that was really fascinating to me, a lot more fascinating than many other things we used to do as computer scientists. So I got more into this field of...

05:09
trying to monitor activity of the heart, of the brain, the muscles, all sorts of things with these sensors. And from there, then I started looking more at the data. I myself got a bit more passionate about endurance sports and monitoring certain aspects and trying to see what we could understand related to different stressors and training responses. And then maybe also how

05:39
as a recreational athlete, but someone that was quite, let's say, committed with his job and work, how non-training related stressors would actually impact me maybe more than training related stressors. And then how we could use data in particular, resting heart rate and heart rate variability to capture these stress responses and possibly implement some changes. So as I, let's say, picked an interest into these kinds of things.

06:08
Later on, after my PhD and after working a few years, I went back then to university to study also sports science and high performance coaching to basically understand a bit better both aspects like the technology, but also the physiology and how we can actually use these things. Yeah, that's a bit how it happened. Yeah. And you mentioned that you sort of became more interested in endurance sport along that sort of journey.

06:38
When you first got into it, were you thinking more, not in the endurance sport, but the sort of senses, were you thinking more along the health sort of side of things at that time that then sort of transgressed into sport? Or was it sort of always for you a little bit about sport and then you thought, oh, health implications? Yeah, I would say the place where we were developing this census and application, it was more health oriented.

07:07
Yeah, just for example, tracking physical activity in the context of being active as something that is good for you and for your health, but not necessarily related to athletic performance. So it would be physical activity and fitness related to health. And then this quickly translates, I think, into the people that maybe are really passionate about certain sports and then try to optimize their performance, I think.

07:35
many commonalities that are in this context when we start to look at exercise and responses from the body. Yeah, yeah, super interesting. Marco, can we sort of start with the 101? Because I think that there'll be people out there that have heard of heart rate variability, yet don't quite understand what it means and sort of what it measures. So can we start there? Yeah, so when the heart beats,

08:04
at a constant frequency. I think that's the first thing we need to understand when we talk about heart rate variability in very simple but also technical terms. We are looking at this variation in heartbeats. So if we measure our heart rate over a certain period of time, let's say a minute, and to make things very simple, let's say our heart rate is 60 beats per minute. So we expect a bit approximately every second as there are 60 seconds in a minute, but.

08:34
It is not exactly every second that we see a bit. There is always some variability between consecutive heartbeats. And that variability is what we quantify. So let's say that technically it's just a measurement of the variability in heart rate over a certain period of time. And the reason why that is interesting is that that variability is not due to chance. It's not random. It's not always the same. It's actually...

09:03
changing based on how the autonomic nervous system responds to stressors. So since we are interested in measuring our stress response, but we cannot really measure stress, this somewhat abstract entity, and then we cannot measure the nervous system either. So we cannot go there and say, hey, what is our...

09:30
parasympathetic nervous system activity, which is the part of the nervous system in charge of recovery and rest. So we cannot just go there and capture that. But the autonomic nervous system is modulating heart rhythm. So it is changing this variability that we just talked about in a way that when we go and measure heart rate variability, we are indirectly capturing the activity of the autonomic

09:59
in response to stressors. So that's why it becomes a useful metric. As we face a stressor, the autonomic nervous system has certain response, which impacts, modulates heart rhythm. And then we measure heart rhythm, heart rate and its variability. Typically, for example, as we face stress, we will have that parasympathetic activity is suppressed. And this results in a more...

10:27
stable and constant, let's say, heart rate. So the variability is actually reduced as we face more stress, and it is increased as we face less stress.

10:42
That is super interesting because I like almost seems counterintuitive to me that more variability is a good thing because you think, oh, no, we want things to be stable and constant and predictable and anything that throws that out must not necessarily be a good thing. But you're saying that the parasympathetic nervous system.

11:07
its ability to respond is reflected in that variability and increases the variability. That's right. Eventually, I think we need balance if we look at the bigger picture and stability, but we achieve that through change. We need to be able to adapt to the different stresses that we face. We are not always in the same condition. We might be exercising or facing

11:35
we need to be able to respond and adapt to that and then renormalize hopefully quickly. That would be a good response. And so that higher variability is associated typically with a better type of response in these terms. Yeah. And did you say that we don't have a good way of assessing parasympathetic nervous system response outside of heart rate variability?

12:05
Yeah, that's right. This is not in non-invasive ways, let's say. So as a non-invasive tool, total invariability is the best we have. We could try to look at certain hormonal changes as well. That happens at the same time, right? For example, maybe if we could measure cortisol or other stress hormones, that might be similarly insightful. But we cannot really do that with the required frequency, right? Maybe...

12:32
We can do it once in a while in laboratory settings, and it's still a rather complex and inefficient and expensive procedure. And I think it's a bit like HRV maybe many years ago, when you were taking a measurement at the lab with certain equipment, and then you were taking another one after four months of a study. And I think that's so far from how we do things now, which is really measuring every day.

13:00
capturing much better how the individual is going through different stressors. So, Marco, on that, which is sort of a tangent, but I'm curious, is to like, do we, has there been research looking at hormone response and HRV to sort of show any like association with say cortisol or anything like that? Yeah, so there is a research. I think there isn't much good research. That's what I would say. Where you have

13:27
Let's say enough data points from different people going through different stressors, looking at the data before and after, how it changes over time. There's a lot of that for our capability. I would say less for cortisol due to the challenges of acquiring high quality data in let's say a more continuous manner. There are associations of course. They're not necessarily the same things, right? So it's different responses. There's a autonomic nervous system response and then a normal response. The timings might differ.

13:57
but they're all, I would say, ways to look at stress. In similar ways, for example, now you have continuous glucose monitors, right? And those are meant to track glucose in response to food intake, for example, or how it changes when you exercise and things like that. But in fact, that's also something that captures stress responses. And we've seen in data...

14:24
maybe more frequent spikes or higher spikes or a higher baseline value in a period in which people are more stressed, which might also be associated with a lower HRV. So I think many of these things, if we look at them from different angles, we are able to capture similar responses or associated responses because still it's what is happening in the body after all. Yeah.

14:53
that you've got to giving you a much more complete picture, I guess. Yeah, yeah, for sure. So a high HRV is a good thing? So in general, that's what people say. Let's say that I think our understanding has evolved a bit in the past few years. And what I mean is that we used to look at maybe average HRV of a group of people. For example,

15:22
someone with chronic disease and someone that is healthy. And we would say, okay, the healthy group has a higher HRV, higher is better. I think this is where it starts from, is from comparing groups of people. But when we look at individuals, I don't think it's that simple. I think for HRV, as for maybe any other parameter, in fact, if you look again as we were just talking about glucose and we all know that, you know, there's an optimal...

15:51
range we can call it, right? So too high or too low is not great. The same we can think for blood pressure and many other parameters, right? Pretty much any marker might have that optimal range. And I think for HRV is a similar thing. The challenge there is that there isn't really a frame of reference that works for everyone, like a universal frame of reference.

16:18
have for certain other parameters, or like we pretend we have for certain other parameters, where maybe actually also there we should individualize a bit more. But still, we use, again, this is the range where you should be, otherwise you are prehypertensive, or pre-diabetic, or things like that. For HRV, we don't have this frame of reference. So typically, we need to collect some data and see where we stand. And then we determine our own optimal range.

16:45
And when we exit from that normal range, I think it's always a situation in which we might want to be a bit more cautious. So if we have a suppression, it is typically higher stress, like we were saying at the beginning. So more stress typically leads to more constant heart rate, lower heart rate variability, and maybe a slightly higher also resting heart rate. But then also particularly high values.

17:10
might be associated with responses that are not ideal. So it's not necessarily higher or it's better. Let's say that maybe over long periods of time, a trend that slightly increases, that's okay, but a very different acute response, that's not necessarily the case. We can see that for example, in the case of large suppression in resting heart rate, we normally know that

17:38
a slightly lower resting heart rate is also a sign of lower stress. But if you have a large suppression, that may be a sign of acute fatigue and similarly peak in a daily HRV score that is quite far from your normal might also signal something like that. So I would say not necessarily always the higher the better. But yeah, hopefully this gives an understanding of where does that come from and what differs when we look at the data daily.

18:09
Yeah, for sure. Because so as I'm here, as I'm understanding it, then it's understanding your own data and finding your sort of average. And also, I suppose, at a time where other biometric markers might also line up that the fact that you are, you know, lots of other things are good for you, like your sleep is good, and your energy is good, and your recovery from training is good. This is your HRV. And I suppose that

18:37
changes in these things or change in your HRV might give you insight into other things that are going on. Yeah, yeah, exactly. And the thing about HRV is that it's sensitive to so many things, but it's not specific of anything, right? So that's why if I look at your data, for example, and I see suppressions or changes, I cannot possibly have any idea about what has caused them, because I don't have the context and anything could be behind it. But

19:05
when we look at our own data, I think that it's different, right? Because we have all the context and then we can understand a bit better. Hey, maybe here I was traveling or here at this stress at work, or I did this training session that was very high intensity. And then we can try to understand what leads to suppression and maybe how to avoid them in the future. Yeah, it's interesting. This is a slight tangent, but my heart rate has always been higher than.

19:33
someone else might be for the sort of given fitness level that I'm at, and for myself, and also in comparison to other people. So I appreciate that, that you have to find your own sort of individual range. I used to, when I was a teenager, would go into the gym and they would do fitness tests, and they would put a heart rate strap on you, and you'd jump on these awful bikes because it was back like in the 90s.

20:03
I would feel so fit, but I would fail every single fitness test. It was so demoralizing because my heart rate was higher than the sort of ranges would be. Um, disappointing, but I appreciate that individual variation that you're just, you're talking about. Yeah. Marco, I suppose one of the questions I have and I had was obviously you've got heart rate variability, which is sort of a step, I guess one step removed or.

20:32
back from that is just your heart rate. Does it tell you a lot? How much more information do you get from HRV than you would otherwise get from just taking your resting heart rate or your recovery heart rate after exercise or things like that? Yeah. I think, let's say, potentially still up for debate. Not everybody agrees, but I would say that in general.

20:58
When we look at resting heart rate and HRV, there is definitely a strong negative correlation. So as one changes, the other tends to change in the opposite direction. At the same time, heart rate variability is a more sensitive marker of stress in my view and based on the data we have collected over the years. And I think also based on the physiology, basically the way this works is that...

21:26
especially during part of the breathing cycle when you exhale, the parasympathetic system is more active and that's when it's quickly changing beat to beat timing to a way that it's slowing down heart rate as you exhale and then it speeds up as you inhale when the parasympathetic system is impacting the data less. So this change that happens on a breathing cycle.

21:55
When you go and just look at resting heart rate, it disappears completely because by definition, heart rate is just the average. So it doesn't matter if you are slowing down and then going faster later during a breathing cycle. That is captured in heart rate variability, but it's not in heart rate. And you can see that very well actually when you do deep breathing. You do deep breathing, which is supposed to stimulate parasympathetic activity. Your HRV will get much higher, might double.

22:23
and your heart rate is probably the exact same, might change a bit, but it's really similar because this variability is not captured at all by the average. So I think that's why physiologically, it is a bit of a more useful marker. And then in response to different stresses, we looked at it and we could see that HRV changes by a greater extent in response to things like training or...

22:50
sickness, the menstrual cycle, alcohol intake, so all sorts of stressors would lead to a larger percentage change in HRV with respect to heart rate. And so I think that in many circumstances we might see no change in resting heart rate, but HRV is a bit more sensitive and might see a change there and that might be an important signal in terms of the stress response.

23:16
discuss with you some of those things you were talking about with regards to the biofeedback, which of course we'll talk about the breathing, but also the things which impact HRV. But first Marco, how do we measure it? I mean, I know I've got a Phoenix, a Garmin, 7S, whatever, and I know I have the ability to measure HRV, although for whatever reason it doesn't.

23:43
And then of course you've got the Aura ring and I know you're an advisor for Aura. People love WOOP for their sort of data. What is the best way for us to measure and at what time of the day? Yeah, so I think there's a couple of ways that we can use these days. One is simply a morning measurement and I myself build a tool that does that, right? The HLV4Training app. So you would use either your phone with a camera or.

24:11
linked to a chest strap, a polar strap for example, typically works very well. That way you wake up, you take your measurement. We recommend to do that while you sit up. I think that can be a bit more useful than measurements taken lying down or during the night when you are asleep because as you sit up, basically you challenge the body with a little stress. The body has to adjust quickly to the change in body position and that basically amplifies a bit the stress response.

24:40
So you might be able to capture with higher sensitivity, even your HRV in that position. So that's what I normally recommend. An alternative, as you mentioned, is many of the wearables that are out there today. So the Garmin watches or the BOOP or the O-ring, they measure during the night. They provide you an average of the night typically, or something close to an average of the night. And that's also a good way to track resting physiology. I would say both are valid. I think there are some differences. Right, one I discussed now.

25:09
terms of maybe sitting up being more sensitive to stress. Other differences are that, of course, the night comes before the morning. So if you have some late stressors, you train in the evening or you add some alcohol or you ate a bit late, even things that don't really matter so much like just having dinner a bit later than your normal time might have an impact, stronger impact in the night data.

25:37
which in turn becomes a bit less representative of your readiness of the day. So it's more linked to your previous behavior than to what you can do in terms of training, readiness and performance on a given day. In that context, maybe the morning data reflects that better. So I think we just need to understand a bit the differences and then each individual has their own preferences, right? Some people maybe they just cannot measure in the morning, maybe they have young kids or sort of other things to do.

26:07
So maybe they prefer a night device. Other people do not like to wear devices in the night. So they might just prefer a morning measurement. So I think there's pros and cons and some differences, but as long as we measure either as we sleep in the night or first thing in the morning, both are considered measurements of resting physiology and can be effective at capturing our stress response and make some adjustments. Okay, how accurate is it?

26:34
Is it more important for me to look at my garment as a trend over time? Or can I reliably, if I was to start wearing my watch to bed, which that explains why I don't get my HRV because I never wear my watch to bed, it feels uncomfortable to do that. Yeah. It's pretty big. Yeah. Um, but, and like, is it more the trend that I should be interested in?

26:59
So I think both can be important, both the daily data and the trend. Of course, the trend comes from the daily data. So the daily data needs to be reasonably accurate. If it's not accurate, then the trend also will not be so accurate. These devices are getting better and better. And I think especially during sleep, when there is absolutely no movement, basically, they tend to be accurate or in the morning, of course, you measure for a short time so you can just sit still. These devices do not work very well when there is

27:27
any movement or even just any muscle activity. We know that even just for heart rate during exercise, they're not so great. But for HRV, all those problems are a lot worse. So it's really not possible to measure HRV accurately with optical sensors unless they're not moving at all. But in the night, that is fine. So they tend to be quite accurate. And I'm saying that it's important to look both at daily data and the trend, because while, yes, the trend and how things are going.

27:57
I think it's more important and we don't want to be overly reactive and always change things on a day-to-day basis. We want to have a plan and then make some small adjustments here and there. But the daily data can also be insightful at times. Maybe we're getting sick or there is something like that. We of course want to maybe look at the data at the right time, so when it's happening and not just wait for the trend to change, if we are not feeling well, for example.

28:27
Maybe daily suppression can be just a way to reflect a bit more with ourselves and be a bit more aware of, hey, how am I feeling? Is it something coming up and getting sick or something? Or is it just, I'm feeling well and it's just a depression that is my due to some minor stressor and I don't need to change anything in my day to day. I think that it is not there to replace our brain, right? And the decision making process should just be a combination of.

28:56
looking at the data and how we feel and then trying to make some small adjustments here and there. No, I like that because you're essentially saying, yes, we use the data, but we also have to use our own understanding of the situation and actually not ignore our own personal signs either. For sure. Yeah.

29:26
picks up that they might be getting sick before they actually get sick. Is that using HRV as an indicator? Yeah. So I think that in general, let's say there might be changes in our physiology before we develop other symptoms. So this is not always the case. It's not that a device can predict that you will get sick. It can only detect it.

29:56
it had already happened in your body. Maybe you haven't noticed yet and you haven't developed other stronger symptoms, but something is off. So this is not always the case. Sometimes you develop before the symptoms at the same time and you realize it when the data is also off, when you wake up and you see your heart rate is very high, your HRV is very suppressed and you also don't feel well. Other times maybe you still feel okay, but then you will get it later in the day and in the night data it was already.

30:25
there or in your morning measurement. So I think in some occasions it can slightly in advance show you that something is off. It's always difficult though to determine what it is, right? Is it sickness or not? Because if you feel good, there is nothing in the data that is specific of sickness. They typically show that, okay, we are sensitive to so many stressors and then there is something in the data. Maybe the extent of the change can be informative.

30:54
If the heart rate and the HRV is just a bit different from your normal, maybe it's just a minor stressor. But if it is very different, then that is likely to be sickness because sickness is one of the largest stressors and one that has the biggest impact in our HRV and heart rate. So the changes can be much more dramatic than the day-to-day changes we would normally expect otherwise. Yeah. And, you know, we were talking about the Garmin, the Aura, the Woot.

31:23
In your sub stack this morning, you said that Apple Watch wasn't a good... I don't know if I should out Apple, but is it... So obviously the wearables are different in how they measure it, but you're not necessarily a fan of how the Apple Watch does it. Was that... Yeah, that's right. Would you like to sleep a bit? Yeah. No, that's okay. The reason there is that the protocol differs from all the other wearables that we mentioned. So it's really important if we collect data.

31:50
that we do it according to certain protocol, which could be either the morning measurement or if we look at the night, then we need to look at the entire night of it. The Apple Watch is sampling at random times, so it could give you a few data points here and there. Also during the day, which is not particularly informative, especially when it is decontestalized and again at random times, not when you're actually looking at a specific structure and maybe measuring before and after, that's a bit different, that can be meaningful.

32:19
And in the night also, if you have just a few data points, depending on, for example, in which sleep stage you were in, that the data could be dramatically different if it is collected one minute before or after. Because, for example, when you're in REM sleep, your HRV is dramatically different from when you are in deep sleep. That's actually why we can try to guess sleep stages using wearable technology. It's based also on how your HRV is changing together with other parameters.

32:49
If you use a device that measures in the night, I think it's very important that that device is measuring continuously and that you can get an average. So all these changing that happen has basically averaged out because if you try to just get a few data points here and there, the data is very noisy. This is why I think the Apple Watch is not doing this well. To be fair, it's not even trying. It's not that the device is marketed as a device to track your night HRV.

33:17
Apple Watch measures a number of things in different ways and they also measure HRV at random times and granted to health. It's a bit of a different thing than I think getting an wearable that was designed for that purpose to get you night HRV in relation to your recovery. So I would go with one of those or a morning measurement more than the watch in this case. Yeah, I find it really interesting actually that Apple has that approach to it just because they

33:46
I mean, I've got so much of what I own is actually Apple. I know a lot of people like that because they do, they've got a lot of amazing products and features and you'd think that they'd wanna try and ace everything. But from what you've just told me there, and also I listened to you on another podcast where you mentioned that even the algorithm I think that they use for HRV is different from everyone else. And I wondered whether they did that from a point of difference perspective.

34:14
But from what you're saying, it's actually just, they haven't really thought much about it. Not really a priority. Yeah, that's right. I think it's not just relevant too much to their business. I also have a sort of Mac and iPhone and things like that. And I love these products, but the Apple Watch is not what I would use to measure my resting physiology. And the feature they use, the metric is different. I think because that is what is used mostly in medical practice. So it's not that it doesn't make sense. It's just not what is used normally.

34:44
these days to capture stress responses, which is another parameter which is more effective in this context, while in the old days another one was used and they were using that one as well. And yeah, you know, all these other wearables, they specifically want to track your HRV as part of what they are offering to you. And the Apple Watch, you know, they sell millions and millions even without HRV, right?

35:10
It's completely irrelevant, I would say, to their business model. So I think it's a bit of a different story there. Yeah. Marco, you sort of ran us through when best is to measure. So either overnight or in the morning. Now I am sure I heard on a podcast, and it might, some super amazing triathlon coach said that he gets his athletes to measure their HRV in the afternoon before training.

35:40
is that's a better indicator of their readiness than the HRV measured in the morning. Can you speak to that at all? Yeah, for sure. So I think that if we want to assess readiness for the day, we need to measure in the morning or maybe in the night. That's because it's not that the morning is special. It's just that anything affects heart rate variability. So if we want to measure in a state that is reproducible every day.

36:09
in the same way, it is not impacted by all sorts of confounding factors, then that is the only moment we have. Because if I measure before training, then maybe I had coffee and then the data is completely useless. So are these athletes not drinking any coffee, not drinking any water for hours prior to the measurement? Because even water will impact HRV. Are they not eating anything? Because if you eat and then you're digesting, your heart rate is a bit higher, your HRV is suppressed.

36:38
you're not moving because if you are moving around, then your HRV is also impacted. Any psychological stressor, they are on Instagram and someone wrote something that was not so nice and now they are upset and their HRV is very suppressed. So pretty much anything impacts your data in a way that I think we really need to have this routine and context in which we do not get impacted by all these confounding factors. And that's why we do it first thing in the morning.

37:07
And then later on, we can use different protocols to look at training specifically. For example, we could measure just before training and just after training and compare those data points, but we are not using the data point before training as an indicator of readiness. We just look it to compare it to the one after training to see the change due to training. So, because there is research that

37:37
shows for example that if you do train at a low intensity, your HRV is back to values that you had prior to training within a few minutes. So that could be a check if I wanted to train below first lactator ventricular threshold, so just say easy, easy pace, easy training. Then if I measure after training and my HRV is still suppressed after a few minutes with respect to what I had before, then probably I overdid it a bit.

38:05
So that could be a check that you do just before and after, but that's not really about the readiness for the training. I think the only way to do that is you to do it far from or sort of stressors and then we need to do that first thing in the morning. Okay. And potentially I misunderstood what I was reading because of my ignorance. It could be the case because some people, I think try different protocols and maybe they had some success with it, but I think it's really difficult to.

38:31
set all the right constraints to do that free from all the confounding factors. Yeah, sure. So you mentioned earlier on of all of the different stresses that do impact on HRV. How does diet impact Marco? Well, I think diet as well is an important part of daily life and the potential stressor.

39:01
might have an impact, even depending on, let's say, the type of diet you have on the structure. There isn't as much research, I would say, in general. Some diets have shown potentially to improve HRV, but it's, I would say it's a bit like saying physical activity improves HRV.

39:27
meaning like, okay, if you have a Mediterranean diet, a healthy diet and the typical things that we know are good for health, they improve a bit of everything, including maybe HRV. But it's not been, I think, individualized the way we look at HRV now, with respect to maybe other stressors that are a bit easier to quantify or isolate so that we can look at how the data changes, that it is complex. There's many different parts of it, even just tracking it can be complex. And then,

39:56
that makes it I think a bit more difficult to study. Yeah, I mean, you mentioned the continuous glucose monitor and the association between the glucose spikes and HRV, which is super interesting. I think that I also saw a study looking at that sort of detailed that out. I've heard that food intolerances could show up in HRV because of an immune response to

40:25
to the food, is it a thing? Yeah, right. I think it could definitely be the case. It's in this case, also a stress response of the body that might change something in physiological parameters like your resting heart rate and injury in a way that you can capture that. So I think, yeah, many of these aspects can show up in the data. Can be more difficult to understand maybe what is a normal change, because still, you know.

40:52
food and digest, you lead to changes in our physiology. It's not that we should not expect any change or no suppression. I think we need to understand also that stress is good and then we respond to stressors. But then some cases, it might not be so good, like if you are intolerant to something specific, and that might show up with maybe a larger change or a different response. And then we might be able to learn from that and make some adjustments as well.

41:22
I remember hearing you on another podcast actually, and the host, I can't recall which one it was, was saying that fasting for them was, they had a really good response with their HIV, so it improved it. Whereas, as you said, that's not necessarily the case for everyone in terms of things like fasting, I guess. Yeah, yeah, for sure. I think that can be also very individual. I would say.

41:50
Most of the research probably there shows something consistent with this, maybe a reduction in resting heart rate and a slight increase in HRB, but probably also there is some optimal range in terms of how long the fasting lasts and this kind of thing. So I think you can easily go into a phase of too much stress on the body and then have the exhaled opposite response. Yeah, for sure. And I guess if the fasting results in a calorie deficit.

42:20
and potentially too aggressive for an individual, then that in itself is a stressor anyway. And if it's a chronic thing, that I imagine would suppress HRV. Exactly, that's a great point. Actually, the acute and chronic often is not, yeah, they're not both studied the same way, right? We always look a lot at acute changes because it's easy. You have a study, you make a change, and you look at the change, but then,

42:48
how does it work when you do that chronically for a long period of time? And sometimes things are obvious, like exercise is such a bad stressor if you look at how HRV is reduced and all of that, but we know that in the long term, the chronic effect is actually the opposite. So that's why I'm a bit cautious with now everybody looking at HRV in real time. And...

43:11
going through very simplistic interpretations, low HRE is bad and if your stress is not good and you have to do this and that to counter the period of stress, but you're looking at everything so acutely. And that is actually possibly the opposite of what will happen chronically. And exercise, everybody understands this example, but how about all the other things that you're looking at acutely only and you don't know the chronic effects or maybe.

43:39
Um, yeah, we should really step back and not get into caught up into this cycle. I think of looking at that, this data, this very high resolution where it is not really informative of the long-term, um, health and performance outcomes. Yeah, for sure. So, so, you know, it's, you sort of, you have an interest in, in what's happening, but you don't necessarily react to the number that you see in front of you. And so you have like, yeah.

44:09
Yeah, like a good sort of data bank, I suppose. And obviously you mentioned alcohol. Alcohol impacts, can impact negatively on sleep. Is that sort of mechanism for changing HRV, Marco? Yeah, well, I wouldn't, I don't think I know the basic mechanism behind it. I would say, well, alcohol is toxic for the body. I would put it as simple as that. So eventually.

44:37
regardless of sleep, you will have a change in physiology because you can measure right after you've had alcohol before you have any sleep and you will see the same changes. So it's just a large effect. Your death is actually basically as large as sickness. So I think that's a good indication that maybe it's not the best way to go. Yes. Okay. Noted. Maybe this is why I don't measure my HRV.

45:07
So, and what about sex differences in HRV? You mentioned that the menstrual cycle can impact the numbers that we see. I'd love to know whether or not that's good or bad, or actually it's just, it is what it is, you know, it's sort of neutral. And then, do we know anything about perimenopause, postmenopause in HRV? Has there been any research in that?

45:34
Yes, so I would say that in terms of the menstrual cycle, it's just associated to hormonal changes. So some of the changes that you have in hormones are reflected also in resting physiology. So I would say it is the way it is. It does not really mean anything that has been... Even studies that look at training and performance and how hormones impact that, I think the

46:04
main researchers behind these studies is always that we cannot derive guidelines based on just menstrual cycle phases and things like that. There is so much variability between individuals and even in the same individual between different cycles. So that things are so different every month that basically it does not make much sense to make changes.

46:33
based on something that is so variable. And I think with HRV, it's just context. So if you see a suppression maybe in the second phase of the cycle, that could be normal just due to the hormonal changes. And then maybe having that context allows you to disregard, for example, other stressors. You might otherwise think, hey, was this, this or that? And then you just know, hey, it's probably just a second phase of the cycle.

47:02
Yeah, maybe I don't have to take action. So I think it's just useful context for people that have a regular cycle to know that there can be these changes. But it's not that we have any evidence-based guidelines that would differ in that case. And in terms of the differences between men and women, typically, we have that HIV, I would say, is quite similar across different age groups. It might change a bit.

47:30
with menopause in a way that it might become even more similar between men and women after menopause, maybe again due to the changes that you have in hormones, activity, and things like that. So before menopause, there might be a slightly higher HRV in women. Studies are very consistent on this. Some say the opposite.

47:59
So I would say still at the population level is very small differences. The ranges are very broad. So if we have the typical HRV number that is used is called RMSSD. It ranges from maybe 10 milliseconds to let's say 250 milliseconds is the range it takes. And then if we look at two groups of people, the overlap is almost complete.

48:25
is almost maybe 95% of the values you can find in both groups. Even if you look at not really young or not really older people, there's a lot of overlap. You can never tell which group you belong to based on your HRV. That's, I think, the most important thing is even when we look at people with a certain health condition and people that don't have it, there is a difference, but based on your HRV, I can never tell if you have the condition or not. So I think that's important to understand. There's so much variability.

48:55
That's why there is the old story of building your normal range, understanding what is optimal to you, because it's quite pointless to look at absolute values often. And the differences are, some differences are there, but there is so much variability in the data that it's difficult to assign you to any group just based on your data. Yeah, no, that makes perfect sense. Marco, you're obviously, like you mentioned your first app.

49:25
HRV for training, but as I understand it, you've also got HRV, I'm going to say for biofeedback and I don't know that I've got that right, but perfect, and which is a method people can use to improve their HRV over time. Is that, have I got that right? Can we just have a chat about what that is and how that looks? Yeah, yeah, for sure. So, when we talk about HRV, the way we talked about it so far.

49:54
The goal is really to assess it. So we wanna take a snapshot, we wanna understand where you stand and then make potentially some changes on a day-to-day basis based on where you stand with respect to your normal. This is all about just measuring and maybe making, let's say taking action in terms of all the things that you can do in your life. It could be training, it could be other stressors,

50:24
I would say there's always two angles. One is you reduce the stressor, for example, training intensity, you reduce intensity if your HRV is suppressed, or you try to improve your chance of recovering that could be taking a nap or eating a bit better, sleeping a bit more and things like that. Among the things you can do to try to potentially either speed up your recovery or increase your HRV.

50:51
There is biofeedback. There's another use of HRV in which you are basically just doing a deep breathing exercise. So there's many forms of it. The ones that are called HRV biofeedback, they are typically requiring you to use some piece of technology that shows you your heart rate in real time so that you see how your heart rate increases as you breathe in and decreases as you breathe out.

51:20
So you do this exercise because it's supposed to stimulate parasympathetic activity. So again, that's basically what increases your HRV. And then you do that. Well, the protocols published in the literature actually require quite a commitment because that's twice per day for 20 minutes each time. So obviously you can start with just a few minutes, but otherwise it's quite a lot of breathing that you have to do. And then in I think 10 to 12 weeks.

51:48
they typically show some changes. I would say not necessarily. So I think it's one of those things that is still also up for debate. So you see a change for sure as you do it. So acutely, when you're deep breathing, your HIV is much higher. But does that lead to a change in your baseline as you assess it in the morning or in the night over time? I think it depends. So maybe in some people, maybe in others it's a bit more difficult.

52:17
In some of our research, we saw changes more marked for people that had a lower HRV to start with. So maybe there is more room for improvement there and they could see some changes. I think when we do these things, interventions in general, we also need to remember that there's seasonality in the data. That means that the data is not always the same in summer or in winter.

52:43
There are changes that are a bit outside of our control, even on things that we do. So if we are going towards winter and we start a protocol and then our HIV decreases even, that is actually normal with winter and shorter days and things like that. So we should not think that our intervention and the things that we are doing that are actually maybe positive for our health.

53:09
are not just because our HRE is changing in the other direction. So I think we always need to understand that. Yeah, the scale of changes, it's not always easy to understand and grasp also based on other aspects that might change the data. So that's always important to remember when we look at things in the chronic, let's say, time frame as opposed to the...

53:35
acute changes that are easy to identify and less impacted by these other aspects. Sure. And thinking about breathing, it's almost like the premise of almost every meditative practice. And people relate meditation to feeling calmer and more present and potentially be more resilient against stress. So I can understand how it could be helpful, actually. You're right, though.

54:04
lots of 20 minutes in a day. I was just trying to figure out what I would have to give up in order to prioritize that. But certainly for a few minutes for sure. Do you do two times 20 minutes a day, Marco? I bet you do. No, no, I don't. So I had periods where I tried this more. I did not see an effect on my HRV. But I think that's important to highlight also. HRV is just

54:33
better, I think. I was calmer. I was dealing with stressors better. So I think it's good for you and it is shown by many studies that subjectively anxiety, for example, is reduced and I'm quite an anxious person. So for me, it was helpful from that point of view. Sleep has improved. So there are many good points, regardless of its ability to actually change your physiology at rest. Also, it might be that it changes your physiology when you're challenged by a stressor.

55:01
We don't look at that, but maybe it does and it doesn't address. So yeah, eventually I lost a bit of the habit, but yeah, maybe something to pick up again. Isn't it interesting? I'm like you. When I tried meditation and did a really good sort of stint of maybe three to six months and did feel noticeably different, yet sort of just...

55:28
I lost the habit as well. Whereas there are other things, and you and I are both runners, and I understand you've got a 100K race that you're either beginning to train for or you will be training for. And I would never consider not going for my training session or not eating well to support the training yet. And I think that we're no different from a lot of people, particularly with people I speak to with my clients and stuff, is that prioritizing

55:54
rest management techniques like breathing that of course could potentially impact our HRV over time. It's harder to do for whatever reason, even though there are notable changes and we have them and it's tangible. Yeah, yeah, that's right. It's not always easy. I think that's one of the most important things though, regardless of these practices, even just a bigger picture.

56:20
our days, our work, how we handle stress. I think those kinds of things are really key, right? When we talk about HRV, people are always looking at, hey, how do I improve my HRV? And I already eat healthy and exercise and I sleep well, but then I have this super stressful job, right? Obviously that is the elephant in the room. So I think that's where we need to work. And sometimes the job needs to stay that way. And then maybe these protocols can help.

56:46
Other times maybe we can make different choices, right? We can say, okay, maybe this is not what I want to do for real, it's just, you know, society pressured me into this, but I am actually another person and I wanna take it easy, so that is totally fine. Yeah, yeah, no, that's really good insight. Marco, I'm mindful of your time. I do just wanna ask you about one final topic, which it's a huge topic, but hopefully we can get some insights because I understand you have...

57:15
You recently, I don't think it's available publicly, but published a paper around sleep and wearable technology. And we talked about how you are, well, I mentioned you are an advisor for Aura and devices like Aura and Woot give people metrics on sleep. Can you give us any insight into the state of the accuracy or reliability of that data? Like how much of that kind of data should be

57:44
can we read into as being useful for us? Or do we just, because a lot of people get quite caught up in it. Yeah, yeah, I see that. And I think it's tricky. It's a bit dangerous maybe even. A lot of people get too caught up into a metrics that sometimes we need to understand are completely made up. For example, sleep scores, they're not a thing, right? It's a toy to keep.

58:12
someone engaged with a device is not necessarily something that is linked to your health or performance. There's no study showing that in any possible way. Every device is different. When you take five devices and you get five different things for the same parameter, like the quality of your sleep, then we should remember that maybe this inconsistency...

58:38
inconsistency is telling you something which is that we just cannot capture that parameter. So that's something useful to understand. When we look at wearables, some things they can estimate reasonably accurately, for example, typically sleep time. Not always, for example, in my case, I read in bed a lot, I kindle and I don't move, right? And then typically it thinks that I slept like two, three hours more than what I slept. So it doesn't always work.

59:07
depending on your behavior, it might work better for you. And that's great. But that's sleep time. On the stages, we get better at the accuracy over the years, right? Maybe now we get close to 80%, but still it's not so great. And even if it is better, or even if it was perfect, like what does that mean? Like we don't have any evidence that having a certain...

59:34
pattern of sleep stages is better than another or when we have a certain pattern, we should do something different or even the idea that we should have this amount of time of deep sleep and REM sleep every day, I think is quite flawed because based on our behavior and what we've done, we most likely need even different...

59:57
amounts of different types of sleep, right? If we had a very busy day and we trained a lot and maybe we need more deep sleep than REM sleep, any day could be different. So I think it's fine if you're curious about it and want to have a look. And the devices, some of them are not too off from reference data. So polysomnography and measuring actually your brainwaves, but still you're not measuring your brainwaves. It's just an estimate. So there's always a degree of error.

01:00:27
And then again, we don't really know what to do with that. So I think those are things to be aware of. Some of the scores are made up. You should really not worry too much about those. Remember that when scores are inconsistent between devices, it means that you are not measuring something there because otherwise they would be consistent. This is why when you look at your HRV, for example, that's going to be very similar, especially in relative changes over time with all of these devices, right? Because we are measuring it. So...

01:00:56
That is something that is consistent and you can rely more on. And I think in this paper also with the other experts that are all sleep experts and scientists and doctors working with these technologies and they are also very positive about the use of these technologies and the opportunities that it enables, but also very critical of some of the aspects that are brought up here. And I think all of us think that the really what the wearables have brought is...

01:01:25
more than trying to mimic other devices, like with the sleep stages and trying to do what the polysomnography does. The really advantage is that the device is actually measuring your physiology. So it's showing, you know, how your heart rate and HIV changes during sleep, your temperature, and those are the things that maybe are more relevant, even during sleep.

01:01:52
maybe the focus should be shifted more towards your physiology, how it changes with response to stressors, how it's associated to how much you sleep and maybe the structure of your sleep. But this really maybe the physiology and how that changes might be more insightful and more useful to look at as it is also more reliable and consistent between devices. So I would say very briefly in the context of wearables, there's a bit, I think how.

01:02:21
conversation started in a way, hey, we just struck sleep and this and that, but I think now it's shifting a bit towards, hey, these things maybe are not so useful, but we are looking at your physiology and measuring it and that can be insightful in so many ways. Yeah, that sounds super sensible. And, you know, as I mentioned, I don't measure, I'm not much of a data person per se, in some things, but I almost always know when I wake up, whether or not I've had a good sleep,

01:02:50
or a bad sleep, you know, and that I think is probably more valuable for me than knowing how, what stage of sleep I was in or, and, you know, making some inferences from it, I suppose. Yeah, yeah, exactly. And, you know, devices and scores can mess with our head in ways that maybe it's sometimes better just to not use them. Yeah, yeah, yeah. And Marco, are you a big data? I mean, you sound like you're a big data person.

01:03:20
like track all of these things that we're discussing for yourself? So I'm curious. So, you know, it's part of my job. So I look at things and I know a lot of things. And I think it's fascinating. I don't let it much mess with my head or impact my day, I would say. I do track my issue. We write, obviously, I make these tools and I want to look at the data and see changes. And that's maybe the data that over the years I ended up using.

01:03:47
the most also to implement some changes or to see how changes I implement in my behavior translate to the data. I think for me it's been a bit of a process of just self-awareness. So the data is not there again to replace us, but it's just to help us become maybe a bit more aware of ourselves, think a bit more about our feeling and fine-tune things. And I think that way it can be useful. No, that is awesome.

01:04:17
And you know, we've mentioned, I mentioned your sub stack and your apps that you've created or businesses, HRV for training, the biofeedback. Where can people find more information on this and you and your research? Yeah, I think those platforms are great. And then I use quite a bit Twitter or X. So you can find me also there. And that's it. Yeah, that is awesome, Marco. And I will put links to

01:04:46
all of that stuff in the show notes. Thank you so much for your time, your evening, my morning. I really appreciate it and I look forward to the stuff you've got coming out this year. Good luck with your training, with your race. Sounds awesome. Thank you so much. Thank you so much, Miki. Pleasure talking to you. See you later. Ciao.

01:05:19
Alrighty, hopefully you enjoyed that conversation as much as I did. And we've put links to Marco's Twitter, his sub stack and Marco on Strava also have links to his research page and the HRV for training app. So check them out in the show notes. And also would like to remind you that or tell you actually that I have a webinar that I'm doing on April the 7th.

01:05:47
to Masterclass in Metabolism, Menstrual Cycles, and Menopause. Would love to see you there. Check out the link in the show notes. Or DM me. You'll find me on Twitter, Instagram, and threads @mikkiwilliden, Facebook @mikkiwillidennutrition to my website mikkiwilliden.com. All right, guys, you have the best week. See you later.