#174 – Biomarkers in bloodwork and labs & how they’re linked to metabolic health | Dr. Azure Grant & Ben Grynol

Episode introduction

Episode Transcript

Azure Grant (00:00:06):

When you eat and you think about your triglycerides, they are going to rise within a handful of minutes, maybe even stay elevated for hours after eating. They tend to kind of oscillate on their own in relation to your hunger. So for instance, if I sit down all day long and I eat the same bite of food every 30 minutes or so, I did this and I did this with a couple of friends in school, and I write down every time I feel hungry, maybe rate it on a scale every half an hour, as well as take a triglyceride measurement that often, you’ll actually see when you get hungrier, may be representative of your ghrelin hunger hormone and your trigs will be a little lower. When you’re feeling relatively satiated, your trigs are a little higher and so you can think about this as your body having naturally oscillated over the day, releasing a little bit more energy into your bloodstream. So it isn’t just the food, it’s also the timing and the recent context of, you might not be thinking about it, but what you did yesterday.

Ben Grynol (00:01:04):

I’m Ben Grynol, part of the early startup team here at Levels. We’re building tech that helps people to understand their metabolic health, and this is your front row seat to everything we do. This is a whole new level.

Ben Grynol (00:01:37):

We talk often about blood markers, blood panels, labs. Why do people need to know what’s happening with the different molecules in their body in real time? Well, there are so many different molecules to monitor, and so Azure Grant, part of our research team, she and I sat down and discussed what are some of the markers that people should pay attention to? How can they think about them and what is the importance? Things like fasting insulin. It’s a very important marker for people to understand what is happening with it in their body in real time. Other things like lipid particle size, cytokine, CRP, cortisol, the list goes on.

Ben Grynol (00:02:13):

Even when you talk about women’s health, things like estrogen, progesterone, these are all very important in the cycle of fertility and these change over time as women go through different stages of life. When it comes to metabolic health, some of these hormones and biomarkers are in constant fluctuation and they change depending on the lifestyle factors that we have in front of us. They change and the way that we metabolize food, the way food affects our health is in constant fluctuation depending on lifestyle factors and the way in which these other markers play into them. Anyway, it was a great conversation where we went into all of these ideas around what should you measure, how often should you measure it, why is it necessary and how can people think about it moving forward? It’s a great conversation to jam with Azure. Here’s where we kick things off.

Ben Grynol (00:03:08):

So one of the things we should get into is this idea of blood work and it’s a product that we started offering last fall, so fall of 2021. And we had started experimenting with it internally as a team in May, the spring of ’21, and the idea was very much aligned with this idea of biological observability and something we started talking more about. Josh is notorious for saying glucose isn’t a panacea and a lot of advisors do the same thing where it’s a great molecule to understand how, what are the implications of having higher low glucose levels and how does it impact your metabolic health? But it is, in isolation, it is one factor. It’s measuring the gas in the tank of your car and saying, my car is doing great.

Ben Grynol (00:03:53):

And so we wanted to give people more insight into what is blood work, why does it matter, and how can people go about paying attention to all these different biomarkers? So thought it would be good to get into what are some of the most important markers that people can pay attention to in addition to glucose, and then we’ll go through some of the factors, like what it means if one of the molecules is high or low and we can just sort of riff on all these different things. So again, there’s no agenda. Some of them we might go deep, others we might be pretty shallow in depth, but we’ll take it as it is and start hammering away.

Azure Grant (00:04:28):

Sounds great. So I think Levels overall takes this approach to blood work of doing exactly what you said, saying we already provide a continuous metric that is super important. Blood glucose, but blood glucose doesn’t tell you everything about metabolic health, so we offer also a lipid panel that gives you some of the major players that tell you about liver function, insulin at your pancreas, and what’s contributing to these blood glucose levels that you see. And then we also offer a measure of inflammation called CRP. And these are, I think, a nice broad brush strokes picture of someone’s metabolic health. There’s also a whole bunch of things that we could get into that were kind of dream metrics to have in the future that have more to do with reproductive access hormones, thyroid, things like stress or hypothalamic-pituitary-adrenal axis, like cortisol, and all of these things contribute to the overall picture of health. So our panel is really, I think, just the first step, but all of these are super important.

Ben Grynol (00:05:34):

The wild thing is we often compare the dashboard of a car or we compare the number of sensors that an airplane or a rocket will have, and it’s kind of alarming to think how many sensors does a human have? And it’s like, maybe one, maybe up to 10. If somebody’s a big wearables nerd, they might have a ton, but when you compare it to something like a rocket, a rocket has 10,000 or more sensors, right? Even I think the new Airbus airplanes have 100 or 93k total. I think they have 10k on their wings alone. And so you go, like, this doesn’t make sense, right, as far as monitoring what’s happening in real time. If you take it to the body, there’s 50,000, roughly 50,000 markers and the number gets pretty big, but 50,000 that we can start to think we could mark all of those in real time.

Ben Grynol (00:06:27):

Now we’re not going to because there are multiple combinations of proteins and hormones and the list goes on. But 50,000, what is that? That’s a number of people that fit in the Stanford Stadium. Imagine trying to track every single person in real time to be like, what movements are they doing now? It’s just wild. If you narrow down though, there’s probably 100, hormones are a major, major one that we want to dig into, but in general, there’s probably 100 markers that would give enough of a lens into overall health and wellness to provide a picture of saying what is happening in real time and why do these matter? So let’s hammer through some of them.

Azure Grant (00:07:03):


Ben Grynol (00:07:04):

We’ll get through things like insulin, we’ll get through cortisol, we’ll get through cytokine, CRP, testosterone, estrogen, progesterone, luteinizing hormone. We’ll get through all these and we’ll talk loosely about them and see where we go. So let’s dig into fasting insulin. So why is it important and how is it so directly linked to metabolic health?

Azure Grant (00:07:27):

So what is fasting insulin? Insulin does a lot. It tells the body to store what you’re not using right now. So for instance, when your liver is storing fat, your pancreas has to start working harder and your pancreas normally talks first to your liver before it talks to anything else. That’s actually shunting out its products directly to it. So basically if your fasting insulin is really, really low, that means that you are effectively telling your body to store what it needs to with a lower volume signal. And the problem in the body when insulin is really high, say when it’s over 10, is that it contributes to self-proliferation and self-proliferation growth in a controlled manner in the right place at the right time is a really good thing. Say you’re a kid or a growing teenager, you absolutely want to grow. But if you’re an adult and it’s not the time to be growing, especially in places that you might associate in the body with cancer or in your brain, then you don’t want a really high insulin signal all the time.

Azure Grant (00:08:36):

So basically the standards for insulin are, if we can get it really, really low, that’s a good thing. So for instance, if someone is a endurance athlete, if they’re going out and riding their bike every day or for a couple hours a day or a long distance runner, they might have insulin levels under two, which are at the low end of the measurement scale. But if someone’s up around six or seven, that’s pretty good. And then once you get over 10, you want to make sure you start working on that. But basically we can measure glucose continuously, but what when we think about when we measure glucose is not just how high is it, but the more interesting question is how hard are we making our pancreases work right now?

Ben Grynol (00:09:18):

One of the interesting things around insulin, and we talk about this how it’s so directly related to glucose, we’ll call it glucose variability, is sometimes you’ll see, actually let’s sort of riff deeply on this. This is very much a Dr. Lustig, Rob, our good friend Rob, a hat tip to him. So he talks a lot about you can’t judge a person by body type, nor should you, right, because we all have different body types and somebody who might appear where they’re not air quotes overweight and you go oh, that’s a healthy person. There could be so many things going on. So let’s use that as an example of somebody doesn’t appear to be overweight, they have low glucose variability and they say, great, I ate fruit loops. I mean this happened with my best friend when he was using Levels, he was eating lucky charms, just absurd things, Dunkin’ Donuts, he was sort of testing it out and he was saying hey, I’ve got low glucose variability.

Ben Grynol (00:10:15):

And the outlook is like, man, you’ve got very likely hyper insulinemia. For sure, if I’m guessing you probably have such a high level of fasting insulin that your body is now not reacting to glucose the same way. And so that is, to your point about growth, that’s a super dangerous zone to be in where it’s you’ve got high, let’s say you’ve got fasting insulin over 10 and it’s making your glucose levels look like they’re relatively flat in oscillation. That’s a huge flag. But you don’t know. If you’re looking at glucose in isolation and you’re saying hey, look, I’m doing okay, but the other marker is so far out, that is super dangerous. And so there are all these downstream implications of it and that’s why it is the and as opposed to the or, or in addition to. They’re not a mutually exclusive thing, this is something that you need to look at together.

Azure Grant (00:11:10):

I think it’s also something that if we could show to people live, it could be a really powerful signal of, hey, even if my blood glucose looks pretty good, it looks like my body is working really, really hard, going overtime, sending out signals that it doesn’t really need or it could be detrimental in other parts of the body in order to keep this one other marker in shape. And when we’re blind to a metric, it’s very easy to excuse ourselves or think we’re doing a good job and everything is fine, but it’s absolutely important to see or to see things that can give us an indication of how insulin is working. So for instance, when there’s a high sugar load, the liver makes VLDL in response. So that’s a good indication of how we’re processing carbs, basically turning that sugar into fat. And that particularly tends to offload as something called small dense LDL, which is one of the things that contributes to heart disease.

Azure Grant (00:12:10):

If you think of it as a small particulate that’s in the solution of your blood going through your vessels, it’s easier for that small dense particulate to [inaudible 00:12:20] a solution and start to clog up arteries. So these are all related, your pancreas function, your liver function, how your body takes in sugar versus how your body processes fat. And I think it’s the reason that we need to have the blood work panel in addition to the CGM signal. And there are a whole bunch of interesting nuances about how do we interpret this panel that we might take a handful of times a year in relation to something like a continuous signal, because when we think about how we assess a single time point, it’s a very different calculation compared to how we look at all of the fluctuations and complexity of the CGM signal.

Ben Grynol (00:13:01):

Over time, that’s what matters. That single point in time, there could be so many reasons why that one marker is out. When I had my blood work done, it was in New York in May, so May of this year, and the night before, my fasting period of 12 hours was such a terrible lead in compared to my average, right. We went for dinner and very conscious about the decisions that were made to the point where my wife texted me and she’s like, you remember 9:00 PM you got to be done. I’m like, it’s 8:56, dessert’s coming. We had pasta and bread and dessert, and so my fasting glucose was so much higher than what it normally, I think it was 108 or something. I mean, it was just so high because I was eating till 9:00 PM and had blood work at 9:00 AM the next morning. And you’re like, of course it’s going to be higher than average when the average is eating high fat, high protein, relatively low carb diet. That’s like my personal diet.

Ben Grynol (00:14:03):

And so I know my fasting level’s usually going to be around 90ish, right. And so you’re like, well yeah, that’s the reason, but point in time if that was the assessment at a doctor’s office, what do you think is going to happen? It’s like, oh, we’re going to monitor. Well, actually not even at 108, but if it was higher it would be like, let’s monitor this. We might talk about metformin, we have to look into it. And you’re like, no, it’s because I made a silly decision the night before I was doing something, and that is only because I knew that. If you don’t know these things then you make all of these, you might make long term decisions by not having awareness of, well, that’s why that happens. So anyways, that’s a bit of a rant and a digression, but you highlighted LDL, so why don’t we get into HCL, LDL and particle size that is integral for understanding all of these different lipid markers.

Azure Grant (00:14:55):

Absolutely. So let’s talk a little bit more about the lipid panel. Well, first let’s step back from HDL and LDL and talk about cholesterol because that might be the average person’s entry point. It was certainly my entry point when I started trying to learn about metabolic health a handful of years back. And basically a total cholesterol marker is something that you’re going to get at the doctor’s office. I think of it as a freebie on our lipid panel. So basically it’s something that represents the total of a bunch of good things, HDL and a mixed bag of some good things and some bad things, being LDL, which contains a good and a bad sub-distribution, so it’s kind of a freebie. If your total cholesterol is really, really high, it’s absolutely a signal to look deeper at the HDL and LDL distributions. But overall, total cholesterol is pretty hard to interpret even though, historically, it’s gained so much recognition, as you’ll see cholesterol free foods, statins are incredibly powerful in the United States as an influencer of recommendations.

Azure Grant (00:16:05):

So basically first thing is there’s this total cholesterol marker. It’s got a really rich complicated history. Basically if you see a strange total cholesterol number and it’s really, really high, look deeper, but take it with a hefty dose of salt overall first thing, because it’s made up of some things that you want to be high and other things that you don’t want to be high. If we look a little bit deeper into cholesterol, we have high density lipoprotein cholesterol and low density lipoprotein cholesterol. The first one, HDL, or good cholesterol you’ve probably heard of, is something that helps take fat away from fat cells and to the liver for processing or excretion and it’s actually associated with better cardiovascular outcomes. Whereas LDL or low density cholesterol does the opposite, takes things to fat cells for storage, is overall associated with, I think it’s around a third increase in risk of heart attack over your lifetime, so pretty decent, and it’s actually composed, we keep going down their rabbit hole, of two different categories of things.

Azure Grant (00:17:14):

And that’s because if we think of these densities of lipoproteins, you’re basically looking at little fatty globlets that have proteins in them and they come in all different sizes. If you think about pouring some oil into a bowl with vinegar and you kind of shake it up, you’re going to get droplets of all different distributions. And so that’s what we’re talking about here is how to bucket these distributions into different categories. And so if we talk about the LDL, we have a fluffier kind and we have a stickier smaller, I think of it as more grainy kind, and basically that small grainy kind like we mentioned earlier, that has an easier time falling out of solution in the bloodstream and contributing to things like inflammation and clogs and eventual atherosclerosis. Whereas that lighter sort of fluffier kind has an easier time kind of bumping through the bloodstream, getting shuttled along and not falling out of solution.

Azure Grant (00:18:09):

So it turns out that cholesterol, which we might just think of as one thing and want to reduce, is actually composed of things that we want to have quite high HDL, almost the higher the better there, and LDL where we mostly want to pay attention to do we have small dense LDL at high levels, which is a bad sign. So that’s actually something I think would be really interesting to offer in the future would be lipid particle size distribution, it’s a pretty expensive assay right now and not offered super commonly. Or we have some other options that you can kind of use as a proxy to tell you some similar information.

Ben Grynol (00:18:49):

Yeah, and the key is understanding. It’s one thing to see the data, right, so it’s like, I see the data, and we always say the so what and the now what, and especially when you’re looking at data that is maybe a bit foreign, so this is new to, let’s say, the average person. On average, people might not have looked as in depth unless they’ve had a reason to at those numbers to really understand them. So the key is how did it get that way? That’s what we always want to figure out, whether it’s high or low. The reason it is at this level is because you eat really well or you exercise a lot or you do all of these different things. And when it comes to HDL and LDL, the key is understanding, hey, you’re doing pretty well because on average you eat a ton of olive oil, we’ll just say a ton, makes it sound like it’s hyperbolic. You eat an appropriate amount. The reason I laugh is I eat way too much olive oil and always have my whole life on a daily basis, but it is enjoyable.

Ben Grynol (00:19:50):

But neither here nor there. You eat an appropriate amount of olive oil on average every day. You eat things like walnuts, you eat… All of these things are contributing really healthy omega-3s. Or the other side of it is you crush a ribeye steak every single day, seven days a week. Those are two drastically different outcomes when you start to talk about lipid particles and the type of cholesterol that is in your body. And so yeah, it’s good to break that down. If you want to talk maybe a little bit about food and how food impacts health from a lipid standpoint.

Azure Grant (00:20:23):

Yeah, absolutely. And maybe as a segue into that, I want to call out something that you were talking about about timing and recent context for your measurements because in some ways it’s also about the steak you had last night and what you did this morning. And I think a lot of the reason that people have trouble interpreting these measurements or that you can go into the doctor get one high measurement and potentially be prescribed a statin is, A, because it’s really easy to go into your fasted test not fasted. And when I was in college I would say, ah, I’ve been fasting for 10 hours, whatever, let me go get my blood test. I don’t mind. But being in the fasted state actually really matters and it especially matters for the last part of our panel that we haven’t talked about as much yet, but triglycerides.

Azure Grant (00:21:16):

So triglycerides are very acutely reactive to fat you eat. So if you’re still processing food, you’re going to have elevated triglycerides compared to if you’re truly fasted. So that’s one thing. But these metrics also vary quite a lot by time of day. So if you’re trying to decide when to get your panel, not only should you get it in a context that’s fairly reflective of your everyday life, not the day after you’ve gone out and had an unusually big meal with friends, you should absolutely get it in the fasted state, truly at least 12 hours fasting. And then on top of that I would say even try to get it in about the same time of day and same time since you woke up. And that has to do with not just the food reaction but the circadian rhythm of things like cholesterol, all types of cholesterol by the way, as well as triglycerides, which tend to rise across the morning and into the early afternoon, and that happens whether you’re fasting all day long or whether you’re eating. And these can be 30% changes, even more, 40, 50 and above if you’re eating.

Azure Grant (00:22:26):

So just as an aside to that and to lead into talking about how food affects these metrics is basically if you’re eating, your leads are going to go up. So for instance, when you eat and you think about your triglycerides, they are going to rise within a handful of minutes, maybe even stay elevated for multiple hours after eating. And additionally they tend to kind of oscillate on their own in relation to your hunger.

Azure Grant (00:22:58):

So for instance, if I sit down all day long and I eat the same bite of food every 30 minutes or so, I did this and I did this with a couple of friends in school and I write down every time I feel hungry, maybe rate it on a scale every half an hour as well as take a triglyceride measurement that often, you’ll actually see when you get hungrier, maybe representative of your ghrelin hunger hormone, your trigs will be a little lower. When you’re feeling relatively satiated, your trigs are a little higher. And so you can think about this as your body having naturally oscillated over the day, releasing a little bit more energy into your bloodstream, kind of independent of what you’re doing. So it isn’t just the food, it’s also the timing and the recent context of, you might not even be thinking about it, but what you did yesterday.

Ben Grynol (00:23:47):

Yeah. So it’s such an important thing to highlight especially with fasting. What happens at a cellular level, let’s say 10 hours fasted versus 16 or 18 hours or even 24, if you don’t do the full 12, you’re supposed to fast for 12 hours before blood work, that difference, especially in autophagy where your body is changing drastically in the way that it is starting to really clean itself out, and so you eat a brick of cheese on nachos before… Right, you went out with your friends, you ate the nachos with the cheese and all of these things, naturally your blood work is going to not look great the next day and it’s that point in time. So lots of interesting takeaways in how we have to think about these things because the other side of it is if you don’t, now actually let’s paint the picture.

Ben Grynol (00:24:40):

You should look at the numbers as is, but you should understand what has led up to it. You don’t want to ignore certain data to be like, ah, it’s because of that. You should still take it into account. But the other side of it is maybe not anchoring too much if there are things that are throwing it off because it can be a little bit alarming and you don’t want to overreact or underreact. So overreaction is like, oh my gosh, the extreme case is, I need to go get this taken care of right now when it actually might not be as big an issue, because you’re like, no, that’s related to the brick of cheese on the nachos and all these things. So it’s having the foundation of knowledge so that people can understand it in an accessible way. That’s one side.

Ben Grynol (00:25:19):

The other is like underreacting where you look at it, you’re like, ah, I’m fine. That’s also, we’ll call it the air quotes, dad thing, because I think it’s hilarious where I know my dad is that way. A lot of my friend’s dads are, it’s like, nah, I’m fine, I’m fine. You’re like no, you should probably look into that. But yeah, see, it’s having a balanced view on what it means. So it’s really painting that picture of the now what.

Ben Grynol (00:25:43):

Let’s go into this idea of cytokines. It is such an interesting thing because, well, you go deep on this, but cytokines will allow people to understand, hey, my immune system is doing well. The caveat is if there is a huge surge, that might not be a good thing too. So we’ll say somebody goes into septic shock. Well, you could actually forecast that beforehand because you’re like, oh, that’s a cytokine storm. You’ve got way too many of these proteins that are being marked right now. That is alarming. And you might know this, going into a state of sepsis, you might actually know this in advance being able to see this is what, here is the trend line. So being able to monitor that in real time is hugely impactful in helping to really save lives or helping to mitigate some of the downstream implications. So why don’t we go into cytokines thinking through why we should mark them and even CRP when it comes to things like inflammation.

Azure Grant (00:26:49):

Yes, absolutely. So if we talk about inflammation, we can, A, talk about acute inflammation versus chronic inflammation. And if we think about acute inflammation, that might be something like your CRP, C-reactive protein, something that is elevated following elevations in cytokines. That might show up as a little high on a blood test if you’ve recently had a cold, if you’ve recently had COVID, if you’ve had a urinary tract infection, it can go up for a lot of different reasons. So on one hand I think of acute immune responses as your body doing its job. You actually want that immune response if you’ve been sick recently and if you see that on a blood test, it’s not really something to worry about if you recovered from your sickness later on.

Azure Grant (00:27:42):

But chronic inflammation is a different story. So chronic inflammation either from an autoimmune disease, think of Crohn’s Disease, the whole family of irritable bowel, very common and increasingly common, or if you think of other things like Hashimoto’s, there’s a wide family of autoimmune diseases that are also on rise in the US that even more prevalent in women and these result in chronic inflammation throughout the body, which is a stress signal and is going to lead to more tissue damage and is absolutely something to worry about. So I think it’s another reason to take recent context into account when getting blood work done where, if you’ve gotten your blood work and today was a representative day, you feel pretty much how you feel every day and you see high CRP, then that’s a reason to worry, whereas if you’ve recently been sick, then don’t worry about it so much.

Azure Grant (00:28:43):

But you talked about cytokines and we also think a lot about interleukins as things that we would want to measure continuously. And if in the future we could wave our magic wand and be able to measure inflammation continuously in the blood, then that would give us an option to probably learn a ton more, A, about continuous immune regulation and to do things like see if we are sensitive to foods that we might not register that we’re sensitive to yet, because imagine that something within our microbiome reacts poorly to a food that we’ve just eaten, we might really not sense the acute immune reaction, but we could still theoretically pick it up in our body.

Azure Grant (00:29:30):

So I think there’s a big difference between what we can do with immune markers that are taken on quarterly blood work and then potentially in the future to use acute immune response to be able to understand how our behavior is interacting with our physiology. And you could imagine things like an inflammatory response to a really stressful situation that you got in throughout your day, not just to having eaten something or acute elevation in immune markers as a result of not sleeping well for the last week. And those are the kind of changes that I think would be really impactful for helping people tune their actions and have that more sympathetic response, not in terms of the autonomic nervous system, but in terms of emotionally sympathetic response to what they’re doing every day.

Ben Grynol (00:30:20):

Yeah, we know that chronic inflammation compounds, that’s where a lot of these downstream health complications come from is this compounding of chronic inflammation over time and it just keeps getting worse and worse and worse. And so understanding that, having some flag to say, hey, here’s how I should start to adapt or change what I’m doing so that it doesn’t get to the point where it’s just something that is past the point of being able to make a huge difference and you have to start talking about different kind of health intervention. So that gets back to this whole foundation of why we should care about paying attention to markers in real time, multiple markers in addition to glucose.

Ben Grynol (00:31:02):

One of them that is interesting and this comes up very frequently is the dawn effect, right. And so you brought up talking about the way that our autonomic system will adapt in sleep states versus states when we’re awake. And so why don’t we go into this idea of how does cortisol play into all of this, not just when we are waking up and what actually happens physiologically in our body, but also cortisol, I think, actually, let’s riff on this for a sec. Cortisol’s an interesting one because if we pair it apart, so we parse it into two things, we associate cortisol with being bad, heightened cortisol, and it’s not actually because it is associated, so cortisol equals stress, we’ll just say it colloquially, cortisol equals stress.

Ben Grynol (00:31:57):

And it’s like, well, there’s also eustress, which is a positive form of stress where you want that with, let’s say, athletic performance. You want a high level of eustress because that’s going to put you into this state where it’s like you are ready to go and you’re firing on all cylinders, right? Whereas you can have high levels of cortisol that are, we’ll call it bad stress, again, keeping it pretty colloquial, but bad stress where it’s like no, you don’t want to have a heightened state of cortisol. So it’s understanding where is cortisol helping us and where is it maybe prohibiting us from having a balanced state of health.

Azure Grant (00:32:35):

I think cortisol is one of the, cortisol probably along with LH, are some of the best examples to give people to think about context, to think about biological rhythms at the hour to hour time scale versus the daily time scale, which is relevant when you mentioned the dawn effect or circadian rises in cortisol. But let’s just step back and talk a little bit about what cortisol is, the hypothalamic-pituitary-adrenal axis and then how to interpret it based on different contexts. So cortisol is something that’s regulated, like we said, both on a daily time scale, so once every 24 hours or circadian rhythm, it actually tends to rise pretty notably across the morning. This is also called the cortisol awakening response. That’s part of how if we… We don’t have to use an alarm, this is one reason I don’t like to use alarm clocks. It allows you to wake up naturally and that’s why I actually think of it, and I think a lot of people tend to think of it this way, as an awakeness or an arousal hormone rather than a stress hormone.

Azure Grant (00:33:43):

So it’s something that helps us feel awake, alert and ready to mobilize energy stores and do what we need to do. So it has this prominent rise in the morning, but then it doesn’t actually stay smooth all day long. It doesn’t look like a sign wave or a smooth wave on the ocean. It actually has these pulses, so it’s called a pulsatile hormone, and they happen every hour, couple of hours. They’ll actually happen in the absence of us doing much else, so it’s regulated this way all the way up to the level of the brain. And then these pulses are both generated on their own and they’re able to react to what’s happening in our environment. So cort is one of these ones where if you saw that your cort was pretty high for a little while, that might be absolutely nothing to worry about.

Azure Grant (00:34:31):

And so it brings up this difference between thinking about an acute elevation, so a spike that comes up and goes back down, and a baseline elevation wherein the resting level of cort or, let’s say, the big wave on top of which these little waves ride, once that starts coming up, that’s a bad thing. So imagine if you are going along in your everyday life, you’re doing things that require you to mobilize energy stores, like you’re getting ready to go and do some exercise, go for a walk, you’re getting ready to, like this, come in and do a podcast recording that we have to be alert and paying attention for and we’re interspersing that with times of rest. So we’re going to see cortisol all go up and down for that.

Azure Grant (00:35:16):

But let’s say we have an event that is very stressful emotionally or physically that we’re having trouble coming down from, and so that could be something like social stress, maybe a fight with a family member, a really bad meeting or kind of exercising yourself to the point of exhaustion. These are things that can keep cortisol elevated for a long time because you’re still needing to mobilize your mental and physical resources and that kind of condition, if maintained for a long enough time, that’s really bad, and that leads to overall elevations in cortisol or eventually to even adrenal fatigue. And that kind of chronic elevation is the kind of thing that if you were to go and take a blood test, you would already have had to have gotten to that chronic elevation state in order to absolutely consistently pick up that elevation on a blood test.

Azure Grant (00:36:12):

And I say all this not just for the sake of cortisol, but because I think these principles apply to a lot of the different molecules that we’re interested in measuring and if you’re someone who wears a CGM, then you might have been listening along to talking about cortisol thinking, hey, this actually seems quite similar to how I think about my spikes when I look at my glucose data. So there’s a lot in there, both in terms of how this acutely reacts to the actions we take in our everyday life. Short term elevations are not so bad, but longer term chronic elevations can lead to baseline elevations over long periods of time and that those are the things that are bad and that we want to catch before we might be able to catch them on a blood test.

Ben Grynol (00:36:51):

And cortisol is very interesting because people are starting to see a link without measuring it through things like their CGM data. So they’ll say, like you associate the event where people have a spike, let’s make an assumption. A person didn’t eat anything, they haven’t eaten for, let’s say, a four or six hour window, some amount of time that’s long enough where you should have, and they didn’t eat something that would give them a spike, they’re not oscillating, it’s relatively average stable glucose. And then they see a spike and they’re like, what happened, I was sitting watching something? And what it was is they were watching the New England Patriots game and that gave them a cortisol spike which ended up in a glucose spike because their sympathetic system was kicked in with the fight or flight response where it’s like, oh, you had a heightened level of cortisol because you were so invested in this thing.

Ben Grynol (00:37:42):

Same thing happens with exercise where it’s like, oh the reason you are getting [inaudible 00:37:47] much… When I say spike-related to watching a game, everyone’s going to have a different response. Some people might be neutral, some people might have a massive spike, but you’re not going to see these ridiculous spikes that would be the same as exercise where somebody is pushing very, very hard and they say, holy smokes, I had a huge spike. Well, that’s because you push it where cortisol kicked in, a bunch of your stored glucose in your liver was dumped back into your bloodstream, so it’s like hey, here’s your energy source, there we go, right. Lots on there. But we also see things like glucose spikes associated with cortisol that don’t have to do with fight or flight, stumbling on my words here, but have to do with those stressful events. So you highlighted somewhere it could be a disagreement or maybe a bad meeting or something that you’re about to record a podcast. Other times it’s like, I had this thing and it’s because your kids are fighting over an iPad or something. That raises our cortisol.

Ben Grynol (00:38:47):

So that is acute. But what we don’t want to have is this heightened state of cortisol over time because that’s going to impact so many other markers and that leads to… This is where it’s such a full circle of chronic inflammation related to other markers being out because this one… it’s all so interconnected. And so the key is to make sure that we understand how to have relatively stable markers across the board so that nothing is, if one is elevated or I shouldn’t say elevated because they can all mean different things at different levels. Out of range is a better, that is the scientific way of saying it. If they are out of range, that is something where we want to pay attention because that, over time, they’re all so interconnected. So it’s like the oil is out in your car, well then you’ve got engine failure and then your gas is low in your car and you can’t drive anymore. And it’s the same thing with the body.

Ben Grynol (00:39:42):

Why don’t we go into women’s health? It is a very, very wide topic to go into. It’s something that you have deep expertise in and have spent all of your time studying. And so let’s go into estrogen, progesterone, and luteinizing hormone and talk through… Why don’t we talk through it in respect to fertility, we’ll talk through it in respect to some of the health considerations, and then also later stage when women are going through menopause. So how some of these hormones will change and tied into diet, and how when these hormones oscillate every month, depending on stage of life, how that changes physiologically the way that you’re metabolizing different food and the way that you can think about even things like exercise. So lots of things to go into. Let’s start with estrogen. Why don’t we actually, let’s do this. Let’s go through each of the hormones, what exactly they are and how they tie in to women’s health. So start with estrogen.

Azure Grant (00:40:48):

Okay. Estrogen, progesterone and then LH, I think, are some of the most valuable hormones that one could measure continuously and we should definitely talk about testosterone as well for guys, even though men and women both have testosterone, estrogen and so forth, just in relatively different concentrations. And the reason that I think these hormones are so interesting and important to think about measuring at high frequencies is because they change in a very structured, patterned way, both in response to everyday life activities and in response to endogenous biological rhythms. And those changes have very strong impacts on metabolic and overall health. So maybe that’s the first take home point with something like estrogen is estrogen level is not just about when you want to have kids and family planning. It’s actually, especially for women, a marker of overall health. And let’s talk a little bit about the patterning of how estrogen changes over time.

Azure Grant (00:41:58):

So one theme I think we’ve had today is that for many, many markers, including markers that we would traditionally just measure once every quarter or so on a blood panel, time of day is really important. Time of day or circadian rhythms are something that show up in most every system across the body, including estrogen. And these rhythms actually couple across different systems, so many of them change in the same direction at the same time or at least in a phase-locked manner. So estrogen tends to rise across the morning. Light cortisol we talked about and actually a bunch of different hormones, believe it or not, estrogen’s upstream hormone, luteinizing hormone from the pituitary, is actually extremely pulsatile as well and one of the most studied pulsatile hormones. So if we think about that wave of estrogen coming up in the mornings, it has these little pulses every one to four hours that ride on top.

Azure Grant (00:42:55):

And if we zoom out even a level bigger, so so far, we have very little waves within the day riding on top of daily rhythm waves, we have the waves of estrogen that are probably most familiar to all of us from maybe high school health class, which is the ovulatory menstrual cycle. And within that cycle, basically, estrogen rises across the first many days. Once it gets high enough, it triggers several reactions in the body that encourage an egg to be released and then estrogen falls. And then in the latter part of the cycle, if an egg is successfully released during a phase called the luteal phase, estrogen rises again a little bit and falls back down. So overall estrogen is a hormone that has these layered rhythms on top of each other. What’s interesting about this is that estrogen actually helps lower blood glucose, lower body temperature, lower heart rate, raise heart rate variability, and those start to be things that we can tie in to how we feel and what we perceive every day.

Azure Grant (00:44:01):

So one practical way this shows up that we’ve talked a lot about at Levels is that insulin sensitivity is a little bit higher during the first part of the ovulatory cycle when estrogen tends to be much higher and insulin sensitivity tends to be a little bit lower, so carb toleration isn’t quite as good during this latter part of the cycle where estrogen remains at relatively lower levels. And so I think this is actually a really interesting topic for research to go into later is to what extent does changing the way a person eats by phase of cycle impact how they feel, impact the glucose volatility that we see, and is that a viable solution that should be more commonly recommended? I think things like ketogenic diet or nutrient-rich but relatively processed carb reduced diets during luteal phase, lower estrogen phase in particular, they’re super interesting and I think we should talk about them more. So that’s estrogen in a nutshell. You want to talk about progesterone in a nutshell?

Ben Grynol (00:45:05):

Yeah, let’s go into that and we should tie it into pregnancy as well. Maybe we’ll do that after because there’s so much that happens during pregnancy from a metabolic health standpoint because there’s so much going on hormonally where the way food affects your health is completely different than when you aren’t pregnant, right. So we’ll shelf it for a sec, but let’s go into progesterone, so exactly how it contributes to fertility during women’s cycles, how it prepares uterus for fertilization of the egg and all these things. So let’s go into that and then we’ll get into LH.

Azure Grant (00:45:44):

Absolutely. So progesterone is something that is maintained at relatively low levels until around the time when an egg is released. So during the first part of the cycle we talked about estrogen being the dominant hormone, it rising, it actually having these pulses, those pulses get faster and faster leading up to the ovulation event, helping trigger it, and then what? So around the time when an egg is getting ready to be released, the ovaries and even the brain actually start to make a little more of this hormone, progesterone. Pro-gestation makes it kind of easy to remember. It has to do with getting the body ready to potentially host a baby for a long time. But progesterone also raises local inflammation and that’s local both around the ovaries and overall. So if you’re someone who has tracked basal body temperature or used a wearable to track your temperature over time, you’ll notice that it starts to relatively increase during the peri-ovulatory phase and especially after ovulation is complete.

Azure Grant (00:46:51):

So after an egg is released, there’s a little piece of tissue that the egg came out of the ovary in called the luteal body, and that starts to make more and more progesterone across the latter part of the cycle called the luteal phase, and that ramp up looks very proportional to a ramp up that you see in body temperature as well as a ramp down in things like heart rate variability. And in addition to that, that higher level of progesterone is associated with a little bit worse insulin sensitivity. So I think this is definitely something to look at if someone is wearing a CGM is to see if your spikes are getting a little bit worse during that luteal phase, and if they improve with the onset of menses, which is a time when progesterone is quite rapidly falling off in preparation for the next cycle.

Azure Grant (00:47:47):

And I think to tie that all together, we should talk a little bit about LH because LH, maybe that would have been easier to do first even, but LH is sitting upstream of both estrogen and progesterone. And one thing that’s interesting about it is the way that it is released, in particular, the patterns of its release can tell you a lot about what estrogen and progesterone are doing. So in our reproductive system we call it the hypothalamic-pituitary-gonadal or HPG axis, LH that’s at the pituitary level and it’s released, and something we’ve talked about a lot so far is the very pulsatile fashion, meaning every one to four hours-ish, it gets a big pulse and then it goes back down to where it was. Those pulses are actually regulated within the brain via interactions between the hypothalamus and potentially even the dopaminergic system, which is kind of a whole separate interesting side story.

Azure Grant (00:48:49):

But basically, what LH does is it travels through the blood stream and communicates to the ovaries, hey, it’s time to release estrogen or progesterone depending on the phase of cycle. And one of the ways that this gets communicated is in how fast those little pulses are. So LH pulses faster and faster leading up to ovulation, then gets a big wave called the LH surge, which is the context that most people, if they’ve heard of LH, have maybe used one of these little pre-ovulation tests where you can see if this hormone is very, very high enough to trigger ovulation or not. And then after that ovulatory event, this hormone LH slows down a lot, once to every three or four hours of pulsing and then does that for much of the rest of the cycle.

Azure Grant (00:49:37):

So the point is here, if one could wave a magic wand and measure LH, you could actually learn a lot about when ovulation was coming, when it was about to happen via that surge and when, via those slowed down pulses, when it had already happened and you’re progressing through the rest of the cycle and you can only know that really if you were measuring it continuously. So LH is something that I think about for if I wanted a proxy for what’s going on with estrogen, if I wanted a proxy for ovulation and if I wanted a proxy for when ovulation had successfully happened, and then downstream of that, wanted to make some assumptions about what was happening with insulin sensitivity for women, I would think about looking at those patterns. And you’re probably starting to get a sign of this by now, but I think that those high frequency patterns carry a ton of information that are going to end up being really important for how we interpret most continuous biomarkers.

Ben Grynol (00:50:38):

Yeah, it’s fascinating. There is so much to dig into with it. And the idea with fertility, it’s so interesting because the window is very small, right, so it’s like when we talk about the period in which women can actually get pregnant, it’s that narrow window where, if you’re able to track LH where you can see, you’re like, oh, I can see what is happening and then that becomes a window. And so without this insight, it becomes a lot harder to say, it is possible to monitor, not in real time, but to monitor certain elevations in hormones right now. But it gets a lot. You’re having to put in the work, it’s so much harder than what actually happens. So typically the women that I know and the women that are in my life, whether it’s friends, whether it’s family members, in general, if they want to get pregnant and they’re family planning, it’s timed based on cycle.

Ben Grynol (00:51:33):

And so you’re trying your best with things, like there’s the tracking apps, but it’s still not going to be perfect because there’s too many things that are fluctuating as far as these hormonal levels. And so you’re doing your best to try to time things, but it’s still not perfect by any means, and if you miss that window, you miss the window, right, and it gets really hard. And so having the ability to, from a fertility perspective alone, having the ability to track things like LH, things like progesterone, things like estrogen in real time, you’re going to be able to monitor what you want to monitor and get the insight that you want a lot easier.

Azure Grant (00:52:09):

I mean, I think we should just come out and say that cycle tracking for the most part is still stuck with 100 year old technology. It’s stuck with, at the most basic level, looking at day of cycle, which is absolutely not an accurate way to understand when you’re ovulating or understand the portion of the cycle leading up to and including the time of ovulation during which a person can become pregnant. And then when you layer on top of that something like oral temperature, which is what natural cycles is approved as a birth control method, if you do this very laborious thing every morning at exactly the right time, you’re going to get kind of a shaky but decent signal if you use any version of a single time point temperature measurement. We haven’t really talked about this, but basically estrogen lowers temperature, progesterone plus estrogen raises temperature.

Azure Grant (00:53:01):

Basically those measurements are single time points and they are never going to give you as much as a continuous signal can. So really for fertility, the value is in the continuity of the signal because for making predictions, it’s all about looking not just at the level of a hormone or the level of a proxy measure. It’s about looking at the pattern of change over time and the frequency of these biological rhythms within the day. Those are what allow you to make strong predictions. And those combined with the levels of the hormones are what are going to allow you to also confirm that something like ovulation has happened. So I think that’s one of the most exciting things about continuous reproductive hormone measurement is to be able to truly do family planning for the first time in a way that doesn’t place all of the onus on the individual self tracker trying to do a perfect job measuring a signal that you cannot capture everything important about using a single time point measurement.

Azure Grant (00:54:02):

And this even, I think we should flip this on its head and think about glucose because as we’ve talked about, estrogen lowers glucose and progesterone raises it. I think one thing that we should absolutely explore is the more continuous interaction between blood glucose as a measurement and something like high frequency estrogen measurement. Because imagine if every time a person got a little pulse of estrogen, they got a little local decrease in blood glucose. How effectively do you imagine maybe a person could look just at their glucose signal and, given the proper view, be able to tell something about the phase of ovulatory cycle that they were in? I think that would be super cool. And we haven’t even talked about pregnancy or lactation yet and the interesting patterns that might show up there. But yeah, I think fertility is one of these key fields where basically people are still relying on the equivalent of blood work or single time point measurements when you can absolutely rely more on a continuous measurement from the periphery or hopefully in the future, from blood work on the hormones themselves.

Ben Grynol (00:55:11):

I think you just signed yourself up for another episode. We might have to do one on pregnancy in general because fascinating what happens.

Azure Grant (00:55:19):

Sounds great.

Ben Grynol (00:55:20):

We’ll shelf that one. Why don’t we go into high and low levels of LH, because that is an interesting thing to give people insight. So when people are going through ovulation, you’re going to see increased levels of LH. Now let’s say somebody is outside of a window of fertility, but they’ve got these high levels of LH, that’s a bit of a flag where you go, wait, I’m outside of my cycle when it should be high. So the indication is, depending on stage of life, you could be, if you got high levels of LH, like abnormally high, you could be in a state, you could be perimenopausal or going through menopause. That’s one flag.

Ben Grynol (00:56:00):

The other is you could have PCOS, you could have a pituitary disorder. There are these flags. And so it’s important to look into this to say, what’s happening, if you are seeing this and you’re tracking it over time, these are things where you can now start to action or take action based on the insight that you get. So we’ve got that. And then we also know that low levels of LH could mean also pituitary disorder, or there could be things like malnutrition or you’re under stress or there are all these interlinked things. So being able to monitor that, that’s a very interesting thing that can help a lot of women with their holistic health.

Azure Grant (00:56:39):

Absolutely. It’s great that you bring up PCOS and menopause. Obviously very different states that one can be in, but with respect to LH, they actually share some things. It’s not just that the levels of LH are elevated in PCOS and elevated around menopause, but one of the ways they get elevated is those pulses that we talked about. So those pulses get even faster and faster in PCOS in menopause. It’s almost like the brain or the hypothalamus where this pulse generator, called the arcuate pulse generator, sits is saying hey, please, we need our hormones to do the right thing. We need to try to get ourselves to ovulate successfully, which is something that’s not often happening in PCOS cycles, and in menopause, the person is transitioning to a time in which they’re not able to ovulate anymore and they’re going longer and longer between menstrual periods or having more and more anovulatory cycles. And so part of that, the brain’s and the pituitary’s response to this is to say, hey, we really want to make more of this signal that should be telling our body, please ovulate.

Azure Grant (00:57:50):

If we go a little bit deeper into PCOS, I think this one is really important because of the tie-ins that it has with metabolic health for women and because of its prevalence. So PCOS is something that might affect somewhere approaching around a quarter of women and for various reasons that have to do with a number of different kinds of tests and providers involved in getting a diagnosis, it’s something that can take a long time to diagnose, and actually the longer that time to diagnosis, the worse likelihood or let’s say the increased chances of going on to develop something like type 2 diabetes. So in PCOS, not only do we have these LH responses that are high levels, high pulses trying to help the body ovulate, but it also seems like there are high levels of androgens and also a degree of metabolic dysfunction, in particular insulin resistance.

Azure Grant (00:58:46):

So what one could imagine down the line is that we could use additional hormone testing and maybe in this case it could be something like LH to look at signs of ovulatory dysfunction combined with something like continuous testosterone testing to look for elevated androgens and maybe improve the time to diagnosis for something like this. And similarly for menopause, one of the big challenges that I’ve read about and have heard expressed by my family members and friends is that perimenopause, so the time at which overall sex hormone levels across ovulatory cycles are getting a little bit lower and lower, cycles are tending to get a bit shorter and this person is heading towards a time where they will stop ovulating, that can be a time when symptoms start to emerge, when insulin sensitivity gets worse, when people start having things like vasomotor symptoms, hot flashes and night sweats, and there’s not a whole lot of guidance for this process experientially. There’s also not a whole lot of guidance on when exactly a person could start something like bioidentical hormone replacement therapy or when they would take it.

Azure Grant (01:00:01):

It’s well known that the kind of sooner that you start taking exogenous sex steroids, if you take them in the right way and they’re formulated the right way, the better the outcome. But another use for continuous sex hormone measurement in the future could be saying, hey, we have watched you as an individual over years progress across your ovulatory cycles as you’ve gotten older, we’re seeing them destabilize, we’re seeing you ovulate less and less frequently, and we think you’re headed towards menopause. So now, before you, maybe even years before you would’ve intervened before, we’re going to get you talking to a clinician about how you want to manage the state, whether or not you want to take bioidenticals, how you might want to change your diet, maybe to a lower carb but still very nutritious lifestyle and all of that. So it’s a lot about prevention and early diagnosis.

Ben Grynol (01:00:54):

Yeah, there’s a lot to dig into with it and, especially when women are going through menopause and things, even progesterone is changing too, that’s helping to regulate blood pressure. So if you’ve got low levels, that’s where having the insight to say, hey, here is what is going to allow you to make the right choices or to adapt your lifestyle choices, adapt your diet, adapt your exercise, your sleep, all of these things, hugely, hugely beneficial in overall long term health, right, because it’s everything and we keep, it sounds like a broken record, but keep riffing on the idea of everything is so interconnected, so poor sleep quality leads to other metabolic dysfunction and other health complications long term.

Ben Grynol (01:01:40):

Well, why are you getting poor sleep? Maybe you’re getting poor sleep because you’ve got low progesterone and it’s because you’re at this phase of life or you’re going through menopause and it’s just like everything keeps going on and maybe you’re getting poor sleep because you ate the nachos when you went to bed and so, and you drank the alcohol and you did all of these things where over time, and we’ll shelf this one altogether, but it gets into mental health too, where certain people start to have, their mental health might shift or change from what their steady state or the way that they felt before was because it’s like, well, you’re getting lower levels of sleep and you’re feeling bad because of all these things physiologically are happening in your body and it’s like that is the interconnected web of all these things. So it’s having that insight to say, hey, this is actually leading to that and here are things that we can change.

Ben Grynol (01:02:30):

So physiologically, we’re not just, menopause is happening, that’s happening. That is physiologically happening. But what you can change is you can change your diet, you can change the way that you sleep, the way that you exercise. You can start to supplement with different hormonal therapies over time, Azure, men and women, because we know, getting into testosterone, we know that depending on the reason for taking something like testosterone, I think we associate, colloquially, we think, oh, somebody is supplementing with testosterone, that’s not good. What, are you trying to get really big and strong? The idea is like, no, actually if you’ve got low levels of testosterone, you can have low mood. It’s really going to affect. Too high levels is when you start to get into things like having more what’s known as the bull aggression, right. So it’s having balance, but you need to make sure that you are able to adapt and supplement as needed based on what is happening in your own body. So it’s not a, hey, this is happening to everyone. It’s, this is physiologically what works for you and here’s how we’re going to personalize it.

Azure Grant (01:03:43):

Yeah. When you talked about balance, I think that’s kind of the key theme in all of this and we all know it or we have some idea of balance that we grew up with or moderation and everything, including moderation. But I think understanding what balance means from a physiological standpoint is not something that we’re traditionally taught. I mean, even for instance, we often talk about is flat blood glucose the best when actually our body regulates everything on a wave like scale. And so it’s rather than, I don’t think thinking about balance as thinking about a flat line or an average, but balance is about being at the right level at the right time with the right pattern. So it’s really more of a shape or a pattern than a level. And testosterone is one that I think everybody knows about because it’s funny and it’s in the media and it’s in sports, but if you have too high testosterone, sure, you’re going to get muscular, but you’re also going to shut down your HPG axis and that’s one of the common things that people talk about when they talk about taking roids.

Azure Grant (01:04:51):

But andropause is a very different story. And andropause is about acknowledging that in the 40s, just like with women having these overall, little bit at a time, declining levels of sex steroids, men go through the same thing and it’s not pleasant. I mean, decreased energy, decreased focus, decreased sports performance, you start to get into eventually prostate problems, sex and libido is all woven into it, and so the idea of testosterone replacement, I think, if done carefully up to the right level and eventually in the right pattern would be tremendously helpful to health. You just can’t go overboard and, A, start giving people cancer or, B, start shutting down the HPG axis by giving it way too strong of a signal. So I think one, have you talked about the whole body is a symphony thing on Whole New Level yet? I feel like that’s an analogy that kind of helps everybody.

Ben Grynol (01:05:57):

You are the first. It is your analogy, so take it away.

Azure Grant (01:05:59):

Okay. I think it relates a lot to what we talked about with how many sensors does a car or a rocket have. But it also takes into account this idea that all of these systems in our body are interconnected and tend to change in a coupled manner over time. And so I think what we are at Levels, especially on the research team and on the data science team, we’re doing something called network physiology, which is studying how different systems in the body change in a coordinated fashion over time and how that coordination falls apart in different ways during different kinds of disease states. So basically the idea is we might want to think of glucose as a silver bullet or maybe one signal that we could look at that would tell us everything we need to know, but it’s more like, I would think of it as the first chair trumpet in a symphony.

Azure Grant (01:06:54):

So if your body is a symphony and your trumpet is your blood glucose, then you know it’s important, it’s flashy, it plays good solos and we need it. But for instance, it’s not going to tell us if the bassoons are out of whack. If your bassoon section in your body is out of whack, you might be pretty unhealthy, but if you only have your head trumpeter to listen to, you might not really know that there was something wrong with the bassoon section until things got so bad that she couldn’t play her first chair solo correctly. So the reason that we need to measure multiple things in the body, even though they should be theoretically playing together all the time, like the chords within a symphony, is because those are still players with individual agency. So they should hopefully be playing all together at the same time in rhythm and in tune to keep us healthy, but still an organ system or a single output can start to drift and it might take a while before it drifts enough that it starts to pull other sections or other players out of tune.

Azure Grant (01:07:56):

So basically I think when we think of the body like a symphony, we can understand both that we need to be listening in continuously to the signals that we’re able to listen to, but that in an ideal world, we would be able to be a very good conductor and listen to every single player. And when we think about blood work, for instance, what we’re really doing there is basically listening to that head trumpet player play her C note for a couple seconds and then stop playing. So we’ll know if she’s able to hit that one note on time, but it doesn’t really tell us how well she could play her solo. So we need it, but there’s a lot more that we can do once we have some continuous information coming in.

Ben Grynol (01:08:38):

You are speaking my language. I love this analogy immensely because it all makes sense, right. So we’re the conductor, right? We’re the conductor of what’s happening. We are in control of this symphony. You could have all the instruments playing in time. There’s sort of two things going on. All the instruments are playing in time, but the wood section is out of tune. They’re still in time, but they’re out of tune. It’s like, ah, it’s fine, they’re playing in time. You could also have things be in tune, but off time where it’s like, well, it’s not working.

Ben Grynol (01:09:12):

And so you’ve got these things that aren’t working the way that they should, and you want to have the symbiosis between everyone to make sure it sounds like a symphony and is playing really well and tight and with the right cadence and crescendos and all these things. But it’s like if you’re off, if you’re out of tune and you’re off time, it’s like that’s the long game of, okay, well this is what happens over years is eventually your whole orchestra falls apart and next thing you know your veal has got a broken string and you can’t recover. And that is not a good thing. So I love the analogy, but it’s so true. It’s like our body really is a symphony and everything has to work in this state of symbiosis to make sure that we’re singing these beautiful songs, painting the picture.

Azure Grant (01:09:56):

Yeah, and I mean, we absolutely want to catch ourselves before we get down the line to when our blood panel comes in and it’s way way off the rails. But I think this also ties back to what you were talking about where how many molecules, hormones, autonomic signals from the skin would we need to measure in order to have a pretty good idea of what was going on overall? So you might think about it as how do we take one player from each one of the sections and give ourselves an overall picture of what this thing is probably sounding like? And I think that’s really the hardest problem.

Azure Grant (01:10:28):

So we might want to think about in a dream world, what can we measure to take a look into what’s happening in each one of these key systems of the body, the metabolism and liver function, the reproductive system, the thyroid axis, inflammation and the immune system. It goes on and on, emotional state and motivation, things like cortisol or dopamine, and we can imagine if we got enough of these, we think we would know enough to be able to either predict the information in the rest of the systems to an acceptable degree, or at least have a representative from each section that’s telling us what’s probably going on in the rest of that system.

Azure Grant (01:11:12):

But it’s a hard problem. And I think one of the reasons to be hopeful is because in general, our body, especially our biological rhythms and our homeostatic or allostatic systems are trying really hard to keep us in time. And that means that, for instance, if you have healthy blood glucose regulation, your insulin should be bumping along every few hours over the day and night, over the ovulatory cycle, even for everyone over the seasons at about the same time. And so I think those coupling constants, which we expect in the state of health to be fairly strong and we expect to maybe get weaker in some very specific ways and disease states, those are the things that we have to work really hard to find out now, and I think the way that we approach that is starting to try to measure as much as we can. So hence us being the people with wearables and with our CGMs and maybe spitting in tubes every half an hour for fun.

Ben Grynol (01:12:05):

So funny. I love the visual. I’m just picturing, it’s almost like a union. You’ve got one representative of the percussion section, one of the brass, somebody’s representing woods, but all these people are coming together, that’s a blood panel, right. It’s like this little union of a trumpet with some legs and some arms going forward to this line, and then we look at the line and it’s like they’re representing the group of all these other things to say okay, this is going to give us enough insight to say, is this section of the orchestra doing okay? Is it sort of out of whack? And if the viola comes forward and all the strings are broke and you’re like, we should probably look into that. So love the analogy.

Azure Grant (01:12:44):


Ben Grynol (01:12:44):

It’s really, really good and super fun to riff on all this stuff. I think what we’ve got to do is we should definitely do an episode on pregnancy. That’s fascinating. But I think there’s so much more we can do as far as digging into a part two around how exactly are we going to use this data? It’s anonymous data as far as the way that we’ve got this data set from our members. But we’ve got so much insight. We’ve got hundreds of millions of health data points that we can look into. We can start to see some trends and some patterns. And we’re in such an early phase of this right now, we’re collecting this data and being able to dig out insight. But yeah, there’s a lot we can do as we’ve looked into it. So maybe we should do a part two around what are you seeing, what are you seeing with our members and how can we think about things and where do we want to head if we start to make some of these adaptations to the insights that we surface so people can really make changes to their health.

Ben Grynol (01:13:42):

Because that’s our goal is like, it’s not just enough to collect the data and say like, hey, this thing happened. It’s like this thing happened and here’s what you can do about it from a personalized standpoint. So if you’re up for it, it might be a fun thing to do.

Azure Grant (01:13:54):

I would love to. And maybe just a teaser before we end up here. That’s really our top problem I think is to say we have all of the leads from a bunch of different systems in the symphony playing their C note and that’s the blood panel. And then we have one player in the symphony, maybe that lead trumpet player, and she’s playing her whole song. So how do we take all the things about her whole song to try to infer what might be going on in the rest of the symphony where they play continuously? And the way that we’re doing that is we’re basically looking at the notes, the melodies, the speed and pattern at which she’s playing, and literally a lot of the same signal processing techniques that you would use to look at sound are what we’re using to look at continuous glucose data. And we’re trying to find the features in that that are associated with better blood work overall and also better, slightly more complex metrics that have been published that have to do with better glucose levels.

Azure Grant (01:14:52):

So I think that’s one of the best things about having collected this data set and about being here is we’re going to get to share what we’re learning through that a lot sooner. And I think the teaser right now is that it doesn’t just seem to be having in range glucose levels, keeping your carbs low enough, locking after you eat enough that you’re not getting big spikes. But there also seems to be something that has to do with the stability of when you eat building on this field of time restricted eating or circadian patterned eating. But it seems like if your blood glucose has more stable rhythms, that also might give you a boost in terms of how well your blood workers is doing. So I think we’re going to spend a bunch of time digging into these questions and I would love to do a separate episode on that.