Rhonda: And so the major, obviously, it seems like lifestyle factor that is regulating glucose disposal, insulin sensitivity, I mean, it seems like both of these things are affected by the contractions of muscle and increasing those glucose transporters, right? So that-

Peter: Exercise is probably the single most important thing we have at our disposal to increase insulin sensitivity. And then there are other things that are very important. So energy balance really matters. Sleep really matters. So both acute and chronic disruptions of sleep will impair that system. It's not entirely clear why, by the way. The experimental evidence is undeniable. And these are experiments that are so easy to do well that they're unambiguous, right? Where you disrupt people's sleep. If you just took a normal group of people and you did what's called a euglycemic insulin clamp, which is an experiment where you run IV glucose and IV insulin into people, and you basically run a fixed amount of insulin into somebody and then determine how much glucose you need to put in to keep their glucose level fixed. That's called a euglycemic, keep glucose fixed. That's the gold standard for measuring insulin sensitivity. So you do that test on somebody, and then for a week, sleep deprive them for down to five or six, four hours a night. Call it four, four would be very dramatic. Within days, you'll see like a 50% reduction in their ability to dispose of glucose. With no other difference, no dietary difference, no exercise difference. So we don't know exactly why that's happening, but it's a very repeatable observation. So sleep disruptions impair this. Energy imbalance impairs this. Hormonal changes impair this, right? So as we age, both the reduction in estrogen and testosterone impair this. Hypercortisolemia impairs this. And then of course, inactivity is the greatest thing that drives this.

Rhonda: I definitely didn't do the exact experiment you're describing, but I've mentioned it to you before. I had my CGM, and when I was a new mother, it was clearly my sleep was being disrupted. I was getting up and breastfeeding, and I mean, it was like night and day difference in my fasting blood glucose, my glucose disposal, my postprandial levels. I mean, it was like clear.

Peter: Oh yeah, we would have asked you to take that CGM off. That would be an awful time to wear a CGM.

Rhonda: But I did find that my going to my HIIT class, even though I was just dogged tired, the last thing I wanted to do, really did normalize it. So is there a postprandial level that like, let's say someone's not trying to do a low carb diet, like they're not trying to like, because that's a whole other area, right? But like they just, they're eating maybe a more omnivore diet and more paleo-ish or Mediterranean-ish, right? Is there a level that you think postprandial, you know, glucose level, like a threshold that would signal like, oh, well, you shouldn't really be going, or is it?

Peter: It's hard to say. I mean, here's what I think we know more clearly. We certainly know with more conviction that the average blood glucose, the lower it is, the better you are. And I say that even outside of diabetic range. Now, I don't have level one data to tell you that because the study's never been done, but I can tell you that by proxy based on hemoglobin A1C data. So the hemoglobin A1C data make it very clear that lower is better even outside of the range of diabetes. So diabetes is defined as a hemoglobin A1C above 6.5%. That translates, 6.5% is an estimate of an average blood glucose of 140 milligrams per deciliter. So assume for a moment that if you have a CGM that says 6.5%, meaning you just trigger the threshold for type two diabetes, your CGM would say your average blood glucose is 140 milligrams per deciliter. Nobody disputes that that's harmful. The question is, is it better to be at 130, 120, 110, 100? Like at what point is it too low? And what the hemoglobin A1C data would suggest is being at 5%, which is about an average of 100, is better than being at 5.5%, which is an average in the one teens. Both of those are normal by our current definitions. Neither of those would be pre-diabetic even. So five and 5.5 are both considered completely normal levels. But the all cause mortality data or the data on all cause mortality suggest a better outcome if you're at five rather than 5.5. That suggests to me, by proxy at least, that an average blood glucose of 100 on a CGM would be better than that of an average blood glucose of 115. So that's the single most important metric we care about. We use other metrics to think about that. So that since we can't measure insulin in real time, looking at postprandial spikes and variability, so looking at the standard deviation, which you can get off the CGM, and just the number of times you exceed a threshold, and that threshold you could say, maybe make it 150 or 140 milligrams per deciliter. And you can just say, how many times in a week do you exceed that threshold? That might give you some indirect proxy of how much insulin are you secreting in response to that. Because, for example, if you took two people who had an average blood glucose of 110 milligrams per deciliter by CGM, but one arrived at it with levels like that, and one arrived at it with levels like that, the former would be a better way to achieve that than the latter. But there are lots of things that raise glucose that are not harmful. For example, that HIT class that you were doing, probably in the short term, really spikes your glucose, because your liver is really trying to meet the demands of all that exercise. So it's putting a ton of glucose into your circulation, and it's going to do the right thing, which is always err on the side of too much. Because in the short term, it's better to have too much than too little. So if I'm wearing a CGM doing a really hard workout, I mean, I'll see that glucose get to 160, which is higher than it will get with a meal.

Rhonda: That goes right back down.