The Peter Attia Drive - #26 - AMA #3： supplements, women's health, patient care, and more

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

00:00:10.160 The drive is a result of my hunger for optimizing performance, health, longevity, critical thinking,

00:00:15.600 along with a few other obsessions along the way. I've spent the last several years working with

00:00:19.840 some of the most successful top performing individuals in the world. And this podcast

00:00:23.620 is my attempt to synthesize what I've learned along the way to help you live a higher quality,

00:00:28.360 more fulfilling life. If you enjoy this podcast, you can find more information on today's episode

00:00:33.020 and other topics at peteratiyahmd.com.

00:00:41.360 Hi, everybody. Welcome to the Peter Atiyah Drive. This is an AMA episode, Ask Me Anything episode,

00:00:47.260 and it's our third such AMA. First time, however, that we use the site to aggregate questions. For

00:00:53.800 the prior ones, we'd done it through Twitter. This seemed a heck of a lot more efficient,

00:00:57.220 so I think we're going to continue to do this. So if you want to ask a question or vote up or down

00:01:02.740 on a question, just go to peteratiyahmd.com and you'll see a little AMA thing to click on the top

00:01:09.180 right part of that, and you can go ahead and do that. In this episode, which by the way, we also

00:01:13.220 did via video. So I think this is the first podcast where we also have a video that'll be released,

00:01:18.400 presumably going to be on YouTube or wherever, but we'll link to it here in the show notes. Maybe

00:01:21.700 there's somebody who prefers to watch this stuff as opposed to listen to it. In the future,

00:01:25.220 I suspect we'll do a couple things by video, but have no plans to do that en masse.

00:01:31.580 Let's see. What do we talk about? We talked about reference ranges for labs,

00:01:34.780 a lot of questions about where I draw my cutoffs versus where the laboratory might draw a cutoff.

00:01:40.220 Talked about women's health, specifically a couple of issues that I think are kind of

00:01:44.240 underappreciated, maybe understudied. Gave some advice to medical students and residents or people

00:01:48.900 who are thinking about going through medical training or perhaps in medical training. Talked about

00:01:53.680 what my plans are to monetize the podcast, which is something I'm going to need to do pretty soon.

00:01:58.860 Talked about the two wearables that I find most sticky, which is my sleep ring and my CGM,

00:02:05.360 my continuous glucose monitor. Talked about heart rate variability, how I use it, why I use it,

00:02:10.540 things like that. Discussed a number of supplements and drugs that people had some very specific

00:02:14.400 questions around, specifically statins, ezetimibe, baby aspirin, lithium, berberine,

00:02:19.480 metformin. And I think we talked also about nicotinamide riboside, which of course is

00:02:24.160 getting a lot of attention as an NAD precursor. We talked about the memoir that I'm working on

00:02:30.220 about being a shepherd. And we talked about that whole coconut oil as a poison thing and a bunch of

00:02:36.340 like recent epidemiologic studies that are sort of goofy. One thing I would like to remind people

00:02:40.960 about is I have a weekly email that comes out on Sundays. All you have to do is give us your email

00:02:45.580 and you'll be on the list. And I promise to make it not lame and worth the time to read

00:02:49.900 on the website. Remember there are show notes and I know I say this every time, but it's probably

00:02:54.620 worth reiterating. Sometimes these podcasts do get a little bit technical. And I think that

00:02:58.980 our team does a really good job trying to put notes in that make it a lot easier to go back

00:03:05.300 and either follow up on a, you know, a term that might be obscure or a study that we reference and

00:03:09.940 go over loosely. And I also want to point out a lot of the times when I'm talking just sort of off the

00:03:15.240 cuff, I'm getting things wrong. And this is where my team is. They're just, they're incredible,

00:03:19.840 right? They're going to go back. They're going to listen to everything I say, and they're going to

00:03:22.820 put the correction in the show notes. So just because I say something, you know, I think that

00:03:27.620 hopefully I don't say things that are wrong that often, but it's going to happen. But I'll cite a

00:03:31.360 study and I'll say, oh, there were 40 patients in it, blah, blah, blah, blah. No, actually, Peter,

00:03:34.440 there were 32 in it. And here's the link to the actual study. So that's the kind of stuff that's worth

00:03:38.020 doing. And then lastly, if you are enjoying these podcasts, please go to Apple and review them. I'm told

00:03:43.720 that that's actually a valuable thing to do. And if you don't like the podcast, yeah, don't go and

00:03:48.020 review it. Actually, that would be much appreciated. Okay. So with nothing else to say, welcome to AMA

00:03:52.740 number three. Hey, Bob. Peter. Welcome back to New York. Thank you. Thanks for having me. It's a huge

00:04:00.320 sacrifice to have you. Hey, everybody. Welcome back to our third slash second AMA. I guess technically our

00:04:09.340 second when you consider that the nothing burger one was very specific to the fast. Again, we're

00:04:14.020 still thinking that this might be a quarterly thing. We've got an AMA page up on the blog. So

00:04:18.160 now folks are asking questions and voting up and down and that's sort of making it much easier for

00:04:23.060 Bob Kaplan here. Who's my, I don't know, my right hand, my main man, my sidekick, my shepherd, whatever

00:04:30.360 to aggregate before that he was doing it through Twitter, which was super painful.

00:04:33.900 So let's see today. We have a lot of questions. I haven't looked at them yet, but I've like seen

00:04:41.620 on your screen, how many there are. It's stressing me out a little bit. Yeah. I'm suspecting it's

00:04:46.400 impossible to get through all of them. So maybe we'll just kind of try to draw a hard stop in a

00:04:51.020 couple of hours and see how far we got. Also, for those of you watching this, this is the first time

00:04:55.960 we're going to record one of these in video. So I have no idea if that's going to prove to be of any

00:05:01.500 value or not. Maybe it won't, but nevertheless, for those who like to be able to watch a couple

00:05:06.040 of knuckleheads drink their Topo Chicos, as opposed to just listen to the bubbles, welcome to my little

00:05:12.560 apartment. All right. All right. I'm going to let you take it away, Bob. Let's get started.

00:05:16.540 And so one of the features on the AMA page is a vote up, vote down feature. So we can see it,

00:05:22.460 which ones are more popular than others. So we'll bias some of those questions towards this.

00:05:27.780 And so one of the most popular questions was curious, what other reference ranges Peter

00:05:35.000 considers too soft? Okay. So this is sort of on the heels of the last podcast or the last AMA,

00:05:42.060 where I talked about my blood values before, during, and after the fast. So because I don't

00:05:47.360 remember what I actually talked about, I'll just probably just start from scratch and there might

00:05:51.380 be some repetition here. So as I just sort of think about going through the labs from start to finish,

00:05:56.060 when it comes to the lipids, I generally take a harder line on triglycerides than is typically

00:06:02.620 given. So the typical reference range on triglyceride is less than 150 milligrams per

00:06:07.440 deciliter. Apologize. I don't know what that is in millimolar for people who are outside of the United

00:06:11.600 States. I generally like to see triglycerides below 100. And if we're really going to get fancy,

00:06:18.280 I'd like to see triglycerides lower than HDL cholesterol when both are measured in milligrams

00:06:23.460 per deciliter. So again, that's probably a bit more of a stringent level that I would put on

00:06:28.820 things, but that that's sort of the first place I think about it. Secondly, looking at the LDL

00:06:32.720 particle number, which is measured in animal per liter, just morally and philosophically,

00:06:37.660 I feel like that number ought to be below a thousand animal per liter, which is about the 20th percentile

00:06:43.500 of the population. And the reason I feel that way is heart disease and atherosclerotic disease are the

00:06:48.080 most ubiquitous causes of death. So to be average on the disease that is the most common strikes me as

00:06:56.240 just backwards mathematics. So you have to, you really do need some alpha when it comes to a metric

00:07:01.020 like that. And I know that there's going to be some people watching this that are saying,

00:07:04.420 oh my God, LDL, there's all this controversy. I don't know when this podcast is going to come out

00:07:09.020 in relation to the discussion that I had with Dave Feldman about this and the discussion with Tom

00:07:14.220 Dayspring. So either by the time you're listening to this, hopefully that point will have been long

00:07:19.080 addressed. And if not, I assure you it will be addressed. So it might make sense to define what

00:07:24.980 reference ranges are because I think people might get confused with what is average and what is

00:07:31.560 optimal. And so people might look at, I don't know, that two thirds of the country is overweight and

00:07:38.900 or obese. And so when you look at the average, it's going to be different from what is actually

00:07:42.840 either normal or healthy. So thanks for bringing that up. So reference ranges are usually given

00:07:50.500 around a population distribution, but it evolves over time. So if you went and looked at a laboratory

00:07:59.140 test from 25 years ago and you looked at in the same units for the same variable, you know, for example,

00:08:06.480 ALT or AST, the transaminases, you would see different numbers. And all that's telling you is

00:08:13.380 every few years, the labs have to kind of update their reference range on what they're seeing in

00:08:17.420 the population. So for some things like hormones, you're often going to see, you know, between X and

00:08:24.360 Y and any lab will tell you upon asking, they usually don't print this on the lab, but we go back and

00:08:30.940 usually ask them, what do your reference ranges represent? And it's usually either 10th to 20th

00:08:38.360 percentile to 80th to 90th percentile, or two standard deviations below the mean to two standard

00:08:44.000 deviations above the mean. And again, that's usually something they could only say if the data

00:08:47.660 approximate a normal distribution, which many of these things do just like height, for example.

00:08:52.500 So you could report height as the average height or the, you know, interquartile range. So the

00:08:58.900 25th to 75th percentile of height, I'm making this up. If for a male in the United States would be,

00:09:04.240 I don't know, five foot eight to five foot 11 and a half. But of course, if you're asking a separate

00:09:10.520 question, which is if you're trying to optimize to be a rower, a heavyweight rower, you know,

00:09:15.280 someone who rows crew, you'd obviously want to be probably between six foot two and six foot four.

00:09:20.600 Again, I'm making that up. But the point is showing you that you would pick a range that's

00:09:24.940 completely outside of the statistical norm, because you're optimizing for something very

00:09:28.520 particular in that case rowing. So I basically just disregard all the reference ranges when I

00:09:33.440 look at labs and I kind of have my own set of standards that are based on my belief system

00:09:39.160 around what should and shouldn't be the case physiologically. So again, going down that list

00:09:43.560 after the LDL, I'd like to see the small LDLP below 500 nanomole per liter, which is represented

00:09:49.660 by the 50th, 25th percentile, pardon me. I like to see the oxidized LDL below 40. That's a very

00:09:56.380 stringent criteria because most labs will acknowledge that anything below 60 is reasonable.

00:10:00.980 I like to see that even lower. I like to see C-reactive protein below one, even though most labs

00:10:06.680 consider anything below two reasonable. Just kind of mentally scrolling through this, I like to see

00:10:11.520 uric acid below 5.0. Again, most labs consider anything below six to be normal. Some labs even

00:10:17.980 consider anything below seven to be normal. And again, I think in part that's because they're

00:10:21.940 optimizing around the prevention of gout and it's very unlikely you're going to have a gout attack

00:10:26.580 with a uric acid of 6.2. However, I think there are a number of other health consequences to that.

00:10:31.700 Did you give ALT and AST?

00:10:32.980 No. So ALT and AST, as of now on the lab that I use, the upper limit of ALT is 44, maybe 42,

00:10:45.180 and AST is about 40. I think both of those are unnecessarily high. And we typically like to see

00:10:52.580 patients below 20 on both of those.

00:10:54.940 Oral glucose tolerance test. You mentioned this, I think, maybe in another AMA. You talked about the

00:11:00.180 one hour, two hour, both insulin and glucose. Actually, you gave those.

00:11:03.740 Right. So again, I don't know. I can't remember what the reference ranges are on our lab, but I know

00:11:08.900 that they're geared primarily towards a diagnosis of type 2 diabetes. There was a day when you would use

00:11:14.300 the OGTT to diagnose type 2 diabetes and not what we do today, which is rely on hemoglobin A1c.

00:11:20.160 I think the use of an OGTT is a better test. And I think the hemoglobin A1c is not a particularly

00:11:24.920 helpful test. So yeah, I generally like to see the fasting glucose below 90 milligrams per

00:11:30.620 deciliter, the fasting insulin below six, and at one hour following a 75 gram glucola challenge. So this

00:11:38.180 is a liquid glucose challenge. Again, if you change the way that you do the test, you have to

00:11:43.980 come up with a whole new set of reference ranges. At one hour for a reasonably muscular male, I'd like

00:11:49.660 to see the glucose below 120 or maybe at a max 130 milligrams per deciliter in the insulin below 20

00:11:55.860 or 30. In a less muscular individual or in a woman who is just smaller, I will tolerate slightly higher 0.53

00:12:01.660 levels. And at two hours, I like to see the glucose back down to below 100 milligrams per deciliter

00:12:07.600 and the insulin ideally below 20 or potentially even only 2x what the one hour glucose was. So if

00:12:14.400 the glucose at one hour was eight, to see it at 16 or less at two hours would be great. You know,

00:12:19.340 on the hormone front, boy, it's complicated there because you're dealing with so many other issues,

00:12:25.140 which includes symptoms that go far and beyond just the numbers. But if someone's TSH is between

00:12:31.720 about 0.5 and 2.0 and their free T3 is above 3.0 and their reverse T3 is below 12, I find it very

00:12:44.000 hard to see how they could have hypothyroidism regardless of the symptoms they have. And if

00:12:49.280 they do have symptoms that sound and smell and feel like hypothyroidism, I'm generally searching much

00:12:55.120 harder for another cause, which again is not to say you can't be fooled. I think you always can be,

00:12:59.780 but that would be what I would consider sort of biochemically optimal for thyroid. And then of

00:13:04.700 course, I think it's probably too nuanced to get into a discussion of what happens when one or more

00:13:09.260 of those is out of whack. How do you begin to make the diagnosis? And I've probably talked about this in

00:13:13.980 the past, but you know, the TSH is really only telling you about what the pituitary gland is seeing

00:13:19.780 in terms of T4 to T4, T4 to T3 conversion. So that's telling you about the diodinase in the brain

00:13:26.720 that's making that distinction. The free T3 is telling you how much peripheral T4 is being

00:13:31.280 converted to T3. And then the reverse T3 is telling you how much peripheral T4 is being converted to

00:13:35.600 reverse T3, which opposes T3. And so the reverse T3 opposes T3 and each of those has a completely

00:13:42.080 different sort of cause. So again, probably a little more complicated and maybe we'll save that for

00:13:46.560 another AMA. That's a great conversation with a whiteboard. Yes, that's a whiteboard discussion

00:13:52.120 because I'm going to screw it up if I try to explain it without being able to show D1, D2,

00:13:56.900 D3 is the diodinases and all the hormones. Probably another whiteboard discussion then would be the

00:14:01.860 male and female sex hormones. That's an even more complicated pathway. There are about nine things

00:14:07.420 that we measure there beginning with DHEA, FSH, LH, all the way to dihydrotestosterone or DHT.

00:14:14.700 Again, these are hormones that are sometimes we're looking to keep things below a certain level or above a

00:14:20.160 certain level. Sometimes we're looking to keep things between a certain level. I think for EPA

00:14:24.360 and DHA, when we look at the red blood cell index of EPA, DHA, I know this is a very messy topic and

00:14:29.820 it's also something we're going to absolutely talk about on a future podcast. So I don't want to get

00:14:34.040 too far down that rabbit hole. I definitely want to have Bill Harris on the podcast to discuss that.

00:14:39.440 But I do like to see the EPA, DHA index above about 8.5%. I used to sort of think anything above 8%

00:14:45.620 was ideal. Frankly, now I let patients up to 10% and even sometimes 12% if they're not having any

00:14:52.240 issues like nosebleeds or something like that. But again, it's more complicated because that's just

00:14:57.320 showing you the total EPA, DHA. And in some patients, you really want to see that DHEA

00:15:01.720 preferentially higher versus the EPA. And it does depend quite a bit on which axis we're most worried

00:15:08.520 about, whether it be the neurodegenerative axis, in which case the DHEA might be a little more

00:15:13.140 important, or the atherosclerotic axis, in which the EPA might be more important. And of course,

00:15:17.560 it all comes without saying, the source of that EPA and DHA is heavily dependent on sort of what you

00:15:23.480 care about and what you're worried about. A lot of the other ranges are kind of complicated. I've

00:15:28.280 talked about it before, so I'll bring it up again. Desmosterol is a very important sterol that we

00:15:33.780 measure. So we measure markers of cholesterol synthesis, of which there are mainly two that can

00:15:39.980 be measured commercially and measures of cholesterol absorption in addition to stanol production. So

00:15:45.800 sterols and stanols are slightly different. They have a slightly different chemical composition.

00:15:50.200 I don't particularly have an interest in how high or low the stanols are or the phytosterols,

00:15:56.480 but I use that very much so when making therapeutic interventions with lipids. But I do care about the

00:16:03.320 desmosterol. And I do, all things equal, if given the choice, like to see that above 0.5.

00:16:09.980 And the reason for that is potentially obscure, but I do have some concern about overly suppressing

00:16:15.960 cholesterol synthesis, which is typically an issue in patients who are taking statins. If you over

00:16:20.300 suppress cholesterol synthesis, I think in aggregate, there's a meaning at the population

00:16:24.700 level, this does not seem to pose a problem, but at the individual level, I would exercise some

00:16:29.440 caution. And so the desmosterol becomes a guidepost. Of course, that's all complicated because there are

00:16:33.880 some medications that interfere with the enzyme that converts desmosterol into cholesterol. And so the

00:16:39.520 whole thing becomes a bit challenging. And you can't really interpret the level in that context.

00:16:45.000 Let's see. Did I forget anything?

00:16:47.760 IGF?

00:16:48.280 Oh, yeah. That's a good one. IGF, boy, this is one where I've really changed my tune a lot over time.

00:16:53.460 I used to be in the camp that said, you know, low IGF is best. And in an ideal world, everybody should be

00:17:00.360 at or below the 50th percentile. And that's one where it's not even worth explaining what the numbers are

00:17:05.440 because it varies so much by age that you just have to look at the table that gives you the IGF

00:17:10.360 breakdown by age. But I actually no longer think that's the case. I really think that IGF should

00:17:14.980 be cycled between high and low. And, you know, for example, like when I did my fast after seven days,

00:17:20.100 my IGF was probably, you know, at the fifth to 10th percentile. And it might rebound to the 80th

00:17:26.600 percentile when I'm not fasting. And so I think epidemiology mostly sucks, especially like 0.93

00:17:33.200 epidemiology that is involving an intervention. Like people who do X get Y. I think that epidemiology 0.92

00:17:39.620 is absolutely the worst. The next layer of epidemiology that's like less shitty is looking

00:17:45.880 at IGF levels and contrasting it with disease because at least there you've simplified a

00:17:50.120 variable. You're not trying to figure out like, did those people eat more eggs or less eggs? Like

00:17:54.900 that becomes a separate question. And I think based on the epidemiology, there's a U-shaped

00:18:00.200 mortality curve with IGF, except it's very skewed. So it's not even a perfect U. And I'm having a

00:18:06.480 hard time being convinced by anyone, including proponents of very low IGF, that an IGF outside

00:18:13.260 of the range of about the 60th to 80th percentile is anything but optimal. So there I've really

00:18:18.360 switched my tune and become a little more liberal in what I like to see. Obviously, the two things that

00:18:24.100 say three things that impact IGF the most are amino acid intake. And there are some

00:18:29.500 amino acids like methionine that seem particularly potent. Insulin levels indirectly through the

00:18:35.660 binding proteins. So therefore, you know, which is largely determined by dietary carbohydrate.

00:18:39.720 And then of course, exogenous hormones like growth hormone, which is obviously very popular

00:18:43.520 in the longevity slash whatever-y circles. But growth hormone, which is an analog, it's the exact

00:18:52.700 analog of the human growth hormone is a hormone secreted by the pituitary gland that tells the liver to

00:18:58.280 make more insulin-like growth factor. So if a doctor is using GH, I hope they're monitoring IGF levels.

00:19:04.220 And so when you're looking at IGF on a blood test, you're looking at a snapshot, whereas you're

00:19:08.700 living your life as a more of a movie. And for example, I think if we, Walter Longo is probably

00:19:16.360 people think of him as a proponent of low IGF. One of the things that he points out in his

00:19:22.440 FMDs and the studies is that you get the shrinking of tissues and organs and the low IGF.

00:19:30.260 And then there's the regeneration, which sort of looks like rejuvenation. I suspect when those

00:19:37.160 tissues and organs are rejuvenating or regenerating, IGF levels aren't necessarily at the floor.

00:19:42.500 That's your rebound. That's where you're actually looking at synthesis and the other thing. So

00:19:49.220 you're looking at an average when you really should be looking at the cycling.

00:19:52.660 That's right. And I think Longo was influenced by his mentor, by the negative effects he saw on his

00:19:57.700 mentor and felt that maybe this having low IGF all the time isn't such a good thing either.

00:20:04.380 Again, I don't want to speak for somebody else's views. I don't know exactly where Longo falls out

00:20:09.420 on that, but you're right. Luckily, IGF is much more stable than GH. So people always ask,

00:20:15.920 should we be measuring GH in people? And my view is that's sort of like measuring ACTH,

00:20:20.980 which is another pituitary hormone that is the one that's most responsible for the secretion of 0.69

00:20:25.880 cortisol. And unfortunately, the answer is outside of very extreme pathological cases,

00:20:31.280 like acromegaly or pituitary Cushing's disease, very difficult to infer what the hell is going on from

00:20:38.240 looking at those hormones because they are so pulsatile. You know, you go in a sauna for 20 0.93

00:20:42.860 minutes, that's going to change your growth hormone level significantly, probably won't change your IGF

00:20:47.020 levels as significantly. So the advantage of looking at very high fidelity, high frequency moving hormones

00:20:52.600 is great. You're closer to the physiology origin. The drawback is the noise can be too much.

00:20:59.940 I'm sure there are a bunch of other things that I care a lot about. You know, I do. I mean,

00:21:03.660 but I just, I think I'm tired of this question. I was going to get into iron and ferritin and TIBC

00:21:09.720 and hematocrit hemoglobin and white blood cell count. I have points of view on everything. I'm

00:21:14.180 a little, probably too opinionated on this topic. Okay. We blew through our time limit on that question,

00:21:19.860 man. Let's see what next. This might be a quick one. What aspect of women's health is the least 1.00

00:21:28.440 studied slash understood. Was that a high scoring question? It was. You wouldn't be asking it this

00:21:35.800 early in the podcast if it wasn't. So ironic that I would be asked that question given that I'm

00:21:41.380 not particularly knowledgeable in this subject and I hope I don't represent that I am. I do have

00:21:46.660 a number of women patients. I would say a quarter to a third of my patients are female. But when you

00:21:51.860 start to talk about things that are uniquely female, a couple of things come to mind. So I guess what I'm

00:21:56.740 going to do is not answer this question because I don't know the answer, but just go off a little

00:22:00.620 bit and hopefully something I say is of value to the people who answer this question.

00:22:04.740 Obviously the most important distinction between men and women are the sex hormones. That's an 1.00

00:22:08.040 enormous difference. And the other difference that comes along with that is that men experience a much

00:22:13.440 more gradual decline in their sex hormones. And as such, it's often harder to appreciate

00:22:20.260 symptomatic changes in men. You got a guy who's walking around with his testosterone,

00:22:25.340 two standard deviations below the mean. He can actually feel pretty reasonable because

00:22:29.000 just like a frog who's been in water that's been slowly getting warmer, he doesn't actually

00:22:33.480 realize how hot the water is because he's been in there the whole time. Whereas if you dropped him

00:22:36.360 in that water overnight, it would be pretty stark. Conversely, women experience two completely, 0.75

00:22:42.540 again, I'm just completely empathetic to what women go through, which is during their menstrual cycle 1.00

00:22:47.640 or during their reproductive years, they have a menstrual cycle, which even over the course of 30 days 0.83

00:22:53.220 or so, 28 to 31 days on average, they're experiencing profound fluctuations in their hormones. And this

00:22:59.380 is another one of my favorite whiteboard topics. And, you know, I remember there was this one patient,

00:23:03.420 I'll never forget. She was, I remember sitting down with her in her living room, actually drawing this

00:23:08.660 out for her and explaining to her why she has PMS because her chief complaint actually, which I thought

00:23:15.120 was an odd reason to come to see me because I don't think like I offer any genius expertise on this.

00:23:20.160 But I mean, mostly it was just to help control her emotional swings during her cycle.

00:23:25.240 And when I actually drew for her the time course of here are the, you know, here's day one,

00:23:31.540 the day your period starts, here's day one of your next cycle. Here's how your FSH,

00:23:37.440 your follicle stimulating hormone, your LH, your luteinizing hormone, your estradiol,

00:23:41.400 your progesterone and your testosterone vary throughout that. And let's highlight what's happening

00:23:46.680 on about day 22 to 28, which is when you have this drop in progesterone and explaining to her that

00:23:52.340 for some women, that central meaning in the, in the brain, that reduction, that withdrawal of 1.00

00:23:58.340 progesterone can create emotional lability and other things. And it's like, you're not crazy.

00:24:03.540 Like that's like saying like someone who's depressed because they don't have enough serotonin

00:24:07.120 is crazy. No, I mean, it's just, everyone has different neurotransmitters and everyone

00:24:10.200 has a different response to hormones. So, I mean, it was a very profound thing for her. She's

00:24:14.400 like, Oh my God, like I've always just thought I was kind of crazy and I'm just like a moody bitch. 1.00

00:24:18.460 And I was like, no, I mean, and your mom went through the same thing. Your sister goes through 0.99

00:24:22.140 the same thing. I mean, this is hereditary and here are two things that you can do about trying

00:24:26.520 to control this. Again, I don't want to get off on the tangent and go what those two things are

00:24:30.180 that we talked about. So that's during a woman's reproductive years. The second thing that's

00:24:34.600 really profound is by the time a woman is in her fifth decade, typically all of a sudden that stuff 0.99

00:24:39.780 gets shut down. So all of a sudden she's losing estrogen, she's losing progesterone and she's

00:24:44.900 losing testosterone. And I include that last one because most people forget about it. So when you

00:24:50.440 talk about the distinguishing characteristics between men and women, we usually think about

00:24:55.020 testosterone being the male phenotype describing hormone, estrogen and progesterone being the female

00:25:01.080 dominant hormones. And that's true. But what most people don't realize is when you actually convert

00:25:05.640 the units to the same numbers. So when you go from nanograms per deciliter and picograms per

00:25:10.640 milliliter, which they're often reported in different units and you do an apples to apples

00:25:13.860 view, when you look at a woman's highest estrogen level in her life, which I mean, outside of pregnancy

00:25:21.340 is during her ovulatory cycle. So when she's ovulating, when she has that burst of estrogen,

00:25:27.960 if you take that estrogen level and compare it to her testosterone level on average, which is

00:25:33.580 testosterone varies a little bit by cycle, but not enormously, the testosterone is about 10 times

00:25:38.740 higher than the estrogen. Now the number never looks that way because testosterone is reported

00:25:43.500 typically in nanograms per deciliter, whereas the estrogen is reported usually in picograms per

00:25:48.180 milliliter. But if you do like the high school, you know, calculation of what your chemistry teacher

00:25:53.020 would have you do and put them both in the same units, you'll realize that testosterone is much

00:25:57.060 higher. So even though a woman's testosterone level is much less than a man's testosterone level, 0.84

00:26:01.820 a free testosterone in a woman might be one nanogram per deciliter. Whereas in a man, 0.53

00:26:08.100 the 50th percentile would be about 14 nanograms per deciliter. So call it 14, 15 times higher.

00:26:14.200 It is still a very dominant hormone, even relative to estrogen. So when all three of those hormones are

00:26:20.440 basically taken away in a period of two to three years, I don't want to get too far on the soapbox,

00:26:26.360 but like definitely top 10 pet peeves are maybe top five pet peeves are doctors that completely

00:26:33.680 disregard perimenopausal and postmenopausal symptoms in women and who without having ever

00:26:39.120 even read a single study or tried to understand the limits and the methodologies of the women's

00:26:44.660 health initiative come to the conclusion, well, no woman should ever take hormones and they should 1.00

00:26:48.780 just deal with their postmenopausal hot flashes and perimenopausal symptoms and eventually,

00:26:54.540 you know, let them lose their muscle mass and bone density because God forbid hormones cause 1.00

00:26:58.600 cancer or some other equally knuckleheaded conclusion. So those are, those are really two

00:27:03.740 huge things. I think a third thing that I've seen, and I gotta be honest, I'm kind of lazy.

00:27:09.540 I don't think I've seen this in the literature and I haven't probably made a strong enough effort to

00:27:12.860 look at it, but empirically it is so overwhelming that I would be surprised if it's not described in

00:27:17.920 literature is there's something about multiple pregnancies and the HPA axis in women.

00:27:25.060 So I usually don't see it after a woman has had two kids, but usually if a woman has had three

00:27:29.680 or four kids, the likelihood that her thyroid bounces back to normal seems not that high.

00:27:37.440 And so I have a couple of patients who came into the practice again, interested in longevity,

00:27:43.320 but the, the, their proximate issue of concern was, you know, ever since two years ago when I had my

00:27:49.340 third child or my fourth child, I have not been able to get back to the same level of energy that

00:27:54.700 I once had. And you look at them and they have a normal TSH typically, but usually their peripheral

00:28:01.440 metabolism of T4 is, is very altered. And, and these are, I think in some ways, like kind of the

00:28:07.500 easiest saves, like you, in a very short period of time, you can make a patient feel a hell of a lot

00:28:12.700 better. And they're just, they're, they sort of fall through the cracks. And again, I think it

00:28:16.580 might come down to just a failure to appreciate some of the, the, these subtle differences between

00:28:22.180 men and women. Again, men aren't giving birth. Fortunately, or else our species would have died 0.96

00:28:26.200 and we wouldn't have a species if you and I were responsible for procreation. Trying to think what

00:28:31.400 else are fundamental differences. I mean, there are many others.

00:28:34.340 Part of the question is least studied or understood. I think it was a Freakonomics

00:28:39.420 podcast. They talked about this a little bit where there's the thalidomide issue. And I believe there

00:28:45.980 were, I think they were maybe doing studies or something where women were involved and obviously 1.00

00:28:49.800 birth defects and horrible things happened. And they sort of took women virtually off the table in 0.91

00:28:55.200 terms of studying. And so there's a long period of time where it was, I think men were predominantly

00:28:59.720 being studied and the subjects of, yeah, we would just sort of assume that's whatever is we see in

00:29:05.840 the men we might see in the women. But I think there's been, as far as epidemiology goes, there's

00:29:10.020 been a resurgence in women's studies and things like that. But I do think that there's overall,

00:29:15.040 they've been understudied just in that regard as well. Just looking at the female body versus the 1.00

00:29:19.740 male body instead of just assuming, you know, they're, they're the same.

00:29:23.540 Yep. That's a great example. I remember that. I remember hearing that and thinking, God, that's,

00:29:27.400 I mean, everybody knows the story about thalidomide. Yeah. You can't get through med

00:29:30.240 school without it, but it's the ramifications of that. It's that, it's that what happened as a

00:29:34.200 result of that. That's a very interesting point. You know, another difference between men and women,

00:29:37.740 I don't know how well it's been studied, but most epidemiologic assessments would make a clear case

00:29:42.440 that women, all things equal, tend to get less cardiovascular disease. Now, I don't know how

00:29:46.760 eloquently these studies have been done, and I don't know if they've been normalized for ApoB in

00:29:50.640 addition to LDL-C. My guess is they may not have been, but nevertheless, if you were to make the

00:29:55.020 assumption that once that analysis is done, if all things equal, inclusive of ApoB inflammation,

00:30:00.880 all of the other things, insulin sensitivity, even if you want it to be really rigorous,

00:30:04.380 women still get less heart disease than men. And after you've normalized for blood pressure,

00:30:08.720 smoking, and really all of the major factors, you'd have to look at iron levels as a next

00:30:14.320 interesting concept. So there's pretty interesting data in men that the more metabolically deranged 0.99

00:30:21.880 they are, the better they do with therapeutic phlebotomy, whereas the less metabolically

00:30:26.620 deranged they are, the less of an effect they have by giving blood, reducing the oxidative stress

00:30:32.520 of iron. So it might be the case that somewhere buried within there is this idea that because

00:30:39.180 for a great number of years of a woman's life, she is going to have less iron than a man due to her 1.00

00:30:44.900 menstrual cycle, that may actually offer a protective benefit against heart disease. Again, something I would

00:30:50.100 be interested to know if that's, certainly people have speculated that, but again, I don't know how

00:30:53.940 rigorously it's been documented. And we have another thing out there that's lingering, and you're talking

00:30:59.240 to Richard Isaacson very soon, actually. He'll probably have more insight into this, which is if

00:31:04.380 you look at Alzheimer's disease and you look at the prevalence, I think it's two out of every three

00:31:08.880 cases are women. And so two out of every three cases of Alzheimer's are women, and you could probably

00:31:15.140 say that age is a factor. Longevity may be a factor in things like that, but I think that that needs to

00:31:19.980 be studied more too. Yeah, because I have a hard time believing that that is purely an age issue.

00:31:24.880 I don't think that the increase in longevity of a woman can affect that much of a gap in, and again,

00:31:32.380 it would be a great discussion with Richard. You're right, because you want to make sure that you're not

00:31:35.880 being fooled by, you know, a diagnostic distinction and things like that as well.

00:31:40.100 So piggybacking on the women's health a little bit, what are your thoughts on fasting and ketosis

00:31:45.060 for females? Well, you know, again, a real tough question because I find all questions of nutrition 0.99

00:31:52.220 to be so individual that it's hard to answer them sort of in a one-size-fits-all, sort of one-stop-shop

00:32:01.280 approach. That said, I think there are a couple things women need to be thoughtful about. If any woman 1.00

00:32:07.000 is having an issue with fertility, I couldn't make a case, I can't make a very compelling case

00:32:12.120 for nutritional ketosis if a woman is trying to get pregnant. And I'm sure this is just going to

00:32:16.600 piss off a lot of the keto herd, but because, of course, if you're in the keto cult, you believe

00:32:21.560 that ketosis is the optimal state for everything, including, you know, global warming. But the reality

00:32:27.940 is if you look at the FGF levels, and also if you just think about it from an ancestral standpoint,

00:32:32.800 the higher the level of ketone during our evolution, the more likely we were separated from

00:32:40.280 food. And the more likely you're separated from food, the less genetic pressure you should have

00:32:46.700 to be reproducing at that point in time. And this concept is so well-preserved in biology. I mean,

00:32:53.400 we had a great discussion with David Sinclair recently to talk about this, and it's even true

00:32:58.440 in the case of the sirtuins, which would be, you know, one of the more important regulators of

00:33:02.800 our aging and including, you know, our reproductive stress. So I think when you look at the FGF 21 data

00:33:09.900 and the ketosis data, there's a very, and I don't want to bastardize this because I wish I'd known that

00:33:14.440 I was going to be asked this question. I would have looked up the paper. It's from a researcher in

00:33:18.140 Texas. I think he's at, he might be at UT Southwestern, but he's, if you search, you know,

00:33:22.720 and his name's first name's David, I'm blanking on his last name, begins with an M though.

00:33:26.000 But if you Mangle, Mangle's door. Yes. Yes. Yep. That's him. I saw him presented a meeting

00:33:30.460 once and it was remarkable data that looked at the differences in a male and female brain

00:33:36.120 in the presence of changing FGF 21 levels. And it was so cool to see this difference.

00:33:42.700 The pituitary gland is one of the very few pieces of tissue in the body that has what's called portal

00:33:48.100 circulation, the liver being the other one. So the pituitary has a direct connection via the

00:33:53.660 pituitary stock to the hypothalamus. The long and short of it is in a calorie restricted state,

00:33:58.640 when ketones are elevated, it suppresses FSH and LH in women, but not in men. This is super

00:34:03.860 interesting to me. And it has an, it has a profound evolutionary. I mean, again, maybe I'm just making

00:34:09.600 up a story to fit this, but if we're going to subscribe to any of Occam's razor, this would be a

00:34:14.200 great application. In a period of famine, you would want women to stop reproducing. You would want to 1.00

00:34:21.020 shut off FSH and LH. You would want men to have no impairment on their testosterone level. That's 1.00

00:34:27.180 all the more time that they should be out there and able to, to get food. So again, I've always

00:34:32.120 been kind of a little bit careful of suggesting that a, that a woman who's trying to get pregnant

00:34:36.580 be in ketosis. Now, look, I know what's going to happen. Everyone's going to say, well, I got

00:34:40.440 pregnant when I was in ketosis. Yeah, obviously it's possible. I mean, I'm not suggesting it's not,

00:34:44.340 but if we're talking about optimization, the second thing of course, is should a woman be in 1.00

00:34:48.240 ketosis during pregnancy? And the, and the answer is I simply don't know the answer.

00:34:52.360 Clearly we evolved with mothers being in ketosis and having children. I mean, it'd be impossible 1.00

00:34:58.220 for our species to be here if mothers were not in ketosis during pregnancy. I mean, we didn't have

00:35:03.360 buffets, but that said, is it optimal? You know, again, just because something happened in, in sort

00:35:09.080 of our evolutionary time history, does that mean it's optimal? No, almost not at all. So as a general

00:35:16.320 rule, and again, I don't like to make general rules when it comes to nutrition, I'm not convinced it's

00:35:22.180 necessarily the best strategy. Just like, I don't think it's the best strategy for kids unless they 0.85

00:35:25.500 have seizures or something else, for example. So, you know, a far better strategy is just like,

00:35:30.780 don't eat junk food, you know, don't eat, don't eat sugar, don't eat highly refined carbohydrates.

00:35:34.980 But, you know, should one restrict carbohydrates to the point where they're in ketosis as a very

00:35:39.380 deliberate act? I'm not convinced that's the case. And of course, when you start to look at things

00:35:44.080 like maternal diabetes or gestational diabetes, rather, that's where it gets a little tougher

00:35:48.480 because ketosis can be a very effective tool for treating type 2 diabetes. And gestational

00:35:53.920 diabetes is not type 2 diabetes, but it has some of its features. So again, I really, I hope there are

00:36:00.400 some obstetricians out there who spend a lot of time on this problem because it is a huge problem.

00:36:06.020 There are many women who go through this. My sister, I probably have talked about this in the past,

00:36:10.360 and I don't think she'd mind me talking about it. And if she does, I'm in trouble. But, you know,

00:36:14.620 my sister had an operation when she was quite young that took out two thirds of her pancreas.

00:36:19.600 So during her first pregnancy, she got gestational diabetes, which was to be expected because she had

00:36:24.820 like a third of the insulin producing capacity. And then it happened again during the second

00:36:29.200 pregnancy. And then after that second pregnancy, she actually got type 2 diabetes. Over the course of a

00:36:34.400 year, going on a ketogenic diet, actually working with a company called Virta Health that I'm an

00:36:39.760 advisor to an investor in, although that is unrelated to the fact that my sister wasn't

00:36:43.880 getting care from them. You know, she'd lost 50 pounds. Her hemoglobin A1C went from somewhere

00:36:49.060 in the twelves to in the fives in the course of a year. And then she got pregnant again. Well, 0.78

00:36:54.880 she decided to go off the ketogenic diet, but still be much more diligent and strict. And despite that,

00:37:00.340 she still, you know, requires insulin during this third pregnancy. I suspect she'll have a much

00:37:05.060 easier time recovering following this pregnancy. But one of the things that's been frustrating,

00:37:09.900 which maybe goes back to this, the question prior to this is her obstetricians, like they have no 0.96

00:37:15.360 insight. Like they have not a thing to offer her as far as how she should be thinking about managing

00:37:20.940 her blood sugars other than just kind of cram more insulin into her exogenous insulin. So, 0.98

00:37:26.140 yeah. And this is why context matters. And there's no bumper sticker that you could have,

00:37:32.180 you know, the question thoughts on fasting and ketosis for females. You could have somebody with 0.58

00:37:36.580 the constellation of abnormalities that is metabolic syndrome, insulin resistant, type two

00:37:42.840 diabetic. And maybe for that person, their fertility might be an issue and going on a ketogenic diet may

00:37:48.960 get them in a healthier state. Yeah, exactly. It might actually improve their fertility because of 1.00

00:37:53.400 the inflammation and all the other stuff that could be happening as a result of what you just

00:37:56.680 mentioned. So, yeah, I, I, well said you should just answer the questions from now on. You do a

00:38:01.680 better job than me. I probably have way too many bumper stickers there that lack context that

00:38:06.740 need it. What advice would you give to medical students and residents? It's pretty broad. With

00:38:13.000 regard to what? Don't do it. Do it. Definitely change your scrubs every day. Yeah. How do I probably,

00:38:19.300 how do I, I don't know, I'm, I'm inferring stuff from this, but probably how do I get through it to,

00:38:26.040 all the way from how do I get through this thing to how do I optimize my time in medical school?

00:38:33.060 And maybe you can, you can sort of look back on your experience.

00:38:36.320 Maybe I'll start even broader. So I definitely re I really enjoy talking to kids in high school and

00:38:42.820 college who want to go into medicine. And I think the most important piece of information,

00:38:47.720 most important piece of advice I would offer these folks, which is probably somewhat biased by my own

00:38:52.760 experience. So I explained that as a caveat and I say, look, you know, here's my bias is don't study

00:38:58.400 pre-med in college. And for a lot of them, they're like, wait, why? But I know I want to do medicine.

00:39:02.620 Like, why wouldn't I study pre-med? And I'm like, that's exactly why all that stuff you learn in

00:39:06.600 pre-med is like the JV version of what you're going to learn in med school. Like, yeah, you're going to

00:39:11.460 learn a little bit of anatomy and some biochemistry and, you know, physiology and a whole bunch of things.

00:39:17.140 And you'll learn them at like, you know, some sort of superficial level of depth.

00:39:20.680 And then you're going to go to medical school and learn it all in real depth.

00:39:24.940 Okay, great. You just wasted four years. What you really ought to do in college is study first,

00:39:31.480 something that you freaking love. In the end, you want to study something where you don't care about

00:39:35.700 the grades. You're going to do well because you are absolutely obsessed with mastering this subject.

00:39:42.280 Look, that could be going an inch wide and a mile deep on something like, you know, I want to really

00:39:48.660 study genetics or biochemistry or molecular biology or something. Alternatively, it could be, you know

00:39:54.640 what, I want to study philosophy and history, or I want to study engineering or whatever, you know,

00:40:00.680 but the point is what you, the absolute last thing you want to do is try to cobble together an

00:40:05.140 education that is geared towards acing the MCAT. Like I can think of no sadder thing to do than to do

00:40:13.040 that. So that's, that's sort of my advice on the people going to medical school. Once you're there,

00:40:19.140 you know, I mean, this sounds kind of touchy feely, but I remember there were some days in

00:40:24.120 medical school where I was super unhappy. A large part of it was, it was the first time in my life I

00:40:30.840 realized you just had to memorize certain shit. And I had prided myself up until that point in life of 0.99

00:40:37.060 never memorizing things. Like I, I could remember the dates of boxing matches and that was it.

00:40:41.520 Couldn't remember a birthday to save my life. Like didn't want to memorize anything. And then you get

00:40:46.780 to medical school and you just can't get through on first principles. You didn't, you just, there's

00:40:50.340 just certain things you have to generate a certain base of knowledge before you can even participate

00:40:54.680 in a thoughtful discussion. And so that took me like a year to just accept. And there were times

00:41:00.240 when I was super pissed off when I was sitting there studying, thinking like, I'm getting so dumb. 0.96

00:41:04.700 Like I'm not solving problems. There are no differential equations. I don't even get to use vector 0.72

00:41:10.920 calculus anymore. One of the things that was very helpful, and I don't remember where this advice

00:41:15.460 came from, but it was very wise advice, which is at any point in life where you're sitting somewhere

00:41:20.120 and you're in a situation and you think it sucks, go and find someone who would give their left nut to

00:41:25.900 be in your situation and go and help them out. So I remember having this thought and thinking,

00:41:31.480 you know what, I'm sitting here at, you know, med school at Stanford. I've got a whole bunch of

00:41:35.160 undergrads at Stanford who were among the smartest undergrads in the country. And, you know, half

00:41:40.240 these kids, one and a half, you know, some percentage of these kids want to get into medical

00:41:43.600 school and they might not. So I remember actually going and just putting up my name on a billboard

00:41:48.380 in the, I forget even the name of the, God, it's been so long. I don't remember the name of the quad,

00:41:52.340 like what the main quad at Stanford was called, but putting up a thing like, Hey, anyone who needs

00:41:56.080 help setting for MCAT, like I'm going to be in this room at this time and I'm just going to answer

00:42:00.360 questions. And a bunch of kids showed up and I didn't charge them anything. It was more,

00:42:04.880 it was actually, I was doing it for me, right? I was doing it to see what their level of

00:42:10.060 enthusiasm was. And could I catch some of that infection again? And it works, you know,

00:42:15.640 it's really a powerful tool. And I did it again in residency. So in residency, there were days when

00:42:20.380 I was like, this fucking sucks. I haven't slept in two days. I haven't eaten a proper meal in a week. 0.99

00:42:27.660 I haven't even changed my underwear in two days. Like I, I just hate my life. I haven't exercised

00:42:32.780 in three days. And then I thought, wait a minute, you are so privileged to be sitting here, right?

00:42:37.300 Like, you know, you are one of six residents that's chosen to be in this categorical program.

00:42:42.520 Look at all of these medical students at Hopkins that would give anything to be here.

00:42:47.340 Do the same thing, grab two of them, sit down and walk them through something, help, you know,

00:42:52.320 teach them something. And you'll see like they'd give anything to be where you are. And so

00:42:57.340 I guess that's, I mean, you could apply that to life, I suppose, but it's particularly helpful

00:43:01.940 in sort of this transition from college to medical school, to residency and beyond.

00:43:06.080 What other advice could I give? I mean, I think a lot of the advice I have is so glib and so cliche

00:43:11.200 that, I mean, I feel a bit embarrassed saying it, hearing it come out of my mouth makes me cringe,

00:43:15.560 but you do have to maintain a balance. In other words, you'd be surprised. Like, let's just say

00:43:20.700 you've, you think there's like a hundred hours of work you need to do to prepare for a certain exam.

00:43:25.380 You're probably better off putting 85 hours into it and spending 15 hours doing something else

00:43:31.380 that is, you know, sort of replenishing you in another way. And for some people that's

00:43:35.780 pleasure reading, like, you know, reading fiction for others like me, that was always going to be

00:43:39.940 exercise for others. It's just sort of blowing off steam and going to get drunk again, not advocating

00:43:45.840 necessarily, but the point is it is important to get out of that, you know, get out of those books

00:43:52.240 for a while. I did not do that in college. So when I went to college, I just decided I just wanted to

00:43:58.520 know everything that was knowable. And that meant studying every minute of every day. I mean, I did

00:44:04.040 exercise, but I didn't socialize at all in college. Very, very little actually. Probably drank twice in

00:44:10.300 all of college. And they were both incredibly epic nights, actually. That's another story. I feel like

00:44:17.460 Chevy Chase. Like, I feel like there's like a Fletch moment here. We're reflecting on that little

00:44:22.280 bubble. We'll go back to you playing basketball in the Lakers. Yeah. With the fro at six inches.

00:44:27.960 But in medical school, I definitely spent way more time having fun. And part of it was like being in

00:44:33.500 such a, it was my first time living in California and it was like, oh my God, like you can ride your

00:44:37.900 bike outside every day. You know, you can swim, you can do all these other things. So I think that is

00:44:43.280 important. I think the other thing in medical school, not to lose sight of is why you're doing

00:44:47.620 it. The more frequently you can interact with patients in medical school, the better. Not for

00:44:54.400 the reasons that I think the consortium would say, which is the problem-based learning. There may be

00:44:59.080 some truth to that. Meaning when you take what's called a problem-based learning approach, which was

00:45:02.900 something that I think Harvard borrowed, actually McMaster University was the first to do it. And then

00:45:08.060 Harvard and two or three other prominent schools started doing it, which is basically this idea of kids

00:45:12.720 weren't going to sit in class anymore and do the traditional stuff. They were going to go straight

00:45:16.300 into the wards from day one and then encounter a patient, encounter a problem, and then go back

00:45:21.700 and learn what they needed to learn to do that. You know, whether that approach is right or wrong,

00:45:25.140 I actually have no insight. But what I think you do get out of seeing patients, whether it's in a

00:45:29.920 traditional path like they had at Stanford, where you spent, you know, the first two years grinding

00:45:34.600 through class and not was, it's really exciting. Like it's, it's, it's, you start to, I mean,

00:45:39.860 again, now I think at my stage, it's very easy to take that for granted and be like, you know,

00:45:43.280 oh, my patients are calling me and I'm trying to enjoy my weekend. And I'm like, you know,

00:45:47.840 stuck answering 27 emails or calling the lab or doing that. I think, but boy, when you're starting

00:45:52.620 out, it is just amazing to realize like you have this bizarre privilege. Like people are going to

00:45:56.660 tell you shit that they would never tell anybody else. You know, you're going to see people at a very 0.99

00:46:01.120 vulnerable state. And that's, I guess it's good to figure out early if that's appealing or not.

00:46:05.920 Because if it's not, if you're not odd by that, you, you probably don't belong in medicine because

00:46:10.900 it's, there's enough things that'll beat that out of you in medicine that if that initially doesn't

00:46:15.800 awe you, there's an issue. This is totally unrelated, but Tim Ferris had a podcast once about how to say

00:46:22.400 no elegantly. And there, I forget who he credits this to, but there's someone who has this criteria,

00:46:28.340 which is if when someone asks you to do something, if the answer isn't hell yes, don't do it.

00:46:33.240 Cause it's all downhill from there, right? If you call me up and you're like, I've got this great

00:46:37.880 idea. We're going to do blah, blah, blah, blah, blah, blah, blah, blah. If I think, okay, I'll do

00:46:44.060 it. Bad, bad idea. I have to think that's the greatest idea I've ever heard. And so similarly,

00:46:48.680 like if as a college student or as a medical student, you aren't bowled over by the beauty

00:46:54.020 of this interaction with a patient, you, it's worth questioning why you're doing this because it,

00:46:58.860 you will slowly have that sort of the forces that'd be will, will do everything in their power to beat

00:47:04.980 that out of you. That makes sense. It sounds touchy feely, but it is. Yeah. It's touchy

00:47:09.680 feely. It's begin with the end in mind. You really have to know what your goal is because

00:47:15.160 more or less you turn the have tos into get tos. So you know, like, Oh, I have to get up early in the

00:47:21.960 morning. I have to be on call. I have to, you know, whatever it is, even like a profound thing is

00:47:27.240 like for a mother saying like, I have to change these dirty diapers or something like I get to 1.00

00:47:30.920 change these dirty diapers because I have a kid that I love and we were able to, you know,

00:47:34.340 make this baby together and things like that. And I think when you go through something like

00:47:38.280 medical school, how rigorous and how tough it is you need, regardless, you're going to have to

00:47:42.200 remind yourself at times because it's going to be tough. Yeah. Yeah. That's again, well said. I don't

00:47:46.680 know why I'm answering these questions. You always say it better. We'll see. You wrote on Twitter

00:47:53.320 that you're considering different avenues to monetize the podcast in order to make it sustainable.

00:47:58.840 What do you think is the best way to do it? So this is something that, so, so we're recording

00:48:03.520 this today on September 11th. I'm not sure when this podcast will actually come out probably

00:48:07.440 later this month or in October, but it'll have been about three months. We started this in July.

00:48:12.200 So first of all, the whole thing was started as an experiment. Have we officially put every podcast

00:48:16.820 out that was in that first trial balloon? Yeah. Okay, cool. So as of yesterday with Lustig's

00:48:22.740 coming out, that is the experiment. Yeah. We might've squeezed one more in. I think it was,

00:48:26.660 we're going to do 12 and he's episode 14 maybe. And there's also the sneak peek. A couple AMAs.

00:48:33.180 Yeah. Yeah. Yeah. Yeah. Deep dive. So I guess I would say now having completed the experiment,

00:48:39.440 I have enjoyed the hell out of this way more than I thought I would. And I know what you and Nick

00:48:43.960 sitting over there, you guys are both rolling your eyes at me because you were saying this forever. And

00:48:47.800 so was Tim and so was Kevin and Jocko and Anahad and Patrick. And like, everybody had been telling

00:48:53.980 me, you got to do this. You got to do this. And honestly, I, this was one of those things where

00:48:57.320 all I thought of was I have to, like, I have to do that. I have to do that. And now I look at it,

00:49:02.460 like I get to do this. Like I get to go and interview people who know more about a subject than

00:49:06.780 I do and pick their brain. And the only thing I have to do is set up a bunch of goofy recording

00:49:10.560 equipment in front of us. It's really amazing. And I don't actually have to do any of the heavy

00:49:14.640 lifting after like you guys do it. Travis does it, right? There's like basically three people,

00:49:20.280 four people who do a whole bunch of work after I do a recording. And I never think about it again.

00:49:25.560 I've never listened to one of the podcasts after the fact, so I'm hoping they're turning out okay.

00:49:29.240 But basically I just get to walk around having spending, you know, about three hours a week,

00:49:33.240 having a cool discussion. So that said, it still is a highly expensive proposition by my back of the

00:49:38.980 envelope calculation. It costs about $2,500 to make each podcast, not including my time,

00:49:43.220 which I wouldn't necessarily want to put a price tag on, um, cause I don't know how to value it.

00:49:47.680 But if I just include the priceless, my time is priceless. Maybe that takes up to $2,500, $50.

00:49:57.540 Um, so, so whatever. So you're going to spend 10 grand a month making a podcast. That's great.

00:50:01.740 But the reality of it is we do need to, we do need to remunerate this for two reasons. One,

00:50:06.160 it's not cool to flush 120 grand down the toilet every year. And two, there are a lot of things that I

00:50:10.680 want to be able to do that upon further inspection. If the podcast can not only pay for itself, but

00:50:16.600 pay for some of the research that we want to do, that would be great. We have a, we just hired our

00:50:20.100 sixth analyst this month. So we have a team of analysts that basically I would like to be as

00:50:25.960 large as we could afford to manage, which means you manage. You can tell I don't do a lot around

00:50:33.380 here. I sort of take care of patients and that's, but right now our research team is basically

00:50:39.320 subsidized by our clinical practice. So everything is under an umbrella called a Tia medical and the

00:50:43.980 patients are basically subsidizing our research. And that's fine because the patients benefit the

00:50:49.240 most from the research, but I would like to figure out a way for the podcast to cover the cost of our

00:50:54.560 research. And if I'm really going to be wishful, not just cover the cost of our analysts, but

00:51:00.540 generate an additional pool of capital to invest in the types of research we want to do at certain

00:51:08.460 universities. So there are a number of investigators around the country who are doing really interesting

00:51:13.700 work who I think could be doing way more interesting work if they had unrestricted funding to do certain

00:51:19.080 stuff. For example, people have heard me talk about my obsession with wanting to understand if there

00:51:24.380 are biomarkers or things that could delineate a metabolic signature for autophagy. This is the type

00:51:30.220 of project that by my estimation could be done for a relatively low sum of money, meaning this is not a

00:51:35.520 billion dollar project. This is a project that a few people could get together and solve. I would

00:51:40.120 like to contribute heavily to that. Have a pretty good sense of who the people are that would need

00:51:43.840 to be involved in that project. If we can get IRB approval to do some of the experiments that I think

00:51:49.280 would need to be done, I think those are elegant. So all of those things are things that I would like

00:51:53.100 to see partially and maybe in a dream state completely funded by the podcast. Here's what I have

00:51:59.600 figured out I probably don't want to do. I don't think I want to sell ads. A couple months ago,

00:52:04.220 I put out some questions on Twitter about, uh, so we can, we can fully talk about the fact that

00:52:10.360 we're drinking Topo Chico because we're just advertising for free. We're not getting paid to

00:52:14.380 do this. It's refreshing. I was actually very touched by the feedback that we got on Twitter

00:52:18.740 when we put that question out about, Hey, what do you guys think about me shucking ads? Because

00:52:23.160 I almost don't remember a single person saying no bad idea. Everyone was like, look, you got to make

00:52:28.740 money too. You've been putting out free content since 2011. You literally have never charged us for

00:52:33.800 anything. You don't have a single ad on your website, blah, blah, blah, by all means sell ads.

00:52:38.560 And so I think I thought, I thought, I thought at the time, okay, well, let's do it. And here's

00:52:42.900 the way we'll do it. We're only going to sell ads for things that we love and care about and blah,

00:52:46.580 blah, blah, blah, blah. But to be honest with you, and I don't want to close the door to this,

00:52:49.920 but I'm having a hard time seeing how I could make it work because there just aren't that many

00:52:53.460 things that I love. And frankly, I have no idea if Topo Chico would want to sponsor this podcast.

00:52:58.780 They might not care. So at this point, I guess there's another deeper point, which is at some

00:53:04.620 point you will probably have to take an advertising dollar from someone who is incoming versus outbound.

00:53:10.020 So today, if we were going to start advertising, I would just say, Nick, by the way, Nick is behind

00:53:15.500 the camera. So at some point, Nick, you should just walk through so people can see who Nick is. But

00:53:19.420 I would say, Nick, can you please go and reach out to these 12 companies? And we've already made a list

00:53:23.860 of like, these are the companies that I love the most and I use their products religiously.

00:53:27.320 And I pay to use their products, blah, blah, blah, blah, blah. Let's go out to them and have

00:53:31.660 them sponsor the podcast. And who knows, maybe like a subset of them would do it. Invariably,

00:53:36.480 you're going to have to, you're going to exhaust those avenues and you're going to need to take

00:53:40.100 inbound. My fear is, even though I do, I think I have, you know, very high integrity. At some point,

00:53:47.280 I'm concerned, like, what if I just start deciding subconsciously that a certain product is better

00:53:53.260 than another product because it's going to pay a lot more than another product would.

00:53:57.320 I don't know. I'd like to think that's not the case. I'd like to think that I'm impervious to that

00:54:01.640 kind of influence, but I don't know if I am. And if I could prevent having to go down that path,

00:54:07.000 I would like to do so. So I think I don't want to do ads. So I think that basically leaves two

00:54:12.440 options. The first option is to do what my really good friend, Sam Harris does. And Sam has been

00:54:18.060 beyond generous in his time that he has spent with me, walking me through what is involved in a

00:54:24.960 listener support model, which is what Sam does, an entirely listener support model.

00:54:29.780 Sam also, if anyone hasn't listened to his podcast, he also does a great job explaining

00:54:34.580 his rationale for it. And he does it far more eloquently than I could reproduce it here.

00:54:39.280 So if you're interested in why Sam feels strongly about that, it's worth doing.

00:54:43.680 Rhonda Patrick is also entirely through a listener support model, though it's a different model than

00:54:48.160 Sam's. So I think that is one really good model. Again, I don't think it's going to be as lucrative

00:54:52.840 as ads. I mean, if we're going to be honest, it's probably not as lucrative as ads. But I think in

00:54:57.260 the long run, it actually could be better. The second model is something that I would only be

00:55:02.380 a participant in. And it would be there are lots of people out there talking about kind of a model

00:55:07.380 like Netflix, where there's a number of people would be part of a podcast network, where you only

00:55:13.960 get to listen to their podcasts if you are a member of the Netflix equivalent. That is something that

00:55:21.020 has only been put on my radar over the last two or three weeks. So it's a little hard for me to sort

00:55:26.060 of quantify and understand the economics of that. I mean, I understand the economics of it, but really

00:55:31.180 pressure test them. But that could also be advantageous because, again, it has the luxury

00:55:35.340 of being it frees you from the constraints of ads. It has its drawbacks, frankly. I mean, you're going

00:55:41.020 to lose listeners if you do that. Even if you make more money, you still lose listeners. It will reach

00:55:46.460 fewer people because the probability that everybody who's listening to a podcast for free will go and

00:55:51.300 join a service for that, I think, is relatively low. So I know that's not a great answer because

00:55:56.600 I haven't given the answer, but I would like to believe that by 2019, we are absolutely monetizing

00:56:02.860 this podcast. And I suspect it will be one of the latter two approaches and not an advertising approach.

00:56:08.960 I will say this. If we end up going down those paths, I would still like to talk about products

00:56:13.620 that I like. Because one of the pieces of feedback we did get from that very informal poll we put out

00:56:19.260 was people want to know what things we like. And to be able to actually shuck the products we like

00:56:26.860 when we're not getting paid to shuck them tells you we're really just shucking them because we like

00:56:30.500 them. So I like this ring. I like this ambulance siren. I like this bottle of water. Whatever it is,

00:56:38.060 we can just talk about it openly. And I think that there's benefit to people in that because then

00:56:42.100 they'll know like, okay, wow, that's an ideal way to get an ad. Do you have anything to say to make

00:56:47.600 what I just said better like you did with every other question? What is it? Buffett and Munger,

00:56:54.060 they do the annual meetings. Yeah. A lot of times I forget what he said, basically. I thought that was

00:57:00.800 well said. What he said. Yeah. All right, good. I have nothing to add. I think that's what it is.

00:57:05.500 He'll say, you go to Charlie Munger after Buffett does like a dissertation on economics. I have nothing to

00:57:10.840 add. That's what I'm going to strive for. Next question. What exactly are you looking to achieve

00:57:18.920 and monitor with your blood glucose monitor, your CGM, your G6, which you, another product you

00:57:26.320 probably, you love. I love the G6 and I'll talk about it all day long without, you know, receiving

00:57:31.640 advertising dollars to talk about it. You know, it's funny the G6 along with the aura ring, which I've

00:57:36.420 talked a lot about are these, and I've worn every wearable that there is, but they're the only two

00:57:41.780 that seems sticky enough that I can't stop wearing them. Like if I, like a month ago, I went to charge

00:57:47.200 my aura ring and I forgot to put it back on my finger when I went to bed. So I slept without it.

00:57:51.180 And I woke up the next morning and realized I didn't have it. And I didn't have the data. Like

00:57:53.880 it made any goddamn difference in my life. I was so pissed. The horror. I was like, God, how could you 0.99

00:57:59.360 forget to put your ring on last night? There's probably a whole separate issue with that. But,

00:58:04.060 and the same thing with like the continuous glucose monitor, I just, it's hard for me to imagine. I

00:58:10.140 used to not know my glucose in real time. So there are a couple of things from it. The first is it's

00:58:15.360 a great way for me to control my behavior. And I know it's tempting to want to believe that I'm

00:58:22.480 somehow impervious to the forces of bad food, but the reality of it is I am not. There was a day,

00:58:29.040 I think when I, you know, had a remarkable resilience and willpower and I could do anything,

00:58:34.580 eat this, don't eat that exercise like this exercise. I could, I was a robot for so much of

00:58:41.000 my life until three years ago. And something just happened in 2015. And I just fell off the rails and

00:58:49.080 I've never got back on. I simply do not possess the intestinal fortitude to be a robot anymore.

00:58:55.360 And I could speculate on several of the reasons for that, which I don't want to get into,

00:59:00.360 but the long and short of it is here I am. I am in an environment where like, you know,

00:59:06.040 yesterday I was on a plane and they were handing out shit, cookies and bullshit. I really wanted a 1.00

00:59:13.780 cookie. I think the only reason I didn't eat that cookie that was bigger than my head

00:59:17.460 is because I knew I'd have to look at my CGM data after.

00:59:20.540 So there is no more powerful behavioral tool for me than my CGM. Because in the end I'm kind of a

00:59:29.580 competitive person internally, much more competitive internally than externally, by the way. And I just

00:59:35.060 can't stand to see spikes of glucose. It just drives me nuts. And so, which is not to say I don't go off

00:59:42.800 the rail sometimes. I absolutely do. We were in Fenway Park the other day and I had fries. Now,

00:59:47.760 luckily I'd fasted all day and worked out. So I didn't actually experience a spike of glucose from

00:59:52.260 the fries. So I got to have the fries without the badness, but I was sort of ready for it. I was

00:59:56.880 kind of bracing myself like, Oh, you might, you might get a little testy seeing this thing,

01:00:00.780 but that actually gets to the second point, which is it has allowed me to very eloquently calibrate

01:00:07.380 how to tether activity levels, nutrient deprivation, the consumption of treats and minimize the damage.

01:00:17.960 I don't know that I could drive a race car very well without seeing my RPM tack. Like if you plugged

01:00:24.700 my ears so that I couldn't actually hear the rev of the engine and you took away my RPM tack and said,

01:00:31.780 drive, could I still drive the car? Yes. Could I drive it half as well as I can drive it when I

01:00:38.080 know exactly where I'm shifting at every moment where I need to shift? No, there's simply no way.

01:00:44.020 Like we just, we're feedback machines. We need feedback. So I'm a huge CGM advocate and really

01:00:50.800 looking forward to what the next few years will bring when these things can become a lot more

01:00:55.160 affordable and a lot more accessible. And the question is, can that be done without them remaining as

01:01:01.700 medical devices? So the one I wear now is the Dexcom G6 is a medical device. It's an FDA approved

01:01:08.200 device and it gives you a number that is in this case, incredibly accurate. It's probably plus or

01:01:13.260 minus two or 3% specifically for the purpose of someone with diabetes being able to dose their

01:01:18.300 insulin. The FDA will very likely not allow such a device into a consumer market because the concern

01:01:26.260 would be that such a device could be used outside of a prescriptive relationship with a physician to

01:01:31.060 dose insulin. So therein lies a whole bunch of issues that would basically, the way it would

01:01:37.780 happen today is the FDA would basically have to neuter the device such that the information couldn't

01:01:43.880 be used for treatment purposes, which means they either take away the real-time nature of it,

01:01:49.720 which is what makes it so valuable, or give you a bunch of ranges and dilute the accuracy. Those are

01:01:55.500 basically the two levers with which you could neuter one of these devices deliberately, which sounds

01:02:00.580 crazy, right? Like it's like a backwards step. Okay. I know you have a Zippo lighter. We're going to

01:02:06.080 start using sticks instead. We're going to rub them together, you know? So, but all that said, you

01:02:12.040 know, hopefully, you know, in an ideal world, the medical device becomes cheap enough that if you want

01:02:18.120 that level of precision, fidelity, and real-time feedback, you'll just, you know, look, doctors write

01:02:23.440 prescriptions for way crazier things than CGMs, right? I mean, you've got docs out there writing

01:02:27.960 prescriptions for pain meds all day long and every hormone under the sun. I don't think it's a big

01:02:32.380 stretch to say, doc, I need a CGM. And I think you might've mentioned this too in terms of what you

01:02:37.520 are looking to achieve and monitor with your blood glucose. You might've said that it's a proxy for

01:02:42.500 your insulin. And maybe you could explain why there isn't a continuous insulin monitor alongside

01:02:48.720 your glucose monitor, because that would be a nice get if it actually exists.

01:02:53.200 Yeah. And I looked into this a lot in 2011 and 2012. I even met with the engineer. He's actually,

01:03:01.100 I don't know if he's still there, but I don't remember his name now, unfortunately, but he was,

01:03:04.700 he might've been an emeritus professor of engineering at UCSD, but he was actually the

01:03:08.100 first guy to figure out actually how to do these real-time glucose monitors, what are called a

01:03:13.700 point of care device. And I actually took him out to lunch one day to pick his brain on, well,

01:03:17.340 why don't we just do this for insulin? And he was like, would that be interesting? And it was just

01:03:21.800 funny to talk to him because he's an engineer. Like why would he know that insulin would be as

01:03:25.400 interesting as glucose or more interesting? And, and so we actually dug into this a lot.

01:03:29.660 And basically the short of it is if you can't measure the assay using an antibody or enzymatic

01:03:37.340 reaction that very quickly without any washing yields an answer, you can't do it at a point of care

01:03:43.640 device. And insulin is pretty hard to measure. So it was initially measured using something called a

01:03:48.820 radioimmune assay. I believe today they're usually done with something called ELISA's,

01:03:53.260 which are these enzyme link, you know, I won't rattle off what ELISA stands for, but it's,

01:03:57.540 it's like, it's a chemical reaction where you have to, you know, puts, put an enzyme on something,

01:04:02.000 rinse it off, put another one on, rinse it off, et cetera. So in other words, it can't be done in a

01:04:05.260 moment. So absent that, I don't really see any direct way to measure insulin in real time. Now

01:04:11.340 I've had discussions with some companies who are interested in using CGM data to impute changes in

01:04:18.880 insulin. And I think that could be done, but I think it's a lot harder than people realize. And

01:04:25.200 you would need a lot of data to do it. Meaning you'd have, you wouldn't just be able to do it

01:04:29.320 off the CGM. You'd have to do the CGM coupled with a lot of blood draws where you actually could,

01:04:34.500 you know, basically build a regression curve off insulin and glucose to predict for future insulin.

01:04:40.140 Therein may lie an answer down the line. So absent that a good proxy for having a low level of

01:04:48.540 insulin is going to be a low level of glucose and a low level of glucose variability. And the CGM

01:04:53.000 spits out those reports. So you go, you know, you go into. In low glucose variability, one might.

01:04:59.480 Yeah. I don't know if people can see this, but I'm going to infer that the A1C might be telling you

01:05:04.520 that. The A1C is not telling you anything about the variability, but I don't know. Let's talk about

01:05:09.740 A1C in a moment, but you can see that I can spit out at any point in time, a 90 day, 30 day, 14 day,

01:05:14.280 or seven day report. And that report gives me average glucose and glucose standard deviation.

01:05:19.840 That's the variability. So why is that relevant? Well, you could have an average glucose of 85,

01:05:27.660 95, whatever, with the standard deviation of 10, which is very low variability, or you could have the

01:05:32.600 same glucose level with a standard deviation of 30. And those are very different insulin profiles.

01:05:37.820 So you want to keep, you want, you want to keep that balance closer. I've largely

01:05:42.060 discounted hemoglobin A1C in an absolute sense as a meaningful number. I think it's,

01:05:48.120 it's directionally tolerable, but mostly shit. And I know that because now I've used CGM in so many 0.99

01:05:57.400 patients with calibration and compared it to A1C. And you realize that the A1C is really at the mercy

01:06:03.680 of its most important assumption, which is a red blood cell lives for 90 to 120 days.

01:06:09.960 Anything that takes it outside of that range leads to an over or underestimation of the A1C,

01:06:15.220 and therefore an over or underestimation of the average glucose.

01:06:18.420 And you can, with an A1C, theoretically, you can say, I have an A1C of 5.4, and you can impute

01:06:24.280 what your average glucose levels were, theoretically.

01:06:27.420 Yeah. So the way the A1C works is you measure the A1C and you impute the average glucose. The way

01:06:33.060 the CGM works is you measure the average glucose, which is actually all that matters, but you can

01:06:37.060 impute the A1C. My A1C runs very high because I have this condition called beta thalassemia trait.

01:06:43.360 So I have a bunch of these little, I think I talked about this in the podcast on one of the- 0.95

01:06:46.540 Shit for blood is what they used to call it. It's one of your nicknames. 0.98

01:06:48.860 Matty used to call me shite for blood, always in a Scottish accent. You got shite for blood. 1.00

01:06:53.000 Blood. So my little shite blood cells live a long, long, long time. I mean, I have no idea, 1.00

01:06:58.900 but it's clearly longer than 120 days. So my A1C is very high. The lowest A1C I've ever seen in

01:07:03.420 myself is 5.6, and the highest is 6.0. So I'm basically just on A1C, I'm a pre-diabetic pretty

01:07:09.600 much all the time. On CGM, when you take a highly calibrated, rigorous look, my average blood glucose

01:07:18.600 imputes that I would have an A1C between 4.5 and 5. That's sort of the range that I would live in.

01:07:25.560 So that's a material difference. And again, I've seen that difference in both directions

01:07:30.160 with patients using CGM. So my hope is that in 10 years, maybe that's ambitious. I would hope that

01:07:36.600 the hemoglobin A1C can't even be ordered on a lab and everyone just has a CGM, you know, and it's like

01:07:42.000 a trivial little, you know, thing that, you know, even if you're getting a life insurance exam,

01:07:46.060 you just wear the CGM for two months and the data comes from that as opposed to actually

01:07:51.840 measuring this nonsensical number. Is it just me or are there like way more like sirens today?

01:07:58.380 Yeah. 9-11 in New York City. I don't know if that's...

01:08:01.700 Yeah. It's interesting. I feel like it's been nonstop sirens today, which is... I'm sorry for

01:08:07.640 the listener, but... I think they get the heads up that you're doing a podcast and just want to

01:08:11.420 buzz the tower. That's just how it works. I'm surprised there hasn't been drag racing and...

01:08:16.280 No. That 14's going by.

01:08:20.200 I didn't get the permission for a flyby there.

01:08:21.840 No. I don't know where that was.

01:08:23.780 Negative, Ghost Rider. The pattern is full.

01:08:27.440 Lithium or lithium supplementation? Can you share your thoughts slash experiences?

01:08:32.520 Yeah. So I think the evidence that groundwater containing higher levels of lithium rather

01:08:40.900 than lower levels of lithium contributing to better mental health is as good as any such

01:08:49.120 data can be, which is to say not great. I call it like tier two epidemiology. So the worst epidemiology

01:08:55.440 in the world is people who exercise this way have this much life or people who eat this way have this

01:09:01.260 much disease or whatever. That's the worst epidemiology because you're measuring... you're

01:09:05.720 trying to accurately assess inputs and outputs. So that epidemiology, I just think we should stop

01:09:11.140 doing that.

01:09:11.960 For the most part, it's epidemiology is hypothesis generating. And this is an interesting

01:09:16.420 observation, more or less. It's an interesting hypothesis.

01:09:19.460 Yeah. But the reason I'm calling this tier two epidemiology is one of those variables is fixed.

01:09:23.540 Like the groundwater lithium concentration doesn't require you guessing.

01:09:28.840 There's no healthy user bias in who's doing that. The only thing you're measuring is the

01:09:33.500 mental health on the other side. So in other words, that's a better type of epi, just like the

01:09:37.740 IGF one I alluded to earlier. But it's actually not subtle. So if you look at those data, it's actually

01:09:43.500 not that subtle. When you overlay like the US map of groundwater of lithium and mental health,

01:09:49.100 it's a very nice association. Now, there is no doubt in my mind that you could think of other

01:09:55.900 explanations for that. I mean, and that's the beauty of trying to be thoughtful about these

01:10:00.440 things as well. Maybe the lithium in the groundwater is a proxy for something else. Maybe it's a proxy

01:10:05.800 for education, for socioeconomic status, for weather. Like you could think of 10 other things that

01:10:11.220 could be... So all of that said, at the first order, it doesn't appear that there's an obvious

01:10:16.600 proxy. But there still may well be. But at the other end of the spectrum, what's the cost of

01:10:22.160 the intervention of trying it out? And so I decided several years ago that I wanted to try

01:10:28.720 taking some lithium to see if it could help stabilize my mood. But that's sort of a crazy

01:10:35.320 thing to do. So I did a ton of homework. And luckily, one of my best friends, Paul Conti, who I've

01:10:41.100 interviewed and will hopefully be on the podcast shortly, maybe even by the time this has come out,

01:10:45.180 well, that's that true. Do you know the order of those, Nick?

01:10:49.800 Oh, sweet. All right. So you'll have you'll have already met Paul Conti by the time this comes out.

01:10:54.240 Paul has an unbelievable experience with lithium because he is one of the few psychiatrists out

01:10:59.540 there remaining who is still very comfortable using lithium in monotherapy for high risk bipolar

01:11:06.380 patients. So usually a patient with bipolar disorder is treated with several drugs. But in some

01:11:12.240 cases, some of the drugs, either the patients don't respond, they're recalcitrant to the drug,

01:11:16.500 or they produce such negative side effects. They can drive up suicidal ideation, things like that,

01:11:22.760 that you basically have to resort to just lithium. And now that's a really scary. That's about as scary

01:11:27.960 a case as a psychiatrist can be. And because you basically have to put enough lithium in that patient

01:11:32.960 to take them almost to a toxic dose without pushing them over. So I wanted to learn everything

01:11:37.560 about lithium toxicity from talking with Paul. And this is going to sound a little crazy. But when I

01:11:42.300 first started taking lithium, I actually took it at a very high dose, not the dose that someone who

01:11:46.560 was taking monotherapy for bipolar would take. But I was taking about half that amount. So I was taking

01:11:50.920 about 600 milligrams a day. And again, I hesitate to talk about this because I know some knucklehead is 0.80

01:11:56.160 going to go and take 600 milligrams of lithium a day. Let me be crystal fucking clear. If you were not 0.99

01:12:01.760 under the care of a physician who is super dialed in on how to measure lithium levels,

01:12:07.440 when to measure lithium levels, which lithiums you take, because it comes in a bunch of forms,

01:12:11.820 doing this is tantamount to suicide. I mean, it's complete stupidity. So with all of that said, 1.00

01:12:18.460 under the most careful, closely monitored conditions that one could have, including getting lithium levels

01:12:24.180 checked constantly, I spent about a year taking 600 milligrams of lithium a day. In two doses,

01:12:30.820 there's two different types of lithium, a slow release, a fast release. I did it as an interesting

01:12:35.660 experiment and I didn't tell anyone I was doing it because I wanted to know if anybody else would

01:12:40.780 appreciate, because I'm an emotionally volatile, crazy guy. So it was like, there should be a

01:12:45.940 noticeable difference. And interestingly, the first person to notice it was my wife, which was about 0.99

01:12:51.600 four months in when she was like, something's different in you. What is it? Again, I have no idea

01:12:57.580 because of course, by me knowing I was taking the lithium, maybe that altered my behavior.

01:13:02.700 In the end, I decided I did not want to take mega doses of lithium because I did notice

01:13:07.760 I could actually have side effects. The first side effects of too much lithium are nausea.

01:13:14.000 Now, even when you're nauseous on lithium, you still don't reach the blood level that approaches the

01:13:19.600 toxic levels, which makes me wonder how those poor patients who are constantly taking 900 to 1200

01:13:25.440 milligrams a day of lithium are tolerating it. But I noticed whenever I traveled through different

01:13:31.400 time zones, if I took the lithium at different times because it was being compressed, if I was

01:13:37.560 stacking time zones flying east, I would start to get nauseous when I took it. And I was like,

01:13:42.560 what the hell is that from? And I realized, oh my God, like you probably just effectively took 900

01:13:46.280 milligrams. So I was like, you know what? I don't think there's enough upside. I feel a bit better on

01:13:50.520 this, but I don't feel that much better that it's worth this hassle. So I just stopped it for a

01:13:54.480 couple of years. And then recently, like about two years ago, I decided I went back and looked at the

01:13:59.600 data and I was like, look, if those groundwater data are correct, you don't need to be taking that

01:14:04.540 much lithium to reproduce the levels. It's we're talking 10 to 20 milligrams. And that's actually

01:14:08.920 an over-the-counter dose. So, so that's why I take it. Now, again, I have certain, you know,

01:14:15.320 things that I take that I put in the category of probably not harmful, not sure how much value I'm

01:14:20.600 getting out of it. This would be one of those things. This is the, the penny in front of a

01:14:24.940 bulldozer. Yeah. This is the two by two, right? So the reward is, are you getting a penny or are you

01:14:30.640 getting a Bitcoin? The risk is, are you picking this up in front? Is it like picking that thing up

01:14:36.020 in front of a tricycle or a train? This to me is trying to pick up a penny, call it a dollar

01:14:41.920 inflation. It's picking up a dollar in front of a tricycle. Yeah. At the end of the day, Hey,

01:14:47.860 it's a dollar. It's a dollar more than I had before. And if, if I'm wrong, the tricycle

01:14:51.740 hitting me is not a lot of work. What you never really want to do is pick up dollars in front of

01:14:56.680 trains. If you are going to step in front of a train, it better be to get a, you know, a basket

01:15:02.260 a bit. Yes, exactly. Yeah. Yeah. Yeah. So I don't know if you, you said this, but when your wife said

01:15:10.160 there's something different, she noticed something different. Did she say what it is or do you know

01:15:14.500 what was that difference? Yeah. She said, you're less of an asshole. Interesting. It wasn't subtle.

01:15:23.100 She said, I don't know how to put my finger on it. It's something about your, um,

01:15:27.220 assholedness has gone down. Decreased. 0.95

01:15:33.080 Okay. And then when you supplemented on, you said maybe like 10 milligrams versus 600,

01:15:38.680 were you, do you think you're achieving the same effect? Now it's impossible to know because the

01:15:43.660 very first time I did all this stuff was before I meditated. And frankly, before I paid attention

01:15:48.800 to any of this stuff, I think I can be just as much of an asshole today as I used to be. I think 0.99

01:15:52.620 the difference is I'm now so much more aware of it that I can correct my assholeness quicker. Like 0.85

01:15:57.780 I can be an asshole, but then I'll snap out of it in like 20 minutes as opposed to two days. 0.99

01:16:02.960 So I think now there are so many confounders that it would be impossible for me to say, 0.98

01:16:07.800 has lithium made any difference in my mental health? Truthfully, uh, insights about berberine.

01:16:13.700 So berberine is a plant derived extract that has two properties, one of which it gets a lot of

01:16:20.160 attention for one of which it might not be as well known. The first is that it is a weak AMP kinase

01:16:26.520 activator. And for those of you that have shown an interest in what metformin does, that's sort of the,

01:16:32.480 that's probably the secret sauce of metformin is it's AMPK activation. So berberine when taken at,

01:16:41.160 you know, sort of recommended OTC doses, which I don't remember the doses anymore. I think it's

01:16:47.400 like a, it might be actually a comparable dose to metformin. It might be like either 500 twice a day

01:16:52.620 or a thousand twice a day does have some of that weak AMPK activation. And what that, the net effect of

01:16:59.680 that is it decreases hepatic glucose output. So it's upregulating AMPK is telling the liver,

01:17:07.100 Hey, you can make less glucose. So send less glucose out of the body. And so in that sense,

01:17:12.940 it's a poor man's version of metformin. I don't find that, that interesting. I prefer to just use

01:17:18.860 metformin. If we're going to go down that path, like let's do it pharmacologically with potent drugs

01:17:23.620 that we understand that are consistent from batch to batch that have a much higher sort of oversight of

01:17:28.680 regulation and that kind of stuff. The other thing about berberine, and this is the, this is when I

01:17:32.380 still do use it in clinical practice is it is also a weak inhibitor of the enzyme PCSK9, which any

01:17:40.920 listener of this podcast is going to be very well up to speed on what PCSK9 is. The word on the street

01:17:47.420 is it probably only works in a subset of people who overexpress PCSK9. So if you, so PCSK9 is a,

01:17:56.840 is a protein that degrades LDL receptors. So if you overexpress this enzyme, you're going to really

01:18:05.280 degrade LDL receptors. You're going to have a higher LDL particle number and probably a higher

01:18:10.260 LDL cholesterol. If you take that subset of patients, they seem to respond quite nicely to

01:18:15.420 berberine. Now the question is, how do you know that? I don't think there's a test anymore to measure

01:18:21.120 that. Atherotech, which I think might've got absorbed into VAP or something like that. I think

01:18:26.840 he used to have a test, but in the end I was like, why am I going to do some test on somebody to see if

01:18:30.580 they overexpress PCSK9 before I give them berberine? I'll just give them berberine, make no other change

01:18:34.800 and see if there's not a significant enough difference in their LDL. I don't care.

01:18:39.120 And every once in a while, you just, you look like a rockstar. You get this patient whose LDL is

01:18:44.340 really high and you check your boxes. Like they don't have, their triglycerides aren't that high.

01:18:51.300 Their phytosterols are not that high. Their stanols are not that high. Their desmosterol is not that

01:18:56.440 high. Hey, they clearly have defective LDL clearance. Maybe they overexpress PCSK9, which would be one of

01:19:02.740 a thousand reasons you could have high, like literally there are 2000 genetic mutations that

01:19:08.220 can lead to an inability to clear LDL receptacles. This is like one of them, but this one amounts to about

01:19:12.560 5% of them. And so in that patient, you hit them with berberine and all of a sudden like

01:19:17.000 everything's fixed. Do you see that with metformin at all? No, no, I don't think metformin possesses

01:19:21.840 that piece of what berberine has. Okay. Now, what is your reasoning for taking baby aspirin and what

01:19:28.900 does the science say? Well, interesting timing. So the science has changed on this as of about a month

01:19:33.480 ago. So my thinking on this has actually changed in the past month. Historically, I took baby aspirin just

01:19:38.740 because I come from a family where heart disease is rampant and that's sort of, you know,

01:19:42.560 in the spirit of our good friend, Charlie Munger, you brought up a moment ago, you know,

01:19:46.180 show me where I'm going to die and I'll try to not be there. So, uh, you know, the crystal ball

01:19:50.520 says I'm probably going to die of heart disease. So let's do everything in our power to mitigate it.

01:19:54.920 Let's keep the lipids as low as possible. Inflammation as low as possible. Don't do

01:19:59.280 anything that's going to exacerbate endothelial dysfunction. Take the baby aspirin, which probably

01:20:04.420 has some effect on inflammation, but probably the majority of its effect is, is obviously on platelets

01:20:09.380 and therefore reducing thrombotic events, which are the final, like that's, that's what pushes you

01:20:14.980 off the cliff when you have an MI. I would have said that prior to a month ago, my view on baby

01:20:20.480 aspirin was if your aspirin works test is positive. So the aspirin works test is a test we do in urine

01:20:26.000 that measures metabolites of platelets and basically tries to predict which patients will and won't

01:20:35.640 respond to baby aspirin. If that level is high, my practice used to be to put patients on a baby aspirin

01:20:41.640 if there was at least one other reason to do so. So if they had an elevated LPPLA2, if they had an

01:20:47.820 elevated LPPLA, if they had a significant enough risk of heart disease, that I would feel that those

01:20:54.100 two things, and I fit in that category, right? So I have a normal LPPLA2, normal LPPLA2, but my family

01:21:00.500 history for heart disease is high and my aspirin works was high. So that's why I take a baby aspirin.

01:21:05.280 And then you have to weigh that against the consequence of a baby aspirin, which is, look,

01:21:08.540 you're going to have a little more easy bruising. You have, you run a small risk, almost unmeasurable

01:21:13.000 of having an ulcer. Nowadays we're using, you know, because you're using a baby aspirin,

01:21:17.600 it's enterically coded. You know, this is a far lower risk than say taking ibuprofen or something like

01:21:22.460 that. The coating helps it get into the small intestine rather than the stomach. Yeah. It basically

01:21:27.600 prevents it from, you know, sort of destroying the gastric mucosa. Now that said, there was a study

01:21:33.580 that came out probably about three weeks ago, we could link to it, that looked at healthy, so low

01:21:39.620 risk populations. And it found that the risk of taking a baby aspirin was very small, which we

01:21:46.140 always knew, but the benefit was also very, very small, such that it was pretty much a wash. And I

01:21:50.960 would say that, you know, that's the nature of clinical medicine. As more and more data become

01:21:54.840 available, you start to revise your criteria. How will this change the way I practice? I will

01:21:59.280 likely be just a little bit less interested in giving baby aspirin to a patient who is otherwise

01:22:05.080 low risk. And certainly patients that had, didn't have an elevated aspirin works, I would rarely use

01:22:09.400 it anyway, unless there were some other overwhelming reason. Will I keep taking it? Eh. As I look to

01:22:15.440 streamline the list, you know, I'm always kind of tweaking what I'm taking. You know, one thing that's

01:22:19.140 interesting about baby aspirin, it's not a great, this is a great example of like what our research team is

01:22:23.900 capable of, right? I used to also think baby aspirin was a reasonable thing to take on a

01:22:28.480 flight. You know, if you're flying, if you're on a long flight, you don't want to minimize your risk

01:22:32.060 of a blood clot, take a baby aspirin, especially if you have another risk factor. So any patient I

01:22:36.440 had that had an elevated LP little a, which is increases your risk of blood clotting probably by

01:22:40.660 two X, although that's still debatable. Those are patients that should take two baby aspirin on a

01:22:45.060 flight kind of thing. Then we had the team look into this, right? I think Dan looked into all of this

01:22:48.920 stuff and wrote a really nice white paper on it. And it turned out that the data didn't support it at

01:22:52.980 all actually, you know? So you were basically in the camp of either taking flight tabs or low

01:22:57.980 molecular weight heparin, which I've taken low molecular weight heparin, but it's injectable.

01:23:02.180 I've taken it on really long flights. And in the end, I'm like, it's just not worth the fricking

01:23:05.820 hassle. Like I'd rather just get up and walk around as much as possible. So, but yeah, and just a great

01:23:10.600 example of how a lot of times things that are intuitive might turn out to be not necessarily standing

01:23:15.480 up to the rigor of a clinical trial. And these are the things that are easy to study in clinical

01:23:19.700 trials. You know, a lot of times clinical trials don't answer questions that we care about, but this

01:23:24.260 is the kind of stuff where they can give you a pretty good answer. Okay. This is switching gears

01:23:28.640 a little bit. How do you use heart rate variability as a metric in your practice and or in your own

01:23:34.340 personal use sleep, pre post exercise, pre post eating every morning readiness? So when I started

01:23:42.740 getting into heart rate variability, it was long before there were commercial, you know, like rings and

01:23:48.680 wearables that were making it easy to track. So God, I'm blanking on the guy's name who got me

01:23:54.160 into it. His name is Will Eden works for Peter Thiel. And I don't know how I met Will, but you know,

01:24:01.240 we had a bunch of friends in common and somehow we met and, um, probably like six years ago. And I went

01:24:06.580 up and spent a day in Peter's office and he was showing me like the raw data on their heart math.

01:24:12.700 So heart math was a company that was running the algorithm on chest strap. So chest strap that was

01:24:17.580 gathering the data and then heart math was the algorithm. You're doing it. And I was really impressed

01:24:22.360 by the data. Roughly speaking, what is heart rate variability? I guess it's worth defining this. Uh,

01:24:26.400 this would be another great whiteboard discussion. Why didn't we get a little, like we should have had

01:24:30.020 like a little post-it, like a little white little flip chart here. We could, we could just draw on the

01:24:34.900 walls too. I'm sure your kids do it sometimes. You want to hear a funny story about this? So I got in

01:24:40.540 really late last night, but it was still early enough that somehow my son Reese was up. But I

01:24:47.400 like, as I landed, he's just getting ready to go to bed. And I'm like, Oh, I just, I want to like

01:24:51.520 FaceTime him. Right. So we have this thing where I tell him stories every night and I don't know where

01:24:56.120 this came from, but like, I just decided the name of the character was Reginald. I just love the name

01:25:00.820 Reginald. Right. So I'm like, Reginald is a four-year-old boy who has a younger brother and an older

01:25:05.360 sister. And it's like, everything is about Reese, except it's not about Reese. It's Reginald. And

01:25:10.020 Reese is obsessed with trash cans, but Reginald's obsessed with helicopters. So we're into a sweet

01:25:15.640 Reginald story. And you can see my wife is sitting next to him and she's like kind of rolling her

01:25:19.700 eyes. Like, come on, get on with it. Like, I got to get this kid to bed. And I'm like, so I was like,

01:25:24.280 I got to make it a quick story. So I'm like, one day Reginald wanted to draw a picture, but he

01:25:31.220 couldn't find any paper. So he grabbed his markers and he drew a big helicopter on the wall

01:25:37.840 and Reese like sits up and he goes, he did what? And I hear my wife go, good story, dad. And I'm

01:25:45.460 like, oh, I mean, but after he drew the helicopter on the wall, he was grounded for a month and never

01:25:50.840 allowed to draw again. And I had to like walk myself back. It's a good save. Um, I think dads just do

01:25:59.660 that. Like, I think we just give kids bad ideas and we don't realize it because we put foot and

01:26:04.840 mouth disease. I think it's term. Yeah, I think so. Even like today I heard, I heard about it again.

01:26:10.880 She's like, he's still talking about Reginald drawing on the wall. Like just so you know, when

01:26:15.980 that happens, you're the one painting it. It was like, fine. I'm sorry. I'm just trying to tell a

01:26:22.060 story. What was I talking about? HRV. Oh, HRV. Right. So, so when you look at this, so, so people listening

01:26:28.700 to this, I'm sorry, this won't make much sense, but if you're watching this, hopefully you'll see

01:26:32.160 this. So when you look at an EKG, which you probably appreciate is there's like this little

01:26:36.280 P wave, which is the atrial contraction. And then there's the little downtick of the Q,

01:26:41.900 the R and the S wave, and that's the ventricular contraction. And then there's the T wave,

01:26:47.400 which is the repolarization. And so that is like one heartbeat P dip Q spike R S T. Can people

01:26:56.200 see that Nick when I'm doing that? I bet I can figure out the technology that it like makes

01:26:59.700 when you're doing that, we can make the sinus wave. That would be very Rhonda Patrick. Rhonda

01:27:04.560 would do that. That's, that's, that's, that's Rhonda would take it to that level. We might be

01:27:09.360 too lazy. Okay. So let's say somebody's heart rate is beating 60 times per minute. They're,

01:27:14.860 they're at rest. The length of time between those R waves, because the spike of the R wave is the most

01:27:20.900 obvious place to measure the beat to beat time would be one second, right? 60 beats per minute.

01:27:26.700 If you're beating 120 beats per minute, it would be half a second or 500 milliseconds.

01:27:33.080 So what heart rate variability is doing is asking the question, how much variability is there between

01:27:39.460 those beats? And you might think, well, if you're just sitting there at rest, how much variability can

01:27:44.160 there be? And the answer is it depends on what time scale you're using to measure it. If you're just

01:27:49.040 measuring it in seconds, not much at all, but if you're measuring it in one thousandths of a second

01:27:53.780 or milliseconds, there could be quite a bit. So somebody who's EKG or, you know, heart rate

01:27:59.000 measurement looks like it's chugging along at 60 beats per minute. It might still be, you know,

01:28:03.720 970 milliseconds, 1030 milliseconds, 1005 milliseconds, 970 milliseconds, et cetera.

01:28:12.260 There's a mathematical algorithm that you applied that you apply to that called the root mean square

01:28:17.400 of the standard deviation that basically our MSSD that basically turns that number. It's a

01:28:24.880 transformation. So it's, it's in mathematics, we call these things transformations where you

01:28:28.500 basically compress these numbers and you can then now measure how much variability there's going on.

01:28:35.360 Okay. I don't want to get much more into the math on that in large part. Cause I actually don't,

01:28:38.780 I'm not, I'm, you know, like I'm not an expert on this topic.

01:28:41.680 There's a little bit of this on the, the Lustig podcast.

01:28:43.900 That's right. Yeah. We talked about the high, the high frequency versus the low frequency.

01:28:47.080 That's right. And we included that in the show notes. So that's, that's helpful.

01:28:49.860 Looking at the difference between the RR intervals or something like that.

01:28:53.180 Well, the RR interval is the raw data. Like, so I'll back up for a moment. So when I got really

01:28:57.640 into this, there were basically only two companies that were doing the analysis, heart math and

01:29:01.640 first beat. I chose to work with first beat, even though I would say at the time,

01:29:05.980 both of their algorithms were excellent. But the first beat was a, you actually,

01:29:11.420 it came with its own like little EKG thing. So it came with a, you would put a sticker here

01:29:16.000 and you'd put a sticker here. And then basically you hung this little battery packed wire across

01:29:21.340 your chest. Like a couple of leads under. You had a couple of leads. Yeah. And it was actually

01:29:25.760 funny because the stickiness of their leads was so profound. I have like leather skin, like nothing

01:29:31.260 can hurt me. I was like completely tore my skin off. I was like for a year or two years after I

01:29:38.160 started using it, I was depigmented where I had those leads. It was like, so we ended up like

01:29:43.220 realizing, well, this is gonna be really hard to do with patients. So they came out with like a less

01:29:47.420 sticky lead and you know, we got away with it, but it would capture the data. Then you would

01:29:51.600 actually have to send the device. You know, you'd have to plug the device in your computer and it

01:29:56.360 would, you know, sort of do the analysis before I get to what I do today. The more important

01:30:00.520 point is what did that data tell us? What were we looking for? So we were looking for

01:30:03.760 several things. So the device could measure heart rate, heart rate variability and respiratory rate.

01:30:09.740 Obviously it has the, it's an, it has the sensitivity to measure the thoracic movement.

01:30:14.740 It would use an algorithm. First beat had an algorithm and I'm sure heart math has its own

01:30:19.120 algorithm and they're both proprietary to impute from that, whether they believed that the user was in

01:30:26.660 a more parasympathetic or sympathetic state or under a period of profound stress, like exercise.

01:30:33.300 So if heart rate variability became very low and heart rate and respiratory rate were high,

01:30:39.240 it would basically impute that to be, you are exercising and it would graphically represent that

01:30:43.780 one way. If heart rate were low, heart rate variability were high. It would interpret that as low sympathetic

01:30:53.320 tone, low fight or flight tone, high parasympathetic tone. It would represent that a certain way. And

01:30:59.680 you can, you know, extrapolate from there. So the advantage of this is you could get really cool data

01:31:05.460 and you would typically have the patient wear this for three days, taking it off only to shower. So they

01:31:10.600 would exercise with it. They would sleep with it. They would change the lead once a day. It became a

01:31:14.640 really helpful way for us to try to look at sleep patterns. The problem is the compliance was very low.

01:31:21.460 Most patients didn't want to do it. And it took so long to get the data. You know, if a patient was

01:31:25.220 in New York doing this, they'd have to mail the thing back to, you know, Mary and San Diego. And

01:31:29.680 it was just kind of a pain in the ass. So the products like Aura Ring or, you know, there's others 0.99

01:31:36.620 like, you know, Whoop tries to do this. Motive tries to do this. I can't speak to their technology

01:31:40.740 as well, but nevertheless, now we can basically get those data every day when we wake up, just looking

01:31:48.300 at the data from our wearable. So as a general rule, you want to see higher heart rate variability

01:31:52.940 because that's a marker of more parasympathetic flow provided again, that's being measured

01:31:58.000 correctly. So, so the things that I've observed, so, so when you, when you look at the recovery

01:32:02.240 index on the Aura Ring, which is one of its menus, what it's basically doing is using a lot

01:32:06.040 of data, but one of them is the parasympathetic, the, the heart rate variability. So as you over

01:32:10.800 train, your heart rate variability will go down. Your resting heart rate will go up.

01:32:14.400 I think I've talked about this in the past, but alcohol and shitty food close to bedtime

01:32:20.140 will absolutely tank my resting heart rate, meaning it will drive it way up and will drive

01:32:25.560 my heart rate variability down. And it's not subtle. It's not like, well, that's on the edge.

01:32:29.420 No, no, no. Like it's because you see the tracing of the data overnight. It's only reporting the

01:32:34.240 average and the max, but you can just look at the raw data and it's pretty clear that, you

01:32:39.180 know, those things, you know, really diminish those parameters. And by extension, then your,

01:32:43.720 your recovery and your sleep quality. So heart rate variability then becomes one of the parameters

01:32:49.740 along with temperature movement, et cetera, that then feeds into the algorithms that predict sleep

01:32:54.980 quality and recovery. Again, I've kind of forgot the question, but I know it had to do with heart

01:32:59.660 rate variability. How you use it in the practice and personally, and I'm wondering, is it actionable?

01:33:05.480 You've talked about doing deadlifts and saying, I forget the number, but it's, it's not small where

01:33:10.220 you say you'll, you're doing deadlifts. You'll warm up. You're going to be warming up to a heavy

01:33:13.600 and then you actually, you walk away cause you're not feeling it that day. Is there anything that

01:33:17.140 you look at it with the HRV, the readiness or things like that and say, absolutely, I'm feeling

01:33:21.120 great, but this thing is telling me. Absolutely. I, I, I, so there's two things that are really

01:33:25.640 valuable. The first is, and this gets to the point we made with the CGM, which is real time

01:33:30.180 feedback is awesome. Take away real time feedback. It's hard to do anything. I mean, everybody's

01:33:35.040 heard the famous experiments when you put on a headphone that delays your voice, you hearing

01:33:41.300 your voice by a few seconds, you can't carry on for more than a sentence. Oh yeah. Right.

01:33:45.640 I've experienced that. Yeah. Yeah. Yeah. So that's actually, that's horrible. That's like

01:33:48.660 negative feedback. That's not just taking away the feedback. That's like actually inserting

01:33:52.040 negative feedback, which is incredibly destructive. So the first thing is when I wake up in the morning

01:33:56.800 and I see that my HRV or my heart rate or my temperature or my respiratory rate is off. And again,

01:34:02.280 I have my parameters. I know what I want to be. I immediately can say, what did you do different?

01:34:07.300 I mean, that's how I sort of figured out what my tolerance for alcohol was. You can have one drink,

01:34:12.960 don't have two. And by the way, one's not even good. Like one gives you slight hit two is a brick

01:34:18.740 over the head. And probably what time are you drinking? Yeah. Like when are you drinking that?

01:34:22.540 And then like, that's how I've certainly figured out like meal timing, like, Oh boy, you're eating a

01:34:26.580 little too close to bed. Like that's not good. Room temperature. Like all, you know, all of the,

01:34:30.960 the crap that I do is largely based on this empirical iterative tinkering approach, 0.74

01:34:37.780 which allows me to have data. And then of course, by giving patients, these devices, 0.98

01:34:42.000 you can figure out what's true for a patient because it might necessarily be true for me.

01:34:46.100 So in many ways, I'm trying to help a patient understand how they can do the tinkering so that

01:34:50.960 they can figure out their optimal state. I mean, I, for example, I like a room to be as cold as is

01:34:55.280 humanly possible. I mean, if it's like, if I don't get under the covers and feel incredibly

01:35:00.800 uncomfortable, I'm going to be too hot by, by my standards, but that, you know, that might be

01:35:05.580 not at all your standard. Right. So that's that. And then to your point, yeah, the readiness score

01:35:09.480 for me, especially when I'm starting to feel a little sick, sometimes that readiness score

01:35:14.480 becomes the harbinger of that. And also it also gives me some inclination. You'll, you'll often see

01:35:20.480 it dip after you've had two consecutive days of really hard workouts. Now I will say this,

01:35:25.100 I don't think it's as sensitive to lifting workouts. So it doesn't seem to possess the

01:35:30.900 ability to distinguish between like you did deadlifts until you puked versus you, you know, 0.51

01:35:36.260 you're at the hotel gym and you did leg press. Like it can't, but what it sure as hell knows is like

01:35:42.120 you went out and, you know, expended 1500 kilojoules yesterday on a bike or a thousand kilojoules

01:35:49.600 in a run where you ran at, you know, seven minute miles. Like it knows that. And it, you,

01:35:54.760 you, you stack a couple of those days on top of each other and it says that's, that's serious

01:35:59.620 illness. I'd love to see the data on, you know, cause I, I think a couple of teams in the Tour de

01:36:05.560 France wore the aura ring this year. I'd love to see what those data were like. Their recovery

01:36:10.040 scores must've been the lowest numbers imaginable because of the physiologic stress they're under.

01:36:14.280 So despite the fact that they're so fit and like, you know, the most remarkable physical

01:36:18.860 specimens, you know, I mean, look at the sort of the trend over the course of the three weeks

01:36:23.800 and see the dip. Yeah. Yeah. Really interesting. Oh, I, I, I, I need to make a note to, cause I,

01:36:30.180 I I'd like to follow up and see if they've, if anyone's ever been willing to share those data.

01:36:34.040 And you mentioned, you may have mentioned this before, but on your, when you fasted,

01:36:37.940 you did a week long fast, you said your sleep was remarkable. Yes. Did the aura ring, did that,

01:36:42.520 did it corroborate it in any way when you looked at the data from the aura ring that it tells you

01:36:46.900 anything? Rem sleep, deep sleep. Yeah. Uh, the biggest, the deep sleep is what went up the most.

01:36:52.600 So the deep sleep went up the most. The light sleep went down the least, went down the most.

01:36:56.760 So lights. So stage one and two went down stage three and four went up. Rem was about unchanged.

01:37:01.820 HRV was higher. Resting heart rate was on par with what it is. If you're not eating and drinking

01:37:06.340 before bed, I'm trying to think what else I have to go back and look. I think respiratory rate went

01:37:10.880 down, which I, I think is a VCO2 thing. I think that the shittier and more carby a meal is before

01:37:20.080 bed, the higher your VCO2, you're really trying to blow off more of that CO2. Yeah. We know that

01:37:24.720 your respiratory quotient is higher. So unless your VO2 is going up, your VCO2 has to be the thing that's

01:37:31.560 going up more. That would, wouldn't surprise me to see respiratory rate go up a little bit.

01:37:36.100 But anecdotally, that's at least what I think I'm seeing. Interesting.

01:37:42.160 Dude, are we halfway through this yet?

01:37:46.700 What are we at? We're at an hour 45? God damn. I, I freaking talk too much, man.

01:37:56.140 I have a bonus round. We've got through like five questions. Yeah. Yeah. We could get through

01:37:59.880 more questions if we, if we do a gimmick. I mean, you love gimmicks. You love bumper. You know,

01:38:04.680 I love gimmicks. Bumper sticker. Love it. Yeah. Wallpaper your cars with, with bumper stickers.

01:38:10.820 I was thinking we could do like a speed round or I used to watch pardon the interruption where it

01:38:15.440 was, I think they had different segments, but probably like 90 seconds. Nick doesn't have a

01:38:19.800 bell that he can ring with the 90 seconds, but maybe he can, he's got a timer. He's got a speed

01:38:24.740 master professional is what he's got. I'm not wearing my, uh, my Casio calculator watch, but

01:38:30.080 I'm pretty sure it's better than the speed master. I could be wrong. I don't know. The moon watch,

01:38:36.780 man, it might be the single most special watch ever made. So do you want to give this a shot?

01:38:41.580 I basically will. I got one minute to answer a question. Yeah. 60 seconds to 90 seconds.

01:38:46.540 All right. So why don't we just say 90 seconds? Nick'll bludgeon you with a club. If you don't,

01:38:50.440 if you go over, yeah, 90 seconds. So here we go with the emergence of the coconut oil is pure.

01:38:58.260 Does he start timing after you finish the question? I was thinking about that as I was

01:39:02.000 reading. Yeah. After the question, after the question, 90 seconds, he'll scream out every

01:39:06.140 five seconds, how much time you have left. Cause I know that that would really be, that'll help.

01:39:09.300 Yeah. Yeah. Yeah. That'll be good. With the emergence of the coconut oil is pure poison article.

01:39:13.960 Can you shed some light on saturated fat in the literature and the types of studies done

01:39:17.680 specifically on coconut oil? It's a great question for 90 seconds and go. I'm going to time it on

01:39:23.420 here too, just so I can keep you honest. Um, 85 seconds. I think that that whole coconut oil

01:39:30.520 literature stuff is sort of ridiculous because I don't even want to get started on nutritional

01:39:34.940 epidemiology. I've already alluded to it a bunch of times. I think it is the lowest form of human

01:39:40.700 inference imaginable, which is not to say it provides zero benefit, but it's at the hazard ratios that

01:39:47.220 these things typically come up coupled with the reliance on food frequency questionnaires,

01:39:52.940 something I've even filled out myself just to prove how ridiculously useless they are. Like

01:39:57.100 ask me what I ate two days ago, not a chance. I'm going to be able to give you anything within an

01:40:02.640 order of magnitude of reality. All that said, I actually think the case for saturated fat may be

01:40:09.540 overstated. Meaning I think that the view that saturated fat is never harmful and we should be able

01:40:15.260 to ingest it at the tune of 90% of our calories. Uh, I think we have to accept the reality that

01:40:20.540 that, that can't be healthy for everyone. Certainly clinically, I don't see that to be the case. Uh,

01:40:25.600 I know I've only got 24 seconds left. So what I would say is rather than rely on some knucklehead

01:40:32.060 study put out by some bottom feeding group of, you know, information providers learn to understand

01:40:39.140 what the biomarkers are that change as a result of it, both inflammatory lipoprotein lipid wise,

01:40:45.260 time go done. Would you discuss the recent meta studies that claim moderate carbohydrate intake

01:40:52.500 may be best for health? You might be able to cite your previous question. So citing my previous answer.

01:40:58.560 Yeah. That study that in particular you're referring to, I'm sure we'll link to it. I don't even

01:41:02.380 remember. It must've come out of the Harvard school of public health, right? Yes. Okay. Yeah.

01:41:05.460 So the bottom line is, look, it suffers from all of the usual problems of epidemiology. Uh, and this,

01:41:12.540 by the way, Zoe Harcum actually wrote a great piece on this. So I, I, she will just point people

01:41:16.800 to that because I think she did a better job explaining this, all the usual epi nonsense.

01:41:21.120 Then on top of that layer in the, like the nutritional epi nonsense, which is like, again,

01:41:25.840 nutritional epi is like its own brand of really bad epi, which is in and of itself a horrible

01:41:32.180 brand of science. And then on top of that, there were some really, you know, systemic issues.

01:41:36.920 They, you know, she points out they didn't correct for alcohol consumption. They had some other really

01:41:41.140 odd confounders in there. The hazard ratios were themselves quite low. And they, you know, again,

01:41:46.340 this is more of a reporting issue than it was an issue of the actual survey. But again, all of these

01:41:52.860 risks always seem to be discussed in absolute terms, in relative terms, when in absolute terms,

01:41:57.200 they're not, I mean, people have heard me rail on the women's health initiative for the same reason

01:42:00.480 in terms of the hormone replacement therapy component of that, you know, talking about a 24% increase in

01:42:06.420 breast cancer in a relative sense when the absolute increase was 0.1%. These things to me are an

01:42:11.560 abomination and I'm going to just time out at a minute 26. Yeah. And I think anytime you look at

01:42:17.180 relative risk, it doesn't really tell you anything unless you also know they tell you what absolute

01:42:21.820 risk is too. You should always look at those two things together. Yes. You have to know, look at,

01:42:26.400 notice how I'm using a little bonus time here. We're outside of the time and he's asking me a

01:42:30.280 question. So I'm just going to take it. That's right. You have to know absolute risk. You have to know

01:42:34.220 relative risk and you have to know the period over time under which the condition was studied

01:42:38.600 relative to the natural course of the illness. It's that third part that doesn't get enough

01:42:43.000 attention. So anyway, I'll off the soapbox. What is the number one recommendation habit you

01:42:50.000 would suggest every person add to their daily regimen besides physical activity in parentheses

01:42:55.300 for wholesome health? Okay. That's exactly the kind of question. Like I never want to talk

01:42:59.040 about it. I don't know. I mean, again, you can't answer this question without sounding glib. Like

01:43:04.700 get a lot of sleep. Do something physical every day. Don't eat for a long period of time. When you

01:43:11.900 do eat, don't eat crap. Meditate. I'm going to leave it at that. Number one. Was there a number one? 0.95

01:43:19.560 The number one? You know, I know we could probably work it in somehow to say like the number one,

01:43:24.660 the number one habit or the number one habit that you could get into is to, is to, it's almost like

01:43:29.460 Feynman's first principle. You could say that like, don't just, don't take someone's number one

01:43:33.140 recommendation. Okay. So that would be the elegant thing to say. The one thing I would say on this

01:43:38.520 that's important to understand is probably sleep is the one we are going to be most sensitive to.

01:43:43.780 In other words, if I said to you, don't eat for this many days or don't eat well for this many days,

01:43:49.380 I mean, you could get away with that for a long time. I think an average person could probably go 30

01:43:53.660 days without eating. They could go years without eating well and still make a recovery. How long

01:43:59.560 could you go without sleeping before you're completely off the rails? I mean, it's days.

01:44:05.700 So in many ways, you know, we say sleep might be the single most important, but that's probably not

01:44:10.920 accurate information. That's probably not the correct way to say it. The right way to say it is

01:44:14.620 probably sleep is the one that you will suffer from the quickest. Yeah. That was close. That was like

01:44:19.600 129. What does it mean if your body has a harder time getting into ketosis via fasting

01:44:24.580 than it used to testing using a precision extra? Oh, that's an interesting question. So the

01:44:30.360 implication is that at one point, this person was fasting, getting into nutritional ketosis,

01:44:35.240 uh, or getting into starvation ketosis, presumably. Is this filler filibustering right now?

01:44:39.560 Yeah. Is this the timer going? Yeah. Yeah. I think it would be similar to, to like people talking,

01:44:44.740 like I went on an Atkins diet and I, you know, the, the weight melted off. And then 10 years later,

01:44:49.200 I went on an Atkins diet and I was like, I could, my weight wasn't really. So I don't know if I infer

01:44:52.740 that from the question. See, to me, they're talking about the numerical values of the BHB level.

01:44:56.980 Yeah. Um, well not, no, yes. Yeah. Yeah. So it's different. It could be a different interpretation.

01:45:00.940 So, so the, those are two entirely different questions. So we should just pick one and answer it.

01:45:04.660 If we, if the question is, I don't get to the same levels of BHB using a precision extra,

01:45:09.120 because they went out of their way to explain how they were measuring it. I think that's what they're asking.

01:45:12.900 It could be a number of things. It could be that they are more efficiently utilizing their ketones

01:45:16.780 now. So you're accumulating fewer of them because you are drawing them off. You are utilizing them

01:45:22.240 more. They have become a more preferred substrate. It could be that they're actually just producing

01:45:26.100 less of them. In other words, they're doing something different that they didn't realize

01:45:29.220 they were doing in the past. Maybe they've changed their composition of fatty acids.

01:45:32.840 Maybe they're eating more protein. Maybe they have more carbohydrate in their diet,

01:45:35.860 things like that. Hard to know without knowing a lot more detail, but those would be the

01:45:39.920 directionally the two things I would explore. Yeah. The first one being impossible to really

01:45:45.320 demonstrate outside of a laboratory. Yeah. I think it's a great point. Yeah. The, a lot of,

01:45:50.680 some of the pee strips is another way to look at ketones. Is it a acetoacetate? Is that the way

01:45:55.340 they look? Yeah. When they're looking at that and you think, well, my ketones are through the roof

01:45:59.080 because I'm peeing out tons of ketones. Like, is that necessarily optimal to be?

01:46:03.360 Yeah. Anecdotally, it seems to go down over time. The more adapted a person gets because their body

01:46:10.260 realizes, Hey, look, I mean, I'm not going to waste this fuel here. I'm going to utilize it.

01:46:13.920 Yeah. Can you give a 90 second reasoning for why you are, uh, why you are taking Zetia and

01:46:18.980 Lipitor? Are you mitigating risk based on your ApoE4 or is there something else going on?

01:46:24.000 So actually I'm not taking them anymore, which speaks to my point of I'm always changing meds in

01:46:28.240 flux, but why I was taking Lipitor and Zetia at the time, and we'll always continue to vacillate

01:46:33.820 into and out of lipid lowering medications is I'm trying to live as long as I can. And for me,

01:46:39.540 that means delaying the onset of atherosclerotic disease as greatly as possible. So, you know,

01:46:45.400 why you wouldn't want to take an all hands on deck approach to that, which is reducing the burden of

01:46:50.120 lipoproteins, maximally inflammation, endothelial health, insulin, all these things. We want them to be

01:46:55.580 all as low as possible. As for the choices of those two agents, Lipitor, it's all empirically

01:47:01.220 derived. So, so through trial and error, I have figured out that, you know, I synthesize a

01:47:05.440 reasonable amount of cholesterol and taking 10 of Lipitor three times a week was actually able to,

01:47:11.040 um, when coupled with ezetimibe, which is a, works via a different mechanism, but I could measure my

01:47:15.660 phytosterols. They were quite high. I knew I'd be a pretty strong responder to it. So by, you know,

01:47:20.660 so the 10 plus 10 stack of those two was a minimum effective dose that produced, you know, very good

01:47:26.920 results in me, meaning it got me to below the 20th percentile, typically to the 10th or 15th

01:47:31.000 percentile. And, you know, so 30 milligrams of atorvastatin a week is almost a placebo dose when

01:47:38.180 you consider what the drug can be given at 80 milligrams daily. But today I'm doing something

01:47:43.880 different time. Okay. You have mentioned a few times on your podcast that you are currently

01:47:48.580 working on writing a book. What will the book be about and what is the expected release date?

01:47:53.580 So the book will be about my experience as a shepherd. What a lot of people don't know about me

01:47:58.280 is prior to, you know, the stuff I do now, I was a shepherd and, you know, I just, I think there's

01:48:05.760 not a, not, there are not, there are not a lot of, you know, pop culture references to shepherds.

01:48:10.300 We don't, you know, there aren't really sitcoms about shepherds. I mean, most people just don't

01:48:13.860 know much about what the day in and day out life of a shepherd is like. And so I thought that would

01:48:20.260 be a good story to tell, to just talk about my experiences doing that. And, um, so, you know,

01:48:26.880 it's been a tough book to write because obviously I have to go back through all of my shepherd journals,

01:48:31.600 um, which are all bound in leather sheepskin, uh, bindings. They're beautiful.

01:48:38.140 And I hope to have the manuscript completed by the end of this year, which would mean, uh, probably

01:48:46.060 a release date in early 2020. Now the drawback of having an early 2020 release is I suspect virtually

01:48:55.380 everyone in this country will be hyper fixated on politics at that time as we'll be spinning up to

01:49:01.160 an election cycle. And if your choice is trying to understand if someone's going to challenge the

01:49:06.940 incumbent president from his own party and who the opposition will be versus the life and times of a

01:49:12.420 shepherd, I think it's going to be an uphill slog to try to get the attention of the masses. But

01:49:17.420 you know, I think one of the other drawbacks of a 2020 release date is people are probably dying

01:49:22.500 for that book to be released and they have to wait a couple of years. That's right. Yeah. I think

01:49:27.440 there's anything that you can do to expedite this process. We'd all be grateful. Believe me. I wish I,

01:49:33.860 I wish I, I wish I could, I wish I could get to this quicker. What are your thoughts on nicotinamide

01:49:39.460 riboside supplementation for longevity? Oh boy. There's no freaking way I get to do this in 90

01:49:44.920 seconds. Well, first of all, the good news is I just had an awesome discussion with David Sinclair

01:49:49.100 about this a week ago. So in the next two months or whatever that, again, I keep saying that like I

01:49:54.060 know when this is going to come out sometime in October, November, the interview with David will come

01:49:57.420 out. Actually, since that interview, a really interesting paper has come to my attention written by one of my

01:50:02.720 medical school classmates, Josh Rabinowitz. So I actually read both of these papers on the plane

01:50:07.140 yesterday. They're freaking phenomenal and emailed Josh a whole bunch of questions to which he

01:50:12.820 responded. I would say the long and short of it is this, and this is not going to make me very popular

01:50:16.940 with anybody. I am completely unconvinced that taking supplemental NR or even NMN by mouth is doing

01:50:25.760 anything other than enriching the companies that make those things. Let me repeat that NAD,

01:50:31.620 which you need in a cell. You could argue having more NAD in a cell is a better thing. That's a

01:50:36.380 second order question. I'm asking for an exemption to my 90 second rule. This is such an important

01:50:40.940 question. Nick is nodding, by the way, just so you guys know. I'm getting the okay.

01:50:44.720 So you could say it like a heavyset, a heavyset judge. I'll allow it, but it better be good.

01:50:49.940 Oh, and then just finding out David Sinclair is going up on November 5th. So you'll,

01:50:52.920 we'll go into much more detail around sirtuins and NAD, but nevertheless.

01:50:55.680 That was Comic Sans too, the font.

01:50:57.180 Yeah. What the hell was that shitty ass font? Calibri. 1.00

01:51:01.000 Calibri. No, no, we should do that. You shouldn't be allowed to have that on your computer.

01:51:05.960 Avenir, Times, there's lots of options, but not Calibri, please. That's unacceptable.

01:51:12.980 I think he did that on purpose. There's no way.

01:51:15.940 Passive aggressive. Hieroglyphics.

01:51:19.420 So cells cannot take up NAD. So a cell has to be able to make its own NAD. So the idea of giving

01:51:28.340 precursors has become obviously the most interesting idea. Now what Rabinowitz's paper showed, and we

01:51:33.760 should link to this. So this was in cell metabolism. It just came out like if, I mean, in the last few

01:51:38.860 weeks, it's a 32 page paper. It took me a while to get through, but basically Josh, who was, like I

01:51:45.060 said, was one of my med school classmates and is still pissed off at me that I didn't invite him

01:51:49.760 to Easter Island two years ago, to which I feel horrible. And I, it's only, and it's a terrible

01:51:54.880 omission. And he has already received an invitation to our next trip to Easter Island next year. He

01:52:00.560 developed, his lab developed a tracer to track all of the intermediaries of NAD and all the precursors.

01:52:06.520 So you can give NR, NMN, and these things can be taken into a cell and then converted into NAD in

01:52:15.060 the cytoplasm. And it appears that according to another paper of Josh's that we could link to if

01:52:19.460 people are interested, it appears that NAD is formed entirely in the cytoplasm and then imported

01:52:24.980 wholly as NAD into the mitochondria, where in theory you would want to have that higher concentration.

01:52:30.020 Here's what the study showed. When you gave oral NR or NMN, the two popular precursors,

01:52:38.000 only the liver could take them up and make NAD using cryptophan. No other cell in the body could

01:52:45.000 take it up. So that would suggest to me that if you're taking oral NR or NMN, you're pretty much

01:52:51.100 just giving it to your liver, which is not exactly the place you want it to be. This would not be

01:52:56.120 changed by using terastilbene. That wouldn't impact it at all. Similarly, all those clinics that are

01:53:02.000 out there giving NAD infusions, which is very popular, that's always struck me as quackery.

01:53:07.400 And it just seems even more quackery today because none of those cells are able to take up NAD.

01:53:12.380 So it would seem to me, and I actually emailed Josh about this last night and he emailed me back today

01:53:17.160 and agreed with my assessment. Based on all of these data, it would seem that the only way to increase

01:53:22.760 cellular NAD would be to use intravenous doses of NR or NMN. And unfortunately, I just hadn't seen

01:53:30.400 that paper because I would have loved to have had that discussion with David because he might have

01:53:33.320 a counterpoint to it. He might be unaware of it. I'm not really sure, but that to me is very

01:53:38.040 important. And I've actually already spoken with a number of our patients who take supplemental NAD

01:53:42.500 and I've, or take supplemental NR and I've already said to them, you know, look, I think you're sort of

01:53:46.620 flushing money down the toilet. Again, I don't think it's harmful. You know, this isn't like a four alarm fire,

01:53:50.140 but I think based on these data, I would have a hard time recommending that anybody take those

01:53:53.440 products. I really went over on that one. I think I heard four hours and four, four minutes and 40

01:53:59.440 seconds. I think Dave Asprey mentioned IV NR. He had Charles Brenner on his podcast. Charles Brenner,

01:54:06.280 I think is involved with, I think it's Niagen, Chromadex, that company. And then there's the other

01:54:10.320 company, Elysium. Elysium spaces. Yeah. But I think they, they talked about it, but I think Charles was

01:54:15.400 making the argument that you got to take NR or NMN, not NAD. Oral, I think it's just NR for Niagen.

01:54:22.620 And then basis is NRPT, terostilbene, which is like a, it's like a methylated resveratrol,

01:54:28.980 more bioavailable supposedly than resveratrol. Although as we learned from David, very, very fat

01:54:34.380 soluble. So if you're not taking it with a boatload of bile, probably not doing a hell of a lot.

01:54:39.620 Yeah. Which brand of supplements have you found effective?

01:54:43.060 Well, I mean, the problem is the whole supplement world's kind of a shit show, 0.99

01:54:46.420 so unregulated. So at some point, maybe we can talk about branch chain amino acids because I know 0.99

01:54:51.960 that they're, I use them, many of my patients use them. And I've learned that most brands out there

01:54:58.660 are not doing anything. But if you talk about like kind of the go-to supplement, my, the companies that

01:55:03.780 I rely on and rely on their products are basically Gero, J-A-R-R-O-W, pure encapsulations.

01:55:11.040 I'm not going to spell that because I can't, I'm like dyslexic when it comes to spelling. Like I

01:55:15.840 can't spell to save my life. Oh, we got spell check. Yeah, I know. But like as I was at the

01:55:21.260 spelling bee, I was always the kid that like got out first. Okay. Spell the T-E-H-T. No,

01:55:28.400 sit down at you. So pure encapsulations, Gero are probably the two companies that I would be my go-to

01:55:34.500 for virtually anything. I'm sure there are a couple of other companies that we will intermittently use

01:55:39.720 depending on the product. Yeah. But those are DHA or EPA, it's Carlson's or Nordic Naturals.

01:55:45.000 Yeah. Yeah. Nordic Naturals and Carlson's I like for EPA, DHA, if not going down the

01:55:50.360 pharmaceutical route, those are also available pharmaceutical. And you alluded to this point

01:55:53.620 with metformin versus berberine that there's different regulation going on for those two,

01:55:58.880 two products. And that's partly why you, you would go to metformin if berberine was a

01:56:03.860 pharmacological agent. I know it's, you know, big pharma. Yes. But the, the scrutiny that they're

01:56:08.560 under is a lot more than a supplement. And I think for berberine, when we do suggest patients

01:56:14.120 take it, I think we use Thorn as the brand that we recommend. I know we went a little over there,

01:56:19.400 Nick. I could see you flailing, but I figured I spent, I wasted part of my time explaining how

01:56:24.540 poorly my spelling was. So. Are you currently accepting new patients?

01:56:28.100 Probably not for the remainder of this year. And I don't manage that. Uh, so Mary who runs sort of

01:56:35.320 the operations of the practice manages the flow of patients. And, uh, she gave me an earful a couple

01:56:41.060 of days ago that we are not accepting any new patients just based on how overworked the team

01:56:46.480 is at the moment. And the followup question probably as unsatisfying as the answer maybe is

01:56:50.820 like, how do I find a Peter Atiyah clone in my area other than maybe your brother?

01:56:55.600 Yeah. My brother is probably the best Peter Atiyah clone. Although it's funny. I don't really 0.98

01:56:59.380 think we look that much alike, but everybody says we look so much alike. I mean, he's so much bigger

01:57:03.220 than me. You know, we just don't look alike. Did I send you guys the video of him in the grocery

01:57:08.000 store pretending to be Borat? Yes. Yeah. You thought that was me? Yeah. You're just saying that

01:57:12.980 because of the point I'm making. No, that was Paul. All right. I guess the point is, yeah,

01:57:19.940 we look, we look a lot like, so, so you could, so my brother lives in Toronto. Got a Tom Morello

01:57:25.680 and have him give you medical advice. That's right. So Tom Morello or, and who, I guess I'm

01:57:30.400 guessing he lives in LA. I don't, I have no idea where Tom would live or my brother lives in Toronto.

01:57:34.680 So that'd probably be the best way to get a clone of me. And then otherwise, I think it, I think it

01:57:39.960 would be a great thing if there's, there would be some way that you could endorse other doctors,

01:57:43.520 but it's not something that is easily done. It is something we talked about actually when we

01:57:47.480 started Nerd Safari and, and I don't, I mean, I think it is something that would be an awesome

01:57:51.280 service is to create kind of like a, I was going to say Tumblr and then I was going to say Grindr

01:57:59.080 to connect doctors and patients, but I'm like, I'm not even going to make the joke, but like a

01:58:03.680 matchmaking service between doctors and patients where you could sort of say, look, without providing

01:58:08.680 an endorsement, cause I don't think I can be in the business of endorsing people. I know there are a

01:58:13.060 ton of docs who think the way I think. And if there are patients out there who want those docs,

01:58:18.600 we, I don't know why we can't have a play, a forum where they can sort of coexist because

01:58:22.520 you know, the, the first hurdle is just the geography of it. Right. It's like, look,

01:58:26.620 if you live in Kansas city, you're not going to want to see some doctor in New York. Even if,

01:58:30.900 even if I had the room in the practice, it just wouldn't be practical. So we just have too many

01:58:35.140 things on our plate right now. And I feel kind of bad cause we've talked about doing that and we

01:58:38.320 haven't done anything about it yet, but, but maybe at some point, you know, we can, cause it's

01:58:42.980 probably one of the questions we get asked consistently more than anything else. I certainly

01:58:45.920 feel like I get a lot of that over Twitter and it would, it would be a great thing to do is to be

01:58:48.940 able to say, look, uh, if you're a doctor who adheres to a certain set of principles, I just don't

01:58:53.480 know how to do it without being kind of hokey and cheesy. Cause like, I don't know what we're

01:58:56.740 advocating. I don't know what, when people say they want a clone of Peter, what do they mean? Like

01:59:02.160 a bald guy? Like, I don't know what they actually want of me. I think that my experience

01:59:08.300 from the people that I know and from, you know, the stuff on the interwebs, the Twitter sphere and

01:59:16.500 Facebook, but like the, the honest response, I think that I, that resonates with me the most is

01:59:23.820 that they, they hear, they'll read what you wrote. They'll hear what you have to say. You'll talk about

01:59:28.600 your strategy with a patient, the objective, the strategy, the tactics, the things that you're doing,

01:59:33.680 the things that you're thinking about. And then they think to themselves, why the fuck is my 0.99

01:59:39.160 doctor not doing any of that stuff? And I bet if I asked my doctor, they wouldn't know, they would 0.99

01:59:43.820 barely even know, or they would say either like, it's not important or I haven't heard of it or

01:59:48.300 whatever the case is. But I think it's like the level of deliberate care that you put into each

01:59:52.840 patient, at least as my personal experience being a part of this. And so I think part of it's like,

01:59:57.820 they're frustrated and they think like, I want to find somebody who can be this comprehensive and

02:00:02.740 deliberate and thoughtful. So what we should do is just have patients come up with a, maybe,

02:00:07.540 maybe we could figure out a way to have patients generate like kind of a questionnaire of things

02:00:11.280 that they want in a doc. And we could, we could be like a central clearinghouse of that where,

02:00:16.640 you know, the, the patients would input and vote on questions that they want asked. And then

02:00:22.540 doctors basically would post their answers to those questions along with where they are.

02:00:28.280 If they're taking patients, do they take insurance? If so, which ones, like, are you Medicare? You

02:00:32.440 know, like there's a whole, there's like 10 things you'd want to know. And again, I don't want to make

02:00:36.520 more work for us because I can see Nick, like giving me the eyes right now. That's like, shut the

02:00:40.700 fuck up. But that sort of puts the onus, I think on the patients and the physicians who are willing to do 0.99

02:00:45.560 the work, I'm sure. Cause I know there are docs out there who are doing the type of work I'm doing.

02:00:49.800 And this would give them an opportunity to sort of explain what they're doing and say, yeah. And

02:00:53.980 like, you know, I I'm in San Diego or I'm in Kansas city or I'm wherever. So maybe let's put

02:00:59.120 that on the list for our call later today. When we talk with Andrew, we can figure that out.

02:01:02.820 And my personal experience too, if I don't get emotional about it, but thank you. My dad was

02:01:06.560 in the ICU for like 30 days. I think it was like 30 days at MGH. Lazarus. Yeah. Like a Phoenix

02:01:13.600 rising from the ashes. He was in rough shape and I would, I would report back to Peter and

02:01:19.780 some other people and my family. And so I think Peter probably appreciated more than my family

02:01:24.240 that we're getting a bunch of numbers spit back and not, you know, you know how he was

02:01:28.180 doing otherwise. Although I was reporting that too. But I mean, I learned a few things, learned

02:01:33.100 how much you know about this stuff, obviously. And that MGH, they, I think they'd like validate

02:01:37.780 your parking after 30 days. So they booted them into the respiratory unit, the rack you.

02:01:42.720 So he went down there for a little while. But one of the things I learned too, as well,

02:01:46.500 is that I would, I would bring back the stuff and the vitals and whatever. And Peter would

02:01:50.500 say, this is, you know, this is sort of what I'm thinking. And then at MGH, I would be doing

02:01:55.140 the rounds with the, with the team over there and everything and being a fly in the wall and

02:01:59.660 adding my two cents and some of the notes that I'd taken and things like that. They're, they're

02:02:03.920 damn good at there at the MGH. And I also thought that that, what a learning experience. Like 0.63

02:02:09.420 I don't, I didn't go to medical school, but my advice would be like, if you want to be

02:02:12.780 immersed in medicine and just the real deal where people are dying and just fast forward,

02:02:19.500 you're learning. Holy shit. Those, those people like the, the ICU is incredible experience. 1.00

02:02:24.240 The ICU is incredible. Yeah. Yeah. And, and, and we joked about it at the time,

02:02:28.960 which was you basically did an ICU mini fellowship because by the end of those 30 days,

02:02:33.440 we'd have our call every morning and you would, you know, we would review in systems because in

02:02:37.680 an ICU, you review everything by systems. You start with the neurosystem, you go to the respiratory

02:02:40.860 system, cardiovascular system, renal system, boom, boom, boom, boom. You know, it only took like a

02:02:44.620 week for you to learn how to gather data and then, you know, how to report it. And, uh, I'm convinced

02:02:51.820 like if you had one year of sitting in there, you would know 80% of what a critical care doc knows.

02:02:57.780 I mean, it's just reps, you know, it's just that kind of exposure. Yeah. And you're right. It is,

02:03:03.540 again, it speaks to this point of feedback that we, this theme keeps coming up today. I like how it

02:03:07.240 keeps coming up. The reason I think critical care is so interesting. And at the same time,

02:03:13.280 so teachable is the cycle time between intervention and feedback is so short.

02:03:18.640 You want to know what happens to the pH on the blood and the VCO2 and the bicarb. When you change

02:03:25.740 the ventilator rate, you're going to find out in about 12 minutes, do it. You want to find out what

02:03:29.580 happens when you tweak this amount of norepinephrine and this much of vasopressin and this much of this

02:03:33.280 and the IV fluids and boom, boom, boom, look at the Swan Gans catheter. Like you're going to see

02:03:37.240 it in a minute. Very few things in life offer you the amount of feedback you get in a critical care

02:03:42.200 unit. It's definitely a cool field of medicine. Yeah. The few patients that you have, I would say

02:03:46.880 that what you do in a sense is critical care, that the level of care that they're providing at the MGH

02:03:52.660 and the ICU. And then I, I see what you do in your practices. I see similarities that you're

02:03:58.380 the, your involvement with the patient. Well, I mean, I mean, I said, I don't want my head to

02:04:04.900 swell too much. I want to be, but I want to point something out, which is, it's really

02:04:07.520 what the patients don't see is what's more impressive. The patients generally just interact

02:04:11.880 with me, but what they don't see is, you know, Ralph and Nicole and Mary and Heather. Like they

02:04:16.740 don't see the whole team behind, they don't see the research team. So, and I was talking about this

02:04:21.860 with Ralph this morning. So, so we saw a patient and spent 90 minutes with him and it was awesome.

02:04:27.160 And he had come in and looked at everything in his labs. So we finished reviewing all of his labs

02:04:33.380 and he goes, Oh, by the way, I noticed you checked my ESR level and it was really low. And I was like,

02:04:39.340 that is so funny. Like, yeah, I checked his ESR cause his CRP was 6.6 last time. And I just wanted

02:04:44.400 to make sure he didn't have something stupid going on. And his ESR came back low and his CRP 0.99

02:04:47.760 returned. And I was like, so after he left, I was like, Ralph, that, how awesome was that? Like

02:04:52.400 that patient had read every single one of the 87 numbers that we checked and was, was asked about

02:04:58.180 the one that I forgot to bring up when we were talking. And I was like, like, I love that. You

02:05:03.400 know, I wish every patient could be that obsessed with their numbers because those are the patients

02:05:08.120 that get the most out of us that extract the most value from us. Um, I find the patients who

02:05:13.240 are least probably engaged with us are the ones that are also getting the least out of us and

02:05:18.280 probably, frankly, don't appreciate the potential. I think that we have as a team to kind of, to,

02:05:24.180 you know, move the needle. You're asking, I think you're asking a lot of your patients,

02:05:27.760 like the be your own advocate. It's really, it's the patient that has the most control over

02:05:32.680 their destiny in many ways. Yes. You, you have to put as much effort and emphasis into this as you

02:05:38.000 would any other thing that you care deeply about. So if you want to, if you want to take your golf

02:05:42.240 seriously, like you can't just think about it three, you know, 30 minutes, four times a year.

02:05:48.000 If you really, really want to be a thoughtful investor, like, yes, you will have professionals

02:05:52.380 help you invest, but you still have to interact with those advocates constantly.

02:05:56.360 Okay. That was a little over 90 seconds.

02:05:59.600 Did that, how, how much over 90 was that Nick?

02:06:04.200 Okay. Are we done? I think we, this has been way too long, right? I think we're, we've,

02:06:08.340 we've, we've exhausted our couple more questions. Who would do 10? The short ones.

02:06:11.900 Oh my God. I can't do any more, dude. Oh, one more question. One question.

02:06:18.100 Oh, okay. Let's see. Can make this one short. So Bob and Peter, you two are awesome and I'm

02:06:26.340 learning. So I'm learning a lot from the AMA. So thanks. Tell us more about the latest and best

02:06:31.620 on ApoE4. Oh, cause you can, you can punt this. Oh, that's a pure punt baby. Tomorrow evening,

02:06:39.000 same bat time, different bat channel. Cause we're not going to record it. Unfortunately,

02:06:43.520 I'm going to sit right at this table and interview my really good friend, Richard Isaacson,

02:06:48.880 DJ rush. That's right. Is he going to spin anything? Is he bringing the techniques 1200s?

02:06:53.120 I don't know. I've met him a few times and every time I met him, he was, he's, he's spinning.

02:06:58.280 Yeah. So, so Richard Isaacson is a neurologist at Cornell, just a few feet from here. And he runs

02:07:05.640 the largest preventative clinic for Alzheimer's disease or Alzheimer's prevention clinic in the

02:07:09.480 country. We have collaborated for several years now, and we are going to have a deep dive nerd on

02:07:16.280 about all things pertaining to Alzheimer's inclusive of ApoE. So I suspect this podcast will be released

02:07:23.560 before that podcast, but please do look for it. Richard Isaacson. Cool. So on that note,

02:07:29.040 is there anything else you'd like to say, Bob, before we disband? I have nothing to add.

02:07:34.340 Nick, do you have anything to add behind the camera laughing at us? Very well. Thanks everyone. I hope

02:07:40.840 people think this was fun and worthwhile and we'll continue our, uh, up vote, down vote,

02:07:45.800 to take a question in AMA style. Bye-bye. You can find all of this information and more at

02:07:54.020 peteratiamd.com forward slash podcast. There you'll find the show notes, readings, and links related to

02:08:00.140 this episode. You can also find my blog and the nerd safari at peteratiamd.com. What's a nerd safari,

02:08:06.760 you ask? Just click on the link at the top of the site to learn more. Maybe the simplest thing to do

02:08:11.220 is to sign up for my subjectively non-lame once a week email, where I'll update you on what I've

02:08:15.820 been up to, the most interesting papers I've read, and all things related to longevity, science,

02:08:20.840 performance, sleep, et cetera. On social, you can find me on Twitter, Instagram, and Facebook,

02:08:26.000 all with the ID peteratiamd. But usually Twitter is the best way to reach me to share your questions

02:08:30.980 and comments. Now for the obligatory disclaimer, this podcast is for general informational purposes

02:08:35.940 only and does not constitute the practice of medicine, nursing, or other professional healthcare

02:08:40.440 services, including the giving of medical advice. And note, no doctor-patient relationship

02:08:45.700 is formed. The use of this information and the materials linked to the podcast is at the user's

02:08:51.060 own risk. The content of this podcast is not intended to be a substitute for professional

02:08:55.220 medical advice, diagnoses, or treatment. Users should not disregard or delay in obtaining medical

02:09:00.720 advice for any medical condition they have and should seek the assistance of their healthcare

02:09:04.880 professionals for any such conditions. Lastly, and perhaps most importantly, I take

02:09:10.100 conflicts of interest very seriously. For all of my disclosures, the companies I invest

02:09:14.540 in and or advise, please visit peteratiamd.com forward slash about.

The Peter Attia Drive - October 29, 2018

#26 - AMA #3： supplements, women's health, patient care, and more

Episode Stats

Summary

Transcript