The Peter Attia Drive - #47 – Matthew Walker, Ph.D., on sleep – Part I of III： Dangers of poor sleep, Alzheimer's risk, mental health, memory consolidation, and more.

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

00:00:10.140 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.

00:00:41.440 Hey everybody, welcome to this week's episode of the drive. I'd like to take a couple of minutes

00:00:46.000 to talk about why we don't run ads on this podcast and why instead we've chosen to rely entirely on

00:00:51.660 listener support. If you're listening to this, you probably already know, but the two things I care

00:00:56.480 most about professionally are how to live longer and how to live better. I have a complete fascination

00:01:02.760 and obsession with this topic. I practice it professionally and I've seen firsthand how

00:01:07.480 access to information is basically all people need to make better decisions and improve the

00:01:12.120 quality of their lives. Curating and sharing this knowledge is not easy. And even before starting the

00:01:17.600 podcast, that became clear to me. The sheer volume of material published in this space is overwhelming.

00:01:22.820 I'm fortunate to have a great team that helps me continue learning and sharing this information

00:01:27.880 with you. To take one example, our show notes are in a league of their own. In fact, we now have a

00:01:33.840 full-time person that is dedicated to producing those and the feedback has mirrored this. So all of this

00:01:39.720 raises a natural question. How will we continue to fund the work necessary to support this? As you

00:01:45.980 probably know, the tried and true way to do this is to sell ads. But after a lot of contemplation,

00:01:51.100 that model just doesn't feel right to me for a few reasons. Now the first and most important of

00:01:56.300 these is trust. I'm not sure how you could trust me if I'm telling you about something when you know

00:02:02.100 I'm being paid by the company that makes it to tell you about it. Another reason selling ads doesn't

00:02:07.580 feel right to me is because I just know myself. I have a really hard time advocating for something

00:02:12.840 that I'm not absolutely nuts for. So if I don't feel that way about something, I don't know how I can

00:02:18.520 talk about it enthusiastically. So instead of selling ads, I've chosen to do what a handful

00:02:23.280 of others have proved can work over time. And that is to create a subscriber support model for my

00:02:29.320 audience. This keeps my relationship with you both simple and honest. If you value what I'm doing,

00:02:36.260 you can become a member and support us at whatever level works for you. In exchange, you'll get the

00:02:41.840 benefits above and beyond what's available for free. It's that simple. It's my goal to ensure that no

00:02:47.860 matter what level you choose to support us at, you will get back more than you give.

00:02:52.860 So for example, members will receive full access to the exclusive show notes, including other things

00:03:00.380 that we plan to build upon, such as the downloadable transcripts for each episode. These are useful

00:03:06.760 beyond just the podcast, especially given the technical nature of many of our shows. Members also get

00:03:12.900 exclusive access to listen to and participate in the regular ask me anything episodes. That means

00:03:19.860 asking questions directly into the AMA portal and also getting to hear these podcasts when they come

00:03:25.800 out. Lastly, and this is something I'm really excited about. I want my supporters to get the best deals

00:03:31.140 possible on the products that I love. And as I said, we're not taking ad dollars from anyone, but instead

00:03:36.280 what I'd like to do is work with companies who make the products that I already love and would already

00:03:40.900 talk about for free and have them pass savings on to you. Again, the podcast will remain free to all,

00:03:48.600 but my hope is that many of you will find enough value in one, the podcast itself and two, the

00:03:56.440 additional content exclusive for members to support us at a level that makes sense for you. I want to

00:04:01.960 thank you for taking a moment to listen to this. If you learn from and find value in the content I produce,

00:04:07.240 please consider supporting us directly by signing up for a monthly subscription.

00:04:11.940 Welcome to the kickoff of this Matthew Walker three-part series on sleep. This is one of the

00:04:16.460 topics we are asked about the most, and I think this will be a fan favorite. This series of podcasts

00:04:21.000 will be released over three weeks with this one being the first. Now at the end of the series,

00:04:25.340 we're going to take questions for a follow-up AMA with Matthew. Asking questions on the AMA forum

00:04:30.940 and listening to the AMA podcast with Matthew along with all AMAs will only be available to subscribers.

00:04:35.960 So if you haven't signed up yet, you can do so now at peteratiamd.com forward slash subscribe.

00:04:41.620 My guest this week is Matthew Walker, professor of neuroscience and psychology at the University

00:04:45.840 of California, Berkeley, and the founder and director of the Center for Human Sleep Science.

00:04:50.000 Matthew earned his undergraduate degree and PhD in neurophysiology in London, and subsequently became

00:04:55.440 a professor of psychiatry at Harvard Medical School before moving to Berkeley. His research examines

00:04:59.740 the impact of sleep on human brain function in healthy and diseased populations. To date,

00:05:04.800 he has published over a hundred scientific studies. He has received numerous funding awards from the

00:05:09.760 National Science Foundation, National Institutes of Health, and he's a fellow of the National Academy

00:05:14.360 of Sciences. He's the author of the international bestseller, Why We Sleep, which also happens to be

00:05:19.380 the favorite book of my not-yet-two-year-old son. He holds many patents covering various consumer-based

00:05:25.980 sleep recordings, sleep tracking, and sleep simulation. He's a sleep scientist at Google, where he helps

00:05:31.380 the scientific exploration of sleep in health and disease. He is also an enormous fan of Formula One

00:05:36.840 and my hero, Ayrton Senna. And in the third part of this installment, we actually spend quite a bit of

00:05:44.360 time discussing this. Here in the first part of this series, we discuss the relationship between sleep,

00:05:49.480 dementia, and Alzheimer's disease, the four pillars of sleep, sleep stages, measurements, and the overall

00:05:55.300 importance of sleep, the lack of sleep and dangers of drowsy driving, and how the conversation of sleep

00:06:01.460 has changed over the past several years. So without further delay, here is the first installment of my

00:06:06.740 three-part series with Professor Matthew Walker.

00:06:11.600 Hey Matt, thank you so much for making time to meet on a Sunday.

00:06:15.280 Absolute privilege and a delight to be with you, Peter.

00:06:17.980 What's with the weather in Berkeley? Is it always this bad?

00:06:19.780 Now, when Peter says, is it always this bad? It's probably in the low 60s, and there isn't much

00:06:27.160 blue sky. It's dry, and there's a very pleasant breeze, and there is some wonderful blossom.

00:06:33.080 So I know it's not San Diego, where everything is utopian, and people are singing Kumbaya at the

00:06:39.100 end of the day and holding hands, but it's a pretty good day when you consider it still February.

00:06:43.520 Right, right. I know. It's like I live in these extremes. Like, you know, in San Diego,

00:06:46.680 the bar is like, it has to be perfect. And then in New York, you just sort of expect it to be sort

00:06:50.680 of, you know, like I was there two days ago, and it stunk.

00:06:53.280 I mean, all three of those are better than the United Kingdom. You know, my home country,

00:06:57.000 where essentially you have nine months of really bad weather, and then three months of winter.

00:07:02.100 And then that's your year. So this day, February, like this low 60s, I'm a happy man.

00:07:08.860 I came from Canada, where it's bad in a different way. But the first time I saw California

00:07:13.380 was my interview for med school. And it was a February day, and it was 76 degrees. And that

00:07:20.820 was the first time it occurred to me that you didn't have to suffer. You know, like I just,

00:07:25.200 I never realized, I mean, I guess intellectually, I understood that it could be warm in the winter,

00:07:29.440 but it never occurred to me that you could choose to do that. So well, now speaking of this, so

00:07:34.120 what brought you to the US?

00:07:35.600 It was science. I did all of my education back in the United Kingdom. I actually started off at

00:07:41.940 medical school. So back in the UK, at age 18, you can actually go to medical school. It's a five-year

00:07:46.980 program. And after about two years, this wonderful professor came up and said, look, you're a nice

00:07:55.060 guy, but you're always asking questions. And you're asking them, I think, genuinely, you're not asking

00:07:59.320 them to be an arse. But I think you're actually a scientist, not a doctor. Because at the time, 0.97

00:08:06.500 you know, he was thinking, doctors are interested in answers. Scientists are interested in questions.

00:08:12.820 I think you're a scientist, not a doctor. So we didn't have MD, PhD programs back at that time.

00:08:18.960 So then, you know, I thought about it for the summer, came back, switched to neuroscience.

00:08:23.460 So essentially brain science, did that as an undergraduate, then went on to do my PhD at the

00:08:28.460 Medical Research Council in London, and then came to Harvard for my first faculty position there.

00:08:35.120 And I was there for seven years as faculty in psychiatry at Harvard. And the winters just got

00:08:43.700 to me. And, you know, the environment at times as well, academic was perhaps more combative and

00:08:51.720 competitive. And I think that's one way to breed productivity in a department, and I'm not shaking a

00:08:57.100 finger at it. But by nature, I'm just a person who is much more collaborative and collegial rather

00:09:02.380 than combative and competitive. So Berkeley fit my mentality better, came out here, weather was

00:09:09.260 great. California, it's an immensely easy place to arrive and a very difficult place to leave. And I

00:09:15.220 think I probably won't.

00:09:17.240 I know that feeling. I'm probably leaving California. I'm going through the mourning process of accepting the

00:09:22.740 departure because you're right. You'll be back. It's a hard place to leave. When did the switch

00:09:27.440 sort of flip in terms of your passion for sleep? I think back at medical school, I was always

00:09:32.500 interested in states of consciousness. I guess you would describe it as I was fascinated by anesthesia,

00:09:39.020 was fascinated by even hypnosis. I remember I actually went out to meet this hypnotist to try and

00:09:46.840 understand not, you know, the act and the show. I wasn't interested in that. I fundamentally wanted to

00:09:51.580 understand what did he think happened within the brain to produce this altered state of consciousness.

00:09:58.640 So I got really interested in how the brain can switch between sort of states, mental states,

00:10:05.280 and then just fundamental awareness states. That then naturally led me to just become fascinated by

00:10:12.560 sleep. I mean, there is no better demonstration of a gigantic shift in conscious state that happens to

00:10:19.520 almost every single living creature on this planet every 24 hours. And back then, this was 20 years ago,

00:10:29.140 we had this fundamental question, why do we sleep? And what was staggering to me is that

00:10:36.300 it took a third of our lives. Every species that we've studied to date does it, sleeps. It seems to have

00:10:44.260 evolved with life itself on this planet. Even more, we've understood the functions of the other main

00:10:50.060 biological drives, eating, drinking, and procreating for tens, if not hundreds of years. But the fourth

00:10:59.900 main biological drive, the drive to sleep, had continued to elude scientific judgment. How was that

00:11:06.780 possible? And back then, 20 years ago, if someone were to ask you, you know, why do we sleep? The crass answer

00:11:13.020 was we sleep to cure sleepiness. And that's the fatuous equivalent of saying, well, we eat to cure

00:11:19.780 hunger. Well, that tells you nothing about the fundamental nutritional benefits of macro nutrition.

00:11:27.740 But that's where we were at with sleep. So it was, for me, this perfect collision of a fascination in an

00:11:35.140 innate biological problem and universal behavior conserved across evolution, together with the fact

00:11:43.440 that we did it for a third of our lives, plus the fact that science had not been able to crack this

00:11:50.520 nut. It was one of the last great remaining scientific mysteries. I thought, well, you know,

00:11:56.960 if you want to pick something that's going to sustain you for the rest of your life, and some of the most

00:12:00.800 acerbic, brilliant minds in science had tried to crack this nut and failed, including Francis Crick,

00:12:07.300 wonderful scientist who discovered the helical structure of DNA, it's unlikely I'm going to do it

00:12:12.320 in my lifetime, but at least it will sustain me as a problem. And I could not let it go. And so

00:12:18.100 that was where the intellectual bite happened. The data, the empirical scientific bite happened when

00:12:24.340 I was doing my PhD. I was studying brainwave patterns in people with dementia. And I was trying to

00:12:29.600 do differential diagnosis very early on in the course of cognitive decline, seeing which type

00:12:35.380 of dementia did they have. Was it frontal lobe dementia? Was it Lewy body dementia? Was it

00:12:39.660 Alzheimer's dementia? And I was failing miserably for two years. I was getting nothing. I used to go

00:12:45.980 home. I was living in this doctor's residence and I had this sort of, in the middle of my room,

00:12:49.780 this little igloo of journals that used to sit in on a Sunday, just this kind of like sort of

00:12:55.060 sanctified circle of knowledge. And I was reading one Sunday and I realized that the different

00:13:01.660 pathologies in these different dementias, some would hit the sleep centers and others would leave

00:13:07.440 those sleep centers completely untouched. Even within forms of dementia that clinically presented

00:13:13.340 very similarly. That's right. Yeah. You could see certain brainstem nuclei, certain centers within

00:13:19.860 the basic base of the brain. Some of those were getting eroded by the dementia and those are sleep

00:13:27.380 generating centers. Whilst in other of those pathologies, those centers were spurred until very

00:13:32.620 late in the disorder. So they probably weren't affecting my early patients. So I realized I was

00:13:38.520 measuring their brainwave activity at the wrong time. I was measuring my patients whilst they were awake.

00:13:43.440 What I should have been doing is measuring them whilst they were asleep. Got a small grant,

00:13:47.760 set up a sleep lab, learned how to do it, got some great results. And then the question became,

00:13:53.140 if their sleep is so disrupted, is it not just a symptom of dementia? Could sleep disruption actually

00:14:00.460 be a biomarker of dementia? Could it actually foretell when you are going to develop dementia and which type?

00:14:09.120 Or even more profound, is sleep disruption an underlying cause of Alzheimer's disease? And I think right now,

00:14:17.220 based on the weight of the data that we have, and we and others have got large research programs doing

00:14:21.960 this work, the evidence I think is causal. I think that sleep at this stage may be one of the most

00:14:28.740 significant lifestyle factors, at least, that determines your risk ratio for Alzheimer's disease.

00:14:34.680 I feel the causal evidence for that now in humans and animals is strong enough to make that statement.

00:14:40.700 And I don't make that statement lightly.

00:14:42.200 Yeah, that is profound because when you think about the other really big ones, vascular disease

00:14:47.980 and insulin resistance, they both feed into a very similar mechanism, which is neuronal energy

00:14:53.560 deprivation. That's right.

00:14:55.500 So do you think that sleep disruption or poor sleep, and I want to come back to the semantics

00:15:01.820 around that because I don't actually think I'm well equipped to talk about this intelligently,

00:15:06.500 but do you think that that also disrupts some aspect of energy metabolism in the brain?

00:15:13.140 Or where in the sort of chain of events do you think that that impacts?

00:15:18.360 I think one of the places that it impacts is in oxidative stress. I think it's safe to say that

00:15:26.520 wakefulness is low-level brain damage. And we need sleep to avert some of that low-level brain damage,

00:15:34.860 that wakefulness. Sleep is the price that we pay for wakefulness, essentially. And something I

00:15:41.000 mentioned in a book that I put out was, and I got a lot of pushback.

00:15:43.920 Just a little book. Just a little New York Times bestseller, yeah. Which, by the way,

00:15:48.540 before we're done, I brought my copy because my youngest son, who's named after our mutual

00:15:54.020 favorite race car driver, loves moons. It's all he thinks about. Moons, moons. His pajamas,

00:16:00.160 moons, everything. And he loves your book, except for the fact that on the inside there are no pictures

00:16:05.460 of moons. So he'll sit there and he'll look at the cover and he'll go, moon, moon, moon. And then

00:16:10.680 he flips through it and he gets so frustrated that there's no moon pictures inside.

00:16:14.400 I will course correct that if there's a second edition, and I will just try to place the words

00:16:19.260 underneath that it is for the benefit of Ayrton.

00:16:23.540 But in the book, I proposed that we didn't evolve sleep. That perhaps we've got it the wrong way

00:16:32.940 around. Perhaps the default state of life on the planet was sleep. And it was from sleep that

00:16:41.060 wakefulness emerged. Why could it not have been that way? Now, I don't have any good evidence to

00:16:47.920 support my hypothesis. I think it could be utter nonsense, but I think it's a tenable hypothesis.

00:16:54.940 It's just one that's difficult to test. But to come back to your question about Alzheimer's,

00:16:58.540 which I think is critical, I think the first thing is that we see that with insufficient sleep,

00:17:03.760 you get increases in oxidative stress. Those oxidative stress processes lead to a whole cascade

00:17:11.460 of kind of a finger flick domino effect of things that lead to neuronal death. One of the areas

00:17:17.920 in the brain that is most sensitive to that is an area that we call the hippocampus, which is a

00:17:23.300 critical memory center. It's probably one of the first structures to undergo damage in Alzheimer's

00:17:28.700 disease. It's part of the reason why the sort of the phenotype of Alzheimer's is forgetfulness,

00:17:33.540 problems with memory. I think perhaps the stronger evidence though, in terms of sleep and

00:17:38.880 Alzheimer's disease risk is a remarkable discovery that was made probably about five or six years ago now

00:17:44.720 by some folks at Rochester University in rats. And what they discovered is that the brain actually

00:17:52.140 has a sewage system inside of it. Now your body has one that you're all familiar with called the

00:17:58.260 lymphatic system, but it turns out the brain has one and it's called the glymphatic system named after

00:18:04.900 the cells in the brain that form this system called the glial cells, also known by sort of from a Greek

00:18:10.820 derivative, meaning glue. So they used to be just thought as the sort of the cells.

00:18:15.320 The sort of irrelevant stuff between the neurons.

00:18:17.500 Exactly. You know, yeah, it was like junk DNA. And of course, what we always learn is that mother 0.79

00:18:22.740 nature is far too efficient to leave inefficiency on the table, like glue or junk. And it turns out they

00:18:31.320 form this sanitization system within the brain. And what happens, they discovered, is that when you go into

00:18:38.200 deep sleep at night, this sewage system kicks into high gear and essentially those glial cells, 0.95

00:18:46.380 which surround the brain cells themselves, the neurons proper, they shrink in size by up to 200%.

00:18:53.460 And then all of a sudden it leaves a vast amount of room for cerebrospinal fluid to start perfusing the

00:19:00.400 brain and washing out the metabolic detritus of wakefulness. So it would be like New York

00:19:08.020 City at night, all of the buildings shrunk down to 200%. You know, they became miniaturized. And then

00:19:15.060 there was this big effluent flush that just happened across Manhattan to clear out all the debris. And

00:19:21.620 it's essentially good night, sleep clean, that you get this power cleanse at night. Why is this related

00:19:27.680 to Alzheimer's? Well, one of the critical ingredients that they found that the glymphatic system washed away

00:19:33.520 was a sticky toxic protein called beta amyloid. Beta amyloid is probably one of the two core proteins

00:19:40.240 that we know underlie your risk for the development of Alzheimer's. Ago, if you're not getting your sleep

00:19:47.460 at night, you're not getting that washing away of the toxic Alzheimer's protein. Every night then,

00:19:53.560 you are building up more amyloid within the brain. If you keep doing that night after night,

00:19:58.620 it's like compounding interest on a loan. It's just escalating your Alzheimer's risk.

00:20:04.760 That's why we can then explain the associational evidence, which I don't like associational evidence

00:20:10.300 in epidemiological studies. And I don't think you're a huge fan of it, perhaps either.

00:20:14.760 Just don't tell anybody at the Harvard School of Public Health.

00:20:17.480 Yeah. Or maybe the NIH when I'm applying for grants for epidemiological studies. But what we certainly see

00:20:22.940 is that if you bucket people into the amount of sleep on average that they've got across the lifespan,

00:20:27.760 and at the end of the sort of life or in late life, in their 70s and their 80s, we do a special type

00:20:33.860 of PET scan to map the amount of amyloid in their brain. And then you just sort of look at those brain

00:20:40.560 scans. Anyone on the street would not need any statistics. They would not need any training in

00:20:46.260 brain science to see that one of these maps is different to the other. It's like Sesame Street.

00:20:51.300 And what is remarkable is people who are getting, you know, seven hours of sleep or less compared to

00:20:57.680 seven hours of sleep or more. There is marked differences in the amount of amyloid that's

00:21:03.040 built up. Now that's just associational evidence. But then when you take it down to the level of

00:21:07.680 animal studies, where they selectively deprive animals of either deep non-REM sleep, or they

00:21:13.260 fragment their sleep, you get an immediate amyloid buildup centrally within the brain.

00:21:19.260 You start to have what looks like a sort of, you know, an amyloid sort of genic process that's 1.00

00:21:25.880 happening. That's rats. So then we can all wave our hands if we want, because we would like to beat

00:21:31.780 our chests and sort of, you know, say, I don't need my sleep. And that's just rat stuff. Well, 0.82

00:21:36.220 the studies have now been done in humans, where we will take a human being, and for one single night,

00:21:41.560 we will take away their deep sleep. And the way that you do that is that you just play these

00:21:45.880 auditory tones at night. So they never wake up. So the total amount of sleep that they get is still

00:21:52.480 eight hours, but I can just selectively excise your deep sleep when that sewage system should be

00:21:59.300 kicking into gear. And then the next day we do something that's really rather unpleasant. We do a

00:22:04.340 spinal cord puncture and we suck out some cerebrospinal fluid, and we can measure the amount of the two

00:22:10.520 toxic proteins linked to Alzheimer's. One is called beta amyloid that we've discussed. The other is

00:22:15.080 called tau protein. And we see the next day after one night of sleep disruption, a significant increase

00:22:23.640 in circulating levels of amyloid and tau. For me, that was kind of like the turning point where I

00:22:29.540 finally felt comfortable going on record and saying, at this point, folks, I really feel comfortable

00:22:36.580 saying that your sleep is a critical component of your prevention of Alzheimer's disease. I really

00:22:45.560 think the evidence is now, you know, in that direction and way in that direction. It's not

00:22:51.180 just epidemiological. It's not just prospective longitudinal studies, which are better than

00:22:55.720 cross-sectional sort of epidemiological. It's not just causal in rats. It's experimentally causal

00:23:02.600 in humans after one night of sleep. Do you see variation there? Because one of the things that I

00:23:08.080 find difficult with understanding all diseases and Alzheimer's would be sort of more so than even

00:23:15.640 cardiovascular disease, because we still have so much less data, is there must clearly be enough

00:23:22.340 variation between people that it's very easy at the heterogeneous level to miss things that you would

00:23:30.100 see in an otherwise homogeneous population. So for example, when you look at the coinage of the

00:23:34.300 term Alzheimer's disease, it's quite likely that that was coined based on a patient who probably

00:23:39.740 had a PSN1 or 2 mutation. It's such an early onset of the disease versus the disease we typically see.

00:23:46.480 I mean, today, those patients make up less than 1%. And it's almost a different disease from the,

00:23:52.100 you know, these are people that are presenting in their 50s, and they're usually dead before they're 60.

00:23:56.000 Whereas 99% of the population are getting a sort of later onset disease. And even, of course,

00:24:02.080 then within that population, all of the variation with APOE genotype, it almost makes you wonder if

00:24:06.720 there's different susceptibilities to this deprivation, because everybody and their brother can point to

00:24:11.980 the exceptions to these rules, right? The person who did everything right, slept like a baby every

00:24:16.400 night, and ends up down this one path. And then, of course, the person who does everything incorrect,

00:24:20.640 and seems to be, you know, completely fine, at least from the perspective of, you know, no obvious

00:24:25.740 disease. So your analogy is really interesting, because I don't really find myself hugely in the

00:24:33.460 amyloid beta camp as the main causal driver. I think it's more of a result. But it's interesting,

00:24:40.000 what you posit is, if it's more about the clearance of that than the generation of that,

00:24:44.960 that would be a congruent view with this idea that, you know, there could be other things on

00:24:50.160 the back end that are driving this primarily. But of course, I think with Alzheimer's disease,

00:24:55.320 it strikes me as just so complicated, because of the genetic predisposition, coupled with the

00:25:02.800 multiple pathways that can seem to drive comparable insults to an otherwise incredibly sensitive organ.

00:25:09.200 I hugely cleave to the idea that probably what we call Alzheimer's right now is going to be subtyped

00:25:16.320 at some point in the future that we'll have. Yeah, imagine it's like cancer, right? Where in

00:25:21.020 30 years, hopefully, the rate of growth has slowed, that it's not, you know, a pandemic. But yeah,

00:25:26.760 you might have five subtypes or something for that. That's right. You know, and I think we will see

00:25:31.020 that, I think, and then we will start to pattern match what features. I think the fact that sleep

00:25:36.940 already, let's just call Alzheimer's disease, this strange heterogeneous sort of mix right now that

00:25:43.120 we feel uncomfortable about in terms of pinning one exact pathological ingredient as the causal

00:25:50.100 trigger for that. In fact, I actually probably think tau is, if you look at the data, I think the data

00:25:55.240 are better in terms of predicting cognitive decline, at least than amyloid is. But the fact that, you know,

00:26:01.500 we've now found that insufficient sleep will also lead to tau aggregation too, I think is concerning

00:26:07.740 and worrying. But let's accept that this Alzheimer's thing, this disease state is heterogeneous. The fact

00:26:15.080 that sleep, and we can find such a strong predictive sleep signal in terms of that risk, despite it being

00:26:22.220 such a diffuse gray matter ball itself of a disease, must mean that the links between insufficient

00:26:29.620 sleep and this heterogeneous thing are, I think, probably quite strong. Otherwise, the dirtiness

00:26:35.940 of the equation would be such that those associations would be far weaker rather than stronger. The fact

00:26:42.360 that they're already present leads me to believe that when we do eventually subtype Alzheimer's disease,

00:26:47.580 we will then find that there are subsets of sleep features that are deterministic of those subsets

00:26:54.600 of Alzheimer's. So for example, deep sleep will be predictive of one form of Alzheimer's. Sleep

00:27:00.700 fragmentation, which is more about sleep quality than it is about physiological quantity. You know,

00:27:08.360 sleep fragmentation will be predictive of something else, another sort of subtype of Alzheimer's

00:27:14.600 disease. Then I think, you know, basic- Duration could matter.

00:27:18.920 I think there probably are four pillars of sleep. There is probably regularity. How consistent is

00:27:25.620 your sleep schedule? Are you undergoing what we call social jet lag, which is where you're bouncing

00:27:30.480 around? You're going to sleep at, you know, 1 a.m., then 10 p.m. and all over the map. So there's sort

00:27:36.540 of regularity. There is the continuity of your sleep. Is it fragmented? Are you waking up many more

00:27:43.540 times or is it just nice one long shot? That's a second feature. I think the third feature is

00:27:49.380 quantity. How much sleep are you getting and how much of the different stages are you getting?

00:27:54.960 And then independent of how regular your sleep is, how continuous your sleep is, and the amount,

00:28:01.380 the quantity, the final is quality. And quality has become big in the last five years in the sleep

00:28:07.180 field. In other words, what is the electrical signature of your sleep? And just because you're

00:28:14.420 getting eight hours of sleep doesn't mean that you're getting eight hours of good electrical quality

00:28:18.720 deep sleep. And, you know, I hope that we get the chance to speak about sort of things like caffeine

00:28:23.020 and alcohol, other substances of abuse that can make you feel like you're sleeping eight hours. But

00:28:30.200 so quantity, just fine. Quality, poor. So to come back to your question, I love the way that you

00:28:36.740 think about that problem. And I agree with it very much that Alzheimer's disease is probably going to

00:28:41.180 be a heterogeneous. I think the pathology already points to that and its complexity. The way that

00:28:46.780 sleep plugs into that heterogeneous disorder that we call Alzheimer's is also going to be heterogeneous

00:28:53.240 in nature. Different sleep features will be predictive of different aspects of the amyloid pathology cascade.

00:29:00.460 And we just submitted a paper, which I'm going to blow now in terms of media sort of outlet,

00:29:06.240 which hasn't been accepted yet. But we just tried to ask the question across someone's lifespan.

00:29:14.140 If you bucket them into 10 year decades and you ask, what was the quantity and quality of their

00:29:20.360 sleep from their 30s and 40s, 40s and 50s, 50s and 60s, so on and so forth. And then at the end of

00:29:26.640 their life, you stick them in a brain scanner and you map how much amyloid is in the brain and how much

00:29:31.540 tau is in the brain, which is what we can do here at the sleep center.

00:29:34.340 And then you essentially can, in a retrospective manner, say, was sleep at certain times of life

00:29:42.240 more critical in predicting how much amyloid you will have late in life? So in other words,

00:29:50.120 are there certain decades during an adult's lifespan where sleep is even more important

00:29:57.460 for determining your... How can you do that retrospectively? So let's say you had this

00:30:02.900 cohort of patients. Maybe at best you could, through questionnaires, infer duration. But how

00:30:10.120 would you be able to infer anything about depth or quality?

00:30:13.320 So depth we can't do, but quality we can do. So we've got a set of standardized questionnaires

00:30:19.460 that have been essentially validated against gold standard, what we call polysomnography,

00:30:24.840 which essentially is sleep laboratory grade recording with electrodes. And what we find is

00:30:30.280 that those questionnaires have good predictive validity relative to objective gold standard measures.

00:30:37.520 So it's a hop, skip and a jump, but you can say, look, these subjective questionnaires seem to go

00:30:43.640 pretty well hand in hand with sleep quantity and quality. Let's take those questionnaires and then

00:30:50.140 let's ask of those patients what they were doing in terms of their sleep quantity and quality in those

00:30:56.540 different bucket decades of their life. And it has all of the problems that you can imagine.

00:31:01.880 Yeah, because this seems like a slightly better version of a food frequency questionnaire, which

00:31:06.460 your peers would be relying on to do the same question about nutrition.

00:31:10.220 So it's reliant on an individual, you know, it's got problems with history because, you know,

00:31:14.540 the further you go back, does that signal of accuracy degrade?

00:31:18.240 How big are the hazard ratios? Because I guess this is where one of those things where

00:31:21.780 if sleep could account for 25% of this disease, or if it increased the risk by say 25% would be a

00:31:31.860 unlikely that you could see that signal because that's not a big enough signal with all the

00:31:36.380 noise you've just described. But if you were talking about a hazard ratio of six or seven,

00:31:40.680 or, you know, it's so funny today, one of my kids asked me about smoking. Like, why do people smoke?

00:31:46.420 And this, I couldn't resist going down the Bradford Hill pathway. Unfortunately, he's four and a half,

00:31:52.340 but he got a probably about a 12 minute lecture on epidemiology and the hazard ratio of small cell

00:31:58.300 lung cancer and smoking. But the punchline of my story to him, Peter Attia, ladies and gentlemen,

00:32:03.600 Peter Attia. I was trying to explain to him what it would mean to have a 14 X increase in risk,

00:32:11.240 which was basically what was observed in the first physician study of smoking. And, you know,

00:32:16.560 the point there is that is such a massive signal that, I mean, as you describe it, I love whenever

00:32:23.300 you can get these things that statistics aren't necessary. When you don't need the P value to figure

00:32:27.760 it out, that's it. So does your intuition, and if, again, I don't want you to scoop what's already

00:32:32.240 in this paper, so if you can't speak about it, no problem. But do you have a sense of that magnitude

00:32:37.920 and therefore if even a crude backwards looking metric is directionally pointing you at something

00:32:45.340 that you think, wow, this isn't a subtle change? Yeah. I mean, we're up in the region of, you know,

00:32:51.040 from 0.45 to 0.6. So it's, you know, it's...

00:32:55.520 In risk reduction for those doing it correctly.

00:32:57.420 Correct. Yeah. So it's substantial. But what was interesting to me, which gave me more confidence,

00:33:02.260 was that the windows of sleep sensitivity were... Now, by the way, across the entire lifespan,

00:33:09.680 if you're getting less sleep at almost any point in the lifespan, the risk that you're going to have

00:33:14.860 higher amounts of these Alzheimer's disease proteins in your brain is greater. That's what

00:33:20.220 we find across the board. But what's interesting is that when it comes to those two different proteins,

00:33:25.980 amyloid and tau, there are different decades that are predictive. So if it's just one global

00:33:34.480 subjective bias in your questionnaire, that you shouldn't see that result. So there is a sensitivity

00:33:42.020 and specificity to these two different proteins that, to me, got me more excited that it transcended

00:33:49.200 some of the limitations that are inherent in this approach. And also the fact that for amyloids,

00:33:56.100 you get a very strong predictive signal, you know, 30s to 40s, and then again, sort of 60s to 70s.

00:34:03.480 And for tau, it's actually somewhere in sort of the... Starts in the 50s to 60s and then goes through

00:34:09.920 until the 80s in terms of 80 years old. So we actually, we didn't go back in terms of childhood

00:34:16.020 just because at that point, it's really difficult to feel confident about that data. You'd have to

00:34:20.560 get parents who are not living at that stage. But to me, I think the fact that you've got these

00:34:25.500 strange kind of hot spots across the lives... And you described a bimodal distribution 1.00

00:34:29.360 for amyloid beta. Exactly, for amyloid.

00:34:31.000 Which then lends me to believe that that result is difficult to explain by sort of just general

00:34:38.380 memory biases, which would be kind of more of a linear approach. But I don't... I mean,

00:34:42.680 I think the point is that, you know, the evidence that continues to build in favor of this story,

00:34:49.400 that insufficient sleep is a significant lifestyle factor in your deterministic ratio odds of developing

00:34:57.340 Alzheimer's disease, is ever stronger now. And the fact that unlike many of the other lifestyle

00:35:03.040 factors, the mechanism underlying it has now been demonstrated causally in humans and adults,

00:35:09.780 and it's been done so acutely within a 24-hour period of time, that to me just knocked me sideways.

00:35:17.900 And the controls did not have... So if you controlled for everything except the perturbation,

00:35:23.200 you didn't see the amyloid or tau rise in the controls. See, this is kind of interesting too,

00:35:27.360 because one of the reasons I've been a little less excited about amyloid... There's two reasons.

00:35:33.280 One is what can only be, I think, described as just the abject failure of the pharmacologic

00:35:38.040 approach to amyloid reduction. At our last counting... Oh, I don't want to be misquoted on this.

00:35:43.900 We'll include it in the show notes what the actual number is. But it's basically one amyloid

00:35:48.980 antibody quasi-success at disease stabilization to the tune of 2,000 failures. That's one thing.

00:35:56.120 The second is the autopsy studies of people who died without any clinical evidence of Alzheimer's

00:36:02.400 disease who still have amyloid, making me wonder if amyloid is necessary but not sufficient.

00:36:09.860 And therefore, what if there's something else going on? And of course, the third thing I would

00:36:14.160 now add, just based on thinking about it for two seconds, so this might not be well thought out, is

00:36:19.400 in the next couple of hours, we're going to talk about a whole bunch of other things that sleep does

00:36:24.740 that are in and of themselves completely devastating to your health that also would feed into that,

00:36:30.880 for example, the impact that negative sleep has on cardiovascular disease. And it's become quite

00:36:35.460 clear that if you're doing bad things to your heart, you're doing bad things to your brain.

00:36:38.940 And of course, we're going to talk about glucose disposal, hyperinsulinemia, and all of those things,

00:36:42.260 which also have a clear impact. So I think this would only strengthen the case that sleep deprivation

00:36:48.540 is bad for you and bad for your brain. It, to me, makes it harder to tease out potentially the

00:36:54.700 mechanistic links, although there's no reason, as you said, that it can't be multifaceted.

00:36:59.680 Yeah. And I think there are certainly those conundrums where you see patients with non-trivial

00:37:05.520 amounts of amyloid in the brain, let's say, at post-mortem, who are even only in the early stages of

00:37:11.480 MCI of what we call mild cognitive impairment. They haven't transitioned yet.

00:37:15.500 I'm glad you remember to explain these things because I'm nodding away. Like,

00:37:18.540 of course I know what MCI is.

00:37:20.340 But that aside, I think the idea that it's necessary, maybe not sufficient, but necessary

00:37:27.300 if you truly don't want necessary in your life. And if you're getting insufficient sleep,

00:37:34.660 you're placing a necessary there where it doesn't necessarily have to be.

00:37:39.180 No, the analogy could be with lipoproteins. You know, there's a...

00:37:42.560 Oh, that's a very... Yeah.

00:37:43.400 Right? I mean, lipoproteins are absolutely necessary, but not sufficient for atherosclerosis.

00:37:48.740 And so some people say, well, there's no efficacy in reducing lipoproteins. And I think part of it is

00:37:54.500 it's diffuse, right? You're only tackling one of the four constitutive elements. Four things have

00:37:59.860 to happen for you to get atherosclerosis, the lipoproteins being one. So that's actually an

00:38:03.240 interesting idea. And it's a great argument for why would you want something that's necessary,

00:38:07.760 but not sufficient there? Why would you decrease your odds?

00:38:10.460 Yeah. I mean, you know, think about it like this. If your goal, the two most feared diseases in

00:38:15.500 developed nations right now, I believe are cancer and Alzheimer's disease.

00:38:18.740 And I would say it's in the other order.

00:38:20.540 And yeah. And I think Alzheimer's is, you know, top of the list there. Do you want to de-risk

00:38:25.960 the probability of Alzheimer's disease in your life? If you truly don't want to invite,

00:38:31.680 write to, you know, this thing called Alzheimer's disease and invite it in, in the mail coming at

00:38:37.240 you, but you're getting, you know, less than seven hours of sleep a night, you're not working

00:38:43.640 in the direction that you wish. You're going against Alzheimer's disease prevention and you're

00:38:50.760 working well towards Alzheimer's invitation. People will often say, you know, how much sleep do you

00:38:56.460 get? And I'll say, look, I give myself a non-negotiable eight hour sleep opportunity.

00:39:01.000 And that's not because I'm trying to practice what I preach. I don't want to be some poster child

00:39:05.700 for sleep. If you understood what I knew about sleep and all cause mortality, as well as most disease

00:39:14.400 processes, you would realize that I'm being nothing short of utterly selfish in my preservation of an

00:39:22.360 eight hour sleep opportunity. I don't want to die young and I don't want sickness and disease in

00:39:27.600 my life. My family, for example, has got a very strong history of cardiovascular disease. And we

00:39:32.720 know that deep sleep is probably the best form of blood pressure medication that you could ever wish

00:39:37.860 for. That would be an easier one to study. I'm guessing we'll come to that later. Yeah. The ability to

00:39:43.480 perturb deep sleep and then see the change in blood pressure. The consequence. Yeah. I think the point is,

00:39:48.840 I think I've bashed people over the head with this sufficiently, but the link between a lack of

00:39:54.780 sleep and Alzheimer's now, I think is very strong. And what I find though positive about that is

00:39:59.880 with Alzheimer's disease, just as you mentioned, when it comes to pharmacological approaches,

00:40:03.920 we just don't have anything that looks robust right now. So medicine in many ways, and especially

00:40:11.160 for Alzheimer's disease has started to shift from a model of late stage treatment to early stage

00:40:16.660 prevention. And I think I feel very comfortable saying that one of the best preventative actions

00:40:23.960 that you can take on the basis of the science right now is start to capitalize on your sleep at night.

00:40:30.980 So let's shift gears just a little bit and go back to the polysom. Let's explain to people,

00:40:38.120 because we're going to eventually have to talk about tracking these things. And I want to

00:40:42.440 bring up some of the sort of commercially available ways that people do that, which

00:40:46.200 I think by definition are still not accurate enough for people to, you know, compare them to their

00:40:52.020 polysom. But if I came into your lab tonight and I was willing to go to sleep there and you would

00:40:58.520 hook me up to an EEG, what are the different patterns of brainwaves and how would you morphologically

00:41:04.640 describe them and then bucket them by these stages? You've already sort of touched on a little bit as

00:41:09.560 deep REM, deep REM, deep non-REM, et cetera.

00:41:14.420 Yeah. So upon falling asleep, human beings, and in fact, all mammals will experience two different

00:41:21.820 stages of sleep. One of them is called non-rapid eye movement sleep. The other is rapid eye movement

00:41:26.620 sleep. Non-REM sleep has been further subdivided into four separate stages, which are unimaginatively

00:41:33.700 called stages one through four. Thank God. Yeah, I know. Yeah. I think that's all that the sleep

00:41:39.160 deprived people who are, you know, experimenting could come up with at that time. So increasing in

00:41:44.780 their depth of sleep, so stages three and four of non-REM sleep are the really deep restorative

00:41:50.960 stages of sleep. Stages one and two are the lighter stages of sleep. And then on the other hand, we have

00:41:56.820 rapid eye movement sleep or REM sleep, which is named not after the popular Michael Stipe pop band,

00:42:02.640 but because of these bizarre horizontal shuttling eye movements that occur during this stage of sleep,

00:42:08.060 these rapid eye movements. So you have these two different types of sleep, and they will essentially,

00:42:15.220 as you fall asleep here in at the sleep center at Berkeley, they will go into essentially a battle

00:42:21.440 for brain domination throughout the night. And that cerebral war between non-REM and REM is going to be

00:42:28.600 won and lost every 90 minutes, and then replayed every 90 minutes to create what we call a standard

00:42:35.960 cycling architecture of sleep, or what we call a hypnogram of sleep. And so what you will see is

00:42:43.160 that upon falling asleep, you'll go into the light stages of non-REM, stages one and two.

00:42:46.840 Before we do that, let's say before I fall asleep, when I'm just laying in the bed,

00:42:50.440 how would you describe my brainwaves under the wakeful condition?

00:42:54.800 So at that point, what we typically see is that when you close your eyes,

00:42:58.500 the back of the brain goes into what's called an alpha rhythm. So when you're awake,

00:43:03.400 your brain shows this remarkably frenetic high frequency electrical activity. In other words,

00:43:10.040 your brain is going, in terms of its brainwaves, it's going up and down many, many times a second.

00:43:15.020 And the amplitude is very small, which kind of is paradoxically, you think if I'm awake,

00:43:23.500 then my brain is active. And so the size of those brainwaves should be big. It's not,

00:43:28.400 it's actually very small. And here's the reason. Different parts of your brain are doing different

00:43:34.180 things at different times. So those brainwaves sort of are all canceling each other out. So the

00:43:41.920 analogy would be, if I were to dangle a microphone above a sports stadium, what I'm picking up is the

00:43:48.840 signal from the crowd of 100,000 individual neurons that sit underneath it. That's how an EEG

00:43:56.040 electrode is. Now, before the game starts, that's wakefulness. And at that point, the 100,000 people

00:44:04.360 in the stadium, the 100,000 brain cells, they're all speaking to each other at different moments in

00:44:09.980 time. So there's a lot of cancellation. So there's a lot of cancellation. They're not all coordinated,

00:44:14.700 but it's very fast and frenetic. So I just get this signal that is very desynchronized. And it's

00:44:21.400 not synchronous. So they're not summing their power together. So the size of that brainwave is not very

00:44:27.740 big, but it's going up and down. But that's why the frequency is so high. The frequency is high, meaning

00:44:32.580 that you're sort of going up and down maybe 50, 60 times per second. Now, as you start to fall asleep,

00:44:40.620 or actually even before you start to fall asleep, as you're lying in bed awake, the back of the brain

00:44:46.220 is the first part of the brain that really starts to settle down. It goes into what we call alpha rhythm

00:44:50.880 when you close your eyes.

00:44:52.360 And I assume that's just because the occipital cortex is in the back and you've closed your eyes.

00:44:56.180 Brilliant. So the back of the brain, visual brain, if you close your eyes, that part of the brain

00:45:01.200 essentially stops processing the outside world. And it goes into its sort of default state when it's

00:45:07.600 awake of slowed frequency, it drops down from maybe 50 times per second down to about, let's say,

00:45:14.940 10 cycles per second.

00:45:16.900 But still relatively low amplitude.

00:45:18.620 But still relatively low amplitude. Then once you start to actually transition, when you make that

00:45:25.480 bridge from the world of wakefulness to sleep, as you sort of cross through that window,

00:45:31.200 from one of those worlds to the next, something bizarre happens. And if anyone out there is lucky

00:45:37.280 enough to be listening to this, and they have a bed partner, you can actually see this. Just look at

00:45:43.220 their eyelids as they're falling asleep. And what you'll see is that their eyeballs start to roll

00:45:49.580 in their sockets. And they're called slow rolling eye movements. Now, we don't know why the eyes roll in

00:45:56.780 the sockets like that. But it's the first sign that you're making the transition from the world of

00:46:02.340 wakefulness into sleep. We use it as almost an indicator of the transition into sleep. Now, by the way, if

00:46:08.300 your partner wakes up, sometimes people will wake up from that state, and they see you just looking over

00:46:13.660 them in bed, staring at their eyelids. If you're in early in a relationship, it's usually the end of the

00:46:20.320 relationship. That's just a point of note. Please don't blame me. Usually wait till you're married

00:46:24.540 and then start to do that one. The divorce usually is harder to come by. So then you start to go down

00:46:30.980 into the light stages of non-REM sleep, stages one and two. Then after about 20 or so minutes,

00:46:37.140 you'll go down deep into stages three and four of non-REM sleep.

00:46:40.760 Between one and two, have we transitioned out of that alpha wave? I mean, where do we start to see

00:46:46.500 the thetas? Theta already happening in stage one. In stage two, you actually go into theta activity,

00:46:54.840 which is now down from 50 cycles per second, maybe down to just sort of six or seven cycles per second.

00:47:02.680 And the amplitude's getting a bit bigger, isn't it?

00:47:04.760 It is getting a bit bigger, yeah. And then every now and again during stage two, the way that I know

00:47:09.400 that you're in this lighter stage of non-REM sleep stage two is that you'll get these synchronous

00:47:15.060 bursts of electrical activity that we call sleep spindles.

00:47:18.860 That's the K-spindle.

00:47:19.840 And that's sort of, yeah, where you get this, actually you get a nice big slow wave,

00:47:24.380 like almost like a single slow wave, and then you get this burst of a sleep spindle. And we actually

00:47:29.820 did this great project here at the Sleep Center where we did the sonification of sleep. And we extracted

00:47:35.860 brain waves from human beings. And then we sort of, you know, smooth them a little bit with some sort

00:47:42.040 of audio software, and then you can play them back. And it's beautiful because when you hear these sleep

00:47:48.880 spindles, they are these short synchronous bursts of electrical activity that last for about a second

00:47:54.620 and a half, and they go up to about 10 or 15 cycles per second. So it's almost like that beautiful

00:48:01.640 rolling R in sort of in the Hindi language, you know, sort of like a cat pairing. So you get these

00:48:08.400 big slow waves, and then you get this ripple of a spindle. And when you hear it on the speakers,

00:48:13.840 it's just sensational. You get this, you know, every time I hear it, I mean, this is spine tingling

00:48:26.900 stuff when you realize this is going on right now, physiologically in someone's brain. And then from

00:48:33.880 there, you start to go down to the really deep stages of non-realm sleep, stages three and four.

00:48:38.520 And at that point, the brain goes into this incredible synchronous mantra chant of these

00:48:46.000 slow waves. So the brainwaves slow down, you go down to maybe just one or two cycles per second,

00:48:53.840 very, very slow brainwave activity. But the size of those waves crashing on the beach of the cortex,

00:49:00.340 as it were, they are huge. And the reason is this, go back to the sports stadium analogy.

00:49:05.760 Let's say that we're here at Berkeley, and we're playing Stanford, Arch Rivals. 0.99

00:49:10.180 Yeah, which hopefully you're going to kick their butt. 0.99

00:49:11.960 Fingers crossed. 0.99

00:49:12.680 Just make sure there's no band anywhere to be found.

00:49:16.600 It always does us a terrible deserve. But at that point...

00:49:19.960 They're chanting in unison.

00:49:20.960 They're all chanting in unison.

00:49:22.020 So it's louder, but they're now making the same sound together. And you can actually hear what's being said.

00:49:29.840 Exactly. So now in that single microphone, you can hear Stanford sucks. Stanford sucks. So all of a sudden, 0.99

00:49:38.520 all of, for reasons that we are only now starting to understand, the brain coordinates hundreds of

00:49:46.760 thousands of neurons, unlike it does at any moment elsewhere in the 24-hour period. Hundreds of thousands

00:49:53.980 of brain cells all decide to join hands, as it were, physiologically, metaphorically. And they all

00:50:00.580 fire together, and then they all go silent. They all fire together, and then they all go silent.

00:50:06.160 And what I find amazing about that, by the way, is how slow that frequency is. You're basically,

00:50:10.920 you could be, this could be only happening once a second, right?

00:50:14.300 That's right. And that's why it got a terrible rap for 20 or 30 years, because it actually looked

00:50:20.280 not dissimilar to some aspects of coma. That's how slow those brainwaves were. So sleep scientists

00:50:26.620 were, you know, understandably fooled into thinking that deep sleep was a time when the brain essentially

00:50:32.560 was just doing nothing. Now we realize the exact opposite is true, because it's during that deep sleep

00:50:41.220 that you get essentially information transfer within the brain. It's a file transfer mechanism.

00:50:47.100 And the way to think about this is, let's say it's like long-wave radio. If you're in the city,

00:50:53.400 you tune into FM, which is short-wave radio, you can pick up a bunch of signals. Then the further

00:50:57.500 you drive out into the desert, let's say here in California, you just lose those channels, because

00:51:01.700 the range from the tower is so short that you just lose that range. But if you go onto sort of,

00:51:08.860 you know, the long-wave radio stations, you can still pick up stations for hundreds of miles. Why?

00:51:13.660 Because the carrier frequency of those radio waves is much slower. And so the distance over which you

00:51:22.480 can transfer information is much further. Deep sleep is a brain state of long-distance information

00:51:31.200 transfer. Is there morphologically a difference that you can see between stages three and four,

00:51:37.500 which are obviously both delta waves, but how do you distinguish those two?

00:51:42.140 Technically, no. The way it's defined is that what proportion of a 30-second period of time,

00:51:48.860 and that's the way that we score sleep. I'll have you sleep for eight hours, then I'll chop up that

00:51:53.660 eight hours into 30-second bits of information. And for each 30 seconds, you'll get a sleep score.

00:51:59.520 And then we add them all up. And what differentiates stage three of deep non-REM sleep from the very

00:52:06.520 deepest stage, which is stage four non-REM sleep, is simply what proportion of that 30 seconds is

00:52:14.200 consumed by that deep, slow brainwave activity. If it's sort of less than 50%, then it's stage three.

00:52:22.160 If it's more than 50%, stage four. So morphologically, no. Although if you really look at it morphologically,

00:52:31.800 on average, yes. The waves are typically larger in amplitude and slower in frequency in stage four

00:52:40.960 than they are in stage three.

00:52:42.520 And do people always progress monotonically through these? Or do sometimes people go one,

00:52:48.120 two, four, three REM, for example?

00:52:51.260 No.

00:52:51.640 You have to go through three to get to four?

00:52:53.640 You have to go to three to get to four. You have to go from two to get to three to get to four.

00:52:59.800 With the exception of probably, let's say, two examples. One is pathological. If you suffer from

00:53:07.520 narcolepsy, one of the things that typically happens is that you go from being awake straight

00:53:13.080 into REM sleep. So to sort of finish the 90-minute cycle, and this will make sense, you go sort of

00:53:19.320 stage one to stage two, stage two, three, three, four. Then after about 70 minutes, you'll rise back up

00:53:26.060 into stage two, and then finally you'll pop up into REM sleep, and then you'll have a short REM sleep

00:53:32.180 period, and then you'll go back down again, down into non-REM sleep, up into REM, down into non-REM sleep,

00:53:38.200 up into REM. And you do that, as I said, every 90 minutes. However, what changes is the ratio of

00:53:45.900 non-REM to REM within that 90-minute period as you move across the night, such that in the first half

00:53:53.780 of the night, the majority of your 90-minute sleep cycles are comprised of lots of deep non-REM sleep

00:53:59.800 and very little REM sleep. But as you push through to the second half of the night, now the seesaw

00:54:06.020 balance changes, and the majority of those sleep cycles are comprised much more of rapid eye movement

00:54:11.280 sleep and almost no deep sleep. And that's why it's always dangerous when people say, you know,

00:54:16.920 I'm one of those who survives on, you know, six, six and a half hours of sleep at night. I wake up,

00:54:21.940 you know, 5, 5.30 to get a jumpstart on the day. And you can ask, well, let's say you're getting six

00:54:27.860 hours of sleep. How much sleep are you losing? And they will say, well, according to your definition

00:54:33.440 of eight hours of sleep, well, I'm 25% off because I'm getting, I lose two hours at the end of an eight

00:54:39.200 hour night of sleep. So I'm sleeping 75%. I'm losing 25%. Yes and no, because yes. It's disproportionate.

00:54:47.060 Exactly. So yes, you're losing 25% of total sleep, but you could be losing up to 70% of all of your

00:54:54.220 REM sleep because it's the REM sleep rich phase that you are shortchanging to get a jumpstart on the day.

00:55:01.000 So you've got to be very careful when you try to think of your sleep dynamics. But to come back to

00:55:05.700 your point, sorry, yes, you always have to go through that linear progression. Narcolepsy is a

00:55:10.560 standout case there. You'll usually go from wakefulness sometimes immediately into REM sleep,

00:55:15.260 very frightening. Most people don't experience, you know, logical, rational waking consciousness,

00:55:21.260 and then go immediately into irrational hallucinogenic consciousness, which is what REM sleep is.

00:55:27.420 It's very disturbing when you speak to these patients. The other is if you are horrifically REM sleep

00:55:32.900 deprived or chronically sleep deprived, occasionally you can make the transition straight into REM sleep.

00:55:39.440 We see this in people who are abusing alcohol heavily, who are alcoholics. Alcohol and will come

00:55:44.880 onto this blocks your REM sleep. You can build up such a hunger for REM sleep that the brain says,

00:55:50.820 well, tonight I'm going to forego the non-REM sleep. I'm just going to go and feast on REM sleep

00:55:54.960 straight away. There's hysteresis in the system, maladaptively so. That's a bad sign when you

00:56:00.220 get hysteresis like that, that type of pressure. Does this explain, and maybe this is not even

00:56:05.800 really correct, but you could see that someone who's constitutively sleeping six hours a night

00:56:11.460 would suffer perhaps fewer of the memory consolidation issues because they're still

00:56:17.780 getting much of their deeper sleep earlier in the evening, presumably where that hippocampal

00:56:24.140 consolidation takes place and the lack of REM's manifestations are just harder to see. Like,

00:56:30.620 does that partially explain any of that or is that, is that even accurate?

00:56:34.480 It turns out that different types of memory rely on different stages of sleep at different times of

00:56:40.520 night. So what we found is that you need sleep after learning to essentially hit the save button on

00:56:46.620 your memories so that you don't forget. And for that, it's all about deep non-REM sleep. So as long as

00:56:52.780 you were getting, let's say the first, you know, half of the night, the first five hours, then it's

00:56:57.560 likely that you may have been able to protect those memories. But here's the other problem that stage

00:57:03.120 two non-REM sleep, which actually happens most in the second half of the night, in the second half

00:57:08.540 of the night, you're having either REM sleep or stage two. And in fact, if you look at those sleep

00:57:12.600 spindles, they exponentially increase towards the end of the night. So you get most of your sleep

00:57:18.880 spindle rich sleep in the last two hours of the night. The reason that's important is because

00:57:24.620 you not only need sleep after learning to consolidate, to save those memories, you also need sleep before

00:57:32.020 learning to get your brain ready in preparation to actually lay down new memories. And there it's stage

00:57:39.920 two non-REM sleep and these spindles. So what we've discovered is that the more of those sleep

00:57:45.820 spindles that you have, the greater refreshment of your memory encoding ability. So think about it

00:57:52.340 like this. And this is a very beautiful experimental demonstration because you can surgically deprive

00:57:57.660 people of each of those. And I've seen other talks you've given, of course, and in your book,

00:58:02.480 you touch on this where you could preferentially from an experimental standpoint, prevent somebody

00:58:07.500 from consolidating memory. And you could similarly prevent them from forming new memory.

00:58:11.640 Even though we can keep the total amount of sleep consistent. So it's not a stressor. It's not

00:58:17.240 that wakefulness. It's, you can selectively excise, just like you said, individual stages of sleep.

00:58:23.640 What's the biggest detriment that persons face clinically with the reduction of REM?

00:58:29.080 I think right now, based on the weight of the evidence, it probably starts with mental health.

00:58:33.500 Anything from depression, anxiety, through to disease.

00:58:37.720 Anxiety, depression, and then suicidality. I think one of the most striking things that we've

00:58:44.540 been seeing in our hands and other people have demonstrated too, and this is probably the second

00:58:48.980 other big area of work that we do here at the Sleep Center is sleep and mental health. You know,

00:58:53.320 sleep is emotional first aid, bottom line period. And what you find is that in young teenagers,

00:58:59.460 one of the strongest predictors of firstly suicidal ideation, in other words, thoughts of taking their

00:59:06.720 own life, strongest predictors of suicide attempts, and tragically, a very strong predictor of suicide

00:59:14.100 completion is insufficient sleep and sleep fragmentation. And what we've been finding is

00:59:20.540 that it's REM sleep that seems to provide essentially a form of overnight therapy. And it's REM sleep that

00:59:27.440 resets or recalibrates the emotional networks in the brain. And there's a series, there's a network

00:59:33.940 of them that involves an area called the prefrontal cortex, and particularly the middle part of the

00:59:39.360 prefrontal cortex, which acts like a top-down regulator of your emotions. So there's deep emotional

00:59:46.400 brain centers, centers like the amygdala, which is a centerpiece region for the generation of strong,

00:59:51.840 impulsive, aversive, negative reactions.

00:59:54.920 Anyone with a young boy knows what the amygdala does.

00:59:57.980 Right, exactly. If they have a temper tantrum in the middle of, you know, the supermarket,

01:00:02.580 it's because firstly, their prefrontal cortex, you know, has just not developed yet. And what

01:00:07.920 little of it that there is, if they haven't slept well, is obliterated.

01:00:12.220 That's actually a really great example of one of the most obvious observations. Like we taught,

01:00:16.940 you know, I mean, any parent listening to this knows how emphatically you protect the nap.

01:00:22.780 Correct. You know, and it's like-

01:00:25.340 For mental, for emotional, mental health.

01:00:27.540 Yeah, just for behavioral health of the child, right?

01:00:30.060 Correct.

01:00:30.640 And it's sort of like your day revolves around the nap. Like, can you go to Disneyland today? I

01:00:35.780 don't know. It depends if we get the nap schedule and, you know, yes or no. And what you're basically

01:00:40.120 providing is a technical explanation for something that is so empirically obvious, it's undeniable.

01:00:46.240 And it's not the opposite, right? So bear with me for a moment. When it comes to the broader mental

01:00:51.700 health stuff, you could easily construct an argument that says the arrow of causality is

01:00:56.960 the other way. The person who is anxious, it's the anxiety that's keeping them awake. Now, my guess

01:01:03.380 is in reality, these two feed off each other constantly.

01:01:06.080 I think you're absolutely right.

01:01:07.200 You might be predisposed to anxiety that fosters lack of sleep. The lack of sleep amplifies and boom,

01:01:12.360 boom, boom, boom.

01:01:12.740 It's a two-way street, but is the traffic flowing more dominantly in one direction than

01:01:17.020 the other? And I think this is what your point you're going to make.

01:01:19.180 But then the behavior of a four-year-old is a perfect example of you only need to screw that

01:01:25.260 up one night. You give a four-year-old one bad night or miss one nap. I mean, it's very obvious

01:01:32.960 how that translates into the behavior. So that would point in the direction of the sleep can really

01:01:38.660 be the driver of the pathology, if you want to use pathology to describe that, which of

01:01:43.480 course is a strong term to describe, you know, outbursts of four-year-olds. But just to make

01:01:47.500 the point, right, that to me is one of the most interesting, albeit somewhat glib, examples.

01:01:52.700 But I think it's so consistent, you know, and I use that frequently where you often see a parent

01:01:58.280 with a child and the child is crying and they look at you and they say, well, they just didn't sleep

01:02:01.800 well last night. As if there's this universal parental knowledge that bad sleep the night before

01:02:06.720 equals bad mood and emotional reactivity the next day. And what's striking is that if you look at

01:02:13.620 the data somewhere between infancy and even now childhood, not only do we abandon the notion that

01:02:20.440 sleep is absolutely essential and non-negotiable, but we start to stigmatize it with this label of

01:02:27.960 laziness, that you're slothful or that your child is slothful for getting sufficient sleep. And the reason

01:02:35.600 that upsets me, and I'm getting a little bit animated, which is odd for a British man, but

01:02:42.140 the, you know, the reason I do is because firstly, you know, it took mother nature 3.6 million years

01:02:49.660 to put this eight-hour thing called a night of sleep in place. And within the space of 70 years,

01:02:56.300 if you look at the data, we've lopped off almost 20 to 25% of that. You know, imagine coming along and

01:03:02.180 saying, you know, in the next hundred years, I think what I'm going to do is for the entirety of

01:03:08.500 human society, I'm going to reduce their oxygen saturation by about 20 to 25%. Do you think that's

01:03:14.700 a good idea? And the answer is no. No, it's such a great example. I'll pause for a moment just to tell

01:03:20.720 a funny story that you and I have talked about off mic, which is up until about 2012, I was in the

01:03:26.680 I'll sleep when I'm dead camp. And I know what led to that. It wasn't that there was a very deliberate

01:03:32.420 decision at the end of medical school when a good friend of mine with all the best intentions,

01:03:37.740 who was a year ahead of me. So he was now in at the end of his internship, as I was about to begin

01:03:43.760 mine, he said, and this is in the days when we didn't have the 80 hour work week requirement in

01:03:49.980 residency. So we averaged, I think about 114 hours a week in the hospital. So he said to me,

01:03:57.500 look, Peter, you're, you're signing up for, you know, whatever, five, seven years of this thing.

01:04:02.620 If you spent every moment outside of the hospital sleeping, you would still be tired. The only

01:04:08.120 difference is you wouldn't have any fun. So make sure you live every moment that you're not in the

01:04:15.620 hospital to the fullest. And so for me, that basically meant if I wasn't in the hospital,

01:04:19.640 I was swimming, I was going out with my friends, I was trying to meet girls, like I was doing

01:04:24.520 anything and everything such that during that period of my life, I, I just know, cause I was

01:04:30.020 pretty adamant about recording how much time, like I was very wed to this idea. There's 168 hours in a

01:04:35.380 week. If I'm spending, you know, 114 of them here and I spend this many driving and I spend this many

01:04:40.000 getting groceries and I spend this many swimming and blah, blah, blah, blah, blah. I think I was about

01:04:44.300 28 hours a week of sleep. So it wasn't four every night because you'd have none and then six and then

01:04:51.680 three and then eight, like you, you could binge sleep from time to time, but it was pretty much

01:04:56.760 28 a week. I'll come back to some of the implications of that, but fast forward a few years, I'm talking

01:05:02.040 to a good friend of mine, Kirk Parsley, who's a physician who like you is adamant about, you know,

01:05:06.380 the importance of sleep. And we're having dinner one night and he says, he's challenging me on this.

01:05:11.800 And he says, so let me get this straight. You've just decided that you're going to sleep half of

01:05:16.520 what is evolutionarily programmed. And I said, yeah. He goes, does it strike you as odd that

01:05:23.780 evolution would have designed us to spend a third of our life, not mating, not watching out for

01:05:30.600 predators, not hunting for food, but doing this thing for some other purpose? Do you think that

01:05:36.620 thing must've been important? And it was such an obvious argument, but it really overnight changed

01:05:42.680 the way I thought about this, which was evolution went to great lengths to do this and superficially

01:05:50.700 at great cost to us, right? I mean, you could argue, well, imagine you didn't need to sleep and

01:05:54.780 you could spend 24 hours a day foraging for food or a mate or some other thing, but it didn't.

01:06:00.100 So it's sort of, it's sort of like, there's probably a reason we are not anaerobic to your

01:06:06.460 point about reducing oxygen saturation by 25%. You know, if you were to think about that, you know,

01:06:10.960 during sleep, just as you said, you're not eating, you're not finding food, you're not finding a

01:06:17.460 mate, you're not reproducing, you're not caring for your young, you're vulnerable to predation

01:06:22.360 on any one of those grounds, but especially all of them put together as a collective.

01:06:27.140 It's completely anti-evolutionary. 0.99

01:06:29.180 It sounds like the dumbest thing. And, you know, often said, and it has been said before, 0.98

01:06:34.040 if sleep doesn't serve an absolutely vital set of functions, it's the biggest mistake that the

01:06:40.700 evolutionary process has ever made. And we now realize from this constellation of evidence that

01:06:47.560 mother nature did not make a spectacular blunder in putting this thing called an eight hour sort of

01:06:53.460 need of sleep in place. It is the greatest life support system that you could ever wish for.

01:06:59.680 It is a remarkable health insurance policy. And what's great is that it's largely democratic.

01:07:05.820 It's mostly free. And in terms of a prescription from a doctor, it's largely painless.

01:07:12.300 So I almost wanted to title the book, Consciousness is Overrated,

01:07:16.420 or just sometimes dot, dot, dot, consciousness is overrated. But when you really look at the

01:07:23.240 evidence in terms of risk, de-risking just about every disease that is killing us in the developed

01:07:30.020 world, it's very hard to look no further than sleep. And that's why I don't want to trivialize

01:07:36.540 diet and I don't want to trivialize movement and activity. But what I would say is that if you want

01:07:43.000 to put sleep up against either one of those two and kind of play the whole head-to-head game,

01:07:48.160 which I don't think we need to do here, I would simply say that sleep is the foundation on which

01:07:54.840 those two other things sit. It's not the third pillar of good health. I think it is the foundation.

01:07:59.340 That's a really interesting way to think about it because I typically describe four pillars or five

01:08:05.120 if you include all of the exogenous molecules that you could lump together. But another way to think

01:08:10.440 about it, which again, I don't think is necessarily the right way to think about it, but sometimes it

01:08:14.180 makes the point. If you deprive yourself of food, how long can you survive? Well, we have one person

01:08:21.000 up to 382 days. Even someone who's as lean as you could survive 30 days with no food. How long could

01:08:28.320 you survive without water? Depends greatly on the temperature, et cetera. But you could make the case

01:08:34.400 that deprivation of sleep would result in the quickest reduction of health, certainly more than

01:08:40.960 not eating or not exercising for a period of time. And those studies have been done in rats.

01:08:46.300 Yeah. And actually we know some of this from humans who have been trying to... In fact,

01:08:50.140 didn't the Guinness Book of World Records, I can't remember if I read this in your work,

01:08:54.000 they've actually banned attempts at longest period of sleep deprivation.

01:08:58.600 So I mentioned this. Yeah. There was a time when you could still try and beat the world record of

01:09:03.560 sleep deprivation. And it got up to about sort of, I think the last true effort was about 24 days.

01:09:11.540 But I think it was debatable, that one. But based on the weight of the scientific data,

01:09:17.980 the relationship between sleep loss and mental health, sleep loss and cancer, sleep loss,

01:09:24.520 cardiovascular disease, sleep loss, and metabolic syndrome, Guinness started to feel very,

01:09:31.160 very uncomfortable. And then when suicide came on the table, it pulled it. So in other words,

01:09:36.480 think about this. You know, there was a gentleman, Felix Baumgardner, I think his name was,

01:09:41.300 who sponsored by Red Bull went up in a capsule in a hot air balloon to the outer surface of our planet.

01:09:48.680 This was about four years ago.

01:09:49.660 Four years ago.

01:09:49.960 He opened the door and then he jumped out and he fell back down to earth at over a thousand

01:09:58.820 kilometers an hour. Using his body alone, he broke the sound barrier and he successfully came down.

01:10:05.840 And now Guinness says for that, just fine. However, to sleep deprive yourself, no,

01:10:15.140 much more unsafe, we're not going to let it happen. Just to put it in context.

01:10:20.940 You're allowed up to 12 jumps off Niagara Falls.

01:10:24.760 Basically.

01:10:25.360 But that's okay. But no, no, you're not going to, that's such a great point. I want to go back to

01:10:28.320 one other thing to close the loop on this. And again, the listener may say, boy, this is,

01:10:33.080 there's a lot of stuff going on here, but there really is. Like this is, I think you have to have

01:10:37.140 a lot, and I'm trying to think about all the questions I can ask that others would have.

01:10:40.020 If you took two individuals who are sleeping six hours a night, if you have one person who just

01:10:45.200 for the argument's sake is doing it from 9 PM to 3 AM, and another person's doing it from midnight

01:10:50.840 to 6 AM, would you expect, or even if it were, let's make it even more clear. It's the same

01:10:55.460 individual, but you do it over a period of weeks. You transition them from nine to three versus

01:11:00.620 midnight to six. Would their sleep wave cycles look the same? Would they still be truncating the

01:11:07.340 same amount of REM at the tail end of that? Or would they start to do some forward truncation?

01:11:13.760 Forward truncation. So essentially what happens is that your brain has a different hunger or different

01:11:20.900 appetite for different types of sleep. So think of those different stages of sleep, like a finger

01:11:26.240 buffet. And at different times of the night, that buffet is going to be sort of, you know,

01:11:32.160 consumed more heavily in terms of, you know, the deep non-REM sleep phase, which would probably be

01:11:38.880 the hours from 9 PM through until about sort of 3 AM. That's really the window of preferential

01:11:46.940 appetite for your brain when it comes to deep non-REM sleep. That's when it will preferentially

01:11:51.860 eat at that type of sleep. But then from about 3 AM through until about midday, then the brain shifts

01:12:00.820 in terms of its pattern. If you let people just sleep it out, you know, if you look at sort of

01:12:05.660 extreme owls who, and we should speak about chronotypes at some point, but people who, you know, like to go

01:12:11.120 to bed very late and wake up very late, essentially there, then the buffet is going to get hit hard in

01:12:16.980 terms of REM sleep devouring. So I make that point that your brain has different taste sensitivities to

01:12:24.400 non-REM and REM across the 24 hour clock face, because it means that if you're shortchanging

01:12:30.820 your sleep by two hours, but you're doing it by going to bed very early and waking up very early,

01:12:36.800 then what's going to happen is that you're probably going to be losing most of your REM sleep because

01:12:41.200 most of your REM sleep happens in the late morning hours and you're already awake by 3 or 4 AM in the

01:12:47.160 morning. And so you're not getting the chance to get into those REM sleep rich phases.

01:12:52.340 And which of those patients would develop more appetite for sleep the subsequent evening?

01:12:59.140 The person who was deprived, and you know why I'm asking, by the way, it's exactly what we were

01:13:03.020 talking about earlier. It's this jet lag question. If you know you've got that flight and you need to

01:13:08.600 force yourself to sleep at a time zone, that's not right. I'm just now thinking about, well, I always do

01:13:13.940 the nine to three. Would I be better off doing the, in other words, I'm depriving myself of REM.

01:13:20.300 Would I be better off just describing, just restricting myself?

01:13:23.880 Well, you're depriving yourself of REM, but you're also still depriving yourself of total sleep as

01:13:28.460 well, you know, when it comes to the jet lag model. And, you know, there are, there's some,

01:13:32.360 I think that that's an interesting model. And I think it does work that you can try to,

01:13:36.960 in the week before you start to fly, let's say you're going, I'm going home to London,

01:13:41.240 which essentially is eight hours ahead. So I need to now essentially start to sort of,

01:13:46.820 by the time I'm waking up, people in London have already been awake for essentially eight hours.

01:13:51.460 So I need to start to get myself on the schedule. So I start to wake up, you know, 10 minutes

01:13:56.960 earlier for about five days before. And, you know, so I'm only about an hour shifted, but it helps.

01:14:04.140 And then the morning of the flight, I'll wake up especially early. And then I won't make the

01:14:10.720 mistake that everyone makes on those transatlantic flights, which is they sleep in the second half

01:14:16.780 of the flight. You shouldn't do that. You should sleep in the first half of the flight, which is

01:14:21.940 when most people in the UK are asleep. And by the middle of that flight, most people in the UK,

01:14:27.260 that's the time to wake up. And so you should be waking up that time so that you build up enough

01:14:32.740 of this sleepiness pressure by the time that you want to get to bed once you've landed that evening

01:14:40.460 in London. And this won't make too much sense right now. And I think we'll, we'll speak probably a lot

01:14:45.820 more about this when we speak about caffeine and adenosine and sleep pressure, but that, but to come

01:14:51.500 to your point, you're building up what's called sleep pressure. The longer that you're awake, the more

01:14:57.360 of that healthy sleepiness that builds up. But to come back to the real question at hand, yes, just

01:15:03.780 because you sleep, you know, five hours from one point in the clock face, maybe earlier in the night

01:15:10.980 to sort of early in the morning versus five hours late in the morning until later in that day, the type

01:15:17.860 of sleep that you get will not be the same. You will be selectively shortchanging yourself of sleep,

01:15:25.040 either of deep sleep or of REM sleep on either side of that equation.

01:15:30.100 This points out another interesting observation, which I do want to spend some time talking about

01:15:34.120 wearables because of course, you know, we're both wearing our wearables at the moment. And I think we

01:15:38.500 both acknowledge that, look, I mean, these things are still not quite at the level of the polysom

01:15:42.540 and maybe they never will be. I mean, that's a very difficult thing to replicate from, you know,

01:15:46.540 something on your wrist or your finger. But it's so funny, this might explain an observation I

01:15:52.300 always experience when I come from New York back to California, which is I generally try to stay

01:15:58.580 on California time, even when I'm in New York, going to bed later, waking up later. And in

01:16:03.260 California, I'm a very early sleeper and an early riser. So even if I'm trying to keep it at eight

01:16:08.420 hours and eight hours, and I can do that very well in California, not as well in New York, my first

01:16:14.500 night back, even my first two nights back in California, I have much higher deep sleep. And that

01:16:20.360 would explain it. It's this phase shift that I'm capturing this. I mean, last night was because

01:16:25.800 I came back from New York last night, you know, went to bed at, I don't know, 8.30 p.m. and got up

01:16:32.260 at five or yeah, probably 4.30 or five in the morning. But had a huge deep sleep. Exactly. So

01:16:38.840 you've dragged your eight hour window, if you think about the sort of eight hour window, and you slide

01:16:43.420 it back on the clock face from midnight to 11 p.m. to 10 p.m. to in fact, 9 p.m., you're dragging that

01:16:51.000 eight hour sleep phase back closer into the deeper sleep rich phase, the deep non-REM sleep that you

01:16:59.200 want. So what your sort of aura ring will be telling you is you got more deep sleep. And perhaps you may

01:17:05.660 actually see a little bit of a reduction in REM sleep. As you go to New York, you may see the opposite

01:17:11.660 as you sleep later in the morning, you'll get more of that REM sleep. I generally do see that

01:17:15.840 pattern, which is I get, and it, you know, it pisses me off too. I'm like, I didn't get as much

01:17:20.140 deep sleep and I got more REM sleep. And not that REM is bad, but I'm generally more upset by the

01:17:25.780 dearth of deep. You know, this is interesting. It's now really, I really need to think about that

01:17:29.980 because of course, on some level, our body is responding to cues in the environment. It's not just

01:17:34.740 responding to the intellectual knowledge that, hey, in California, it's this time, it's time to go to bed.

01:17:38.980 I mean, light is playing a role. Meal timing is playing a role.

01:17:42.760 Huge meal. And that's one of the things.

01:17:43.780 I want to definitely talk with you about this because this is, to me, it's such an important

01:17:47.580 topic. And then of course, the accumulation of adenosine and of course the melatonin. So

01:17:51.940 by the way, I would just say that, you know, I think REM sleep, some people, I have this kind

01:17:57.160 of bimodal distribution of questions. Some people will say, how do I maximize my REM sleep? And I say,

01:18:03.580 why do you want to maximize your REM sleep? And say, well, that's dream sleep and that's the good

01:18:07.860 stuff, right? As if, you know, it's like saying, you know, how do I maximize the amount of yolk in

01:18:12.640 my egg? You know, do I get the, like the, can I go the ostrich egg? Whatever it is. I don't know

01:18:17.780 what the analogy is. And the question is, mother nature has found this equilibrium. Don't mess with

01:18:22.660 it. It's the Da Vinci code of sleep and stick to it. You're good to go. But also then other people

01:18:28.620 will say, you know, how do I get more deep sleep? Or, you know, I, you know, I could get upset by

01:18:33.240 that lack of deep sleep to come back onto REM sleep and how important it is. The studies have

01:18:38.820 been done where firstly, they deprived, and by the way, these studies have never been done again

01:18:44.100 since the 1980s. And they probably never will be because of the ethics regarding them, because they

01:18:49.020 are just so devastating that we probably won't be allowed, and rightly so, I think, to do these studies

01:18:54.720 to animals ever again. But they essentially deprived animals of sleep until they died. And what

01:19:01.980 they found was that rats will firstly die as quickly, if not quicker, almost by about 20%

01:19:09.420 from total sleep deprivation as they will from total food deprivation. So if you want to put sleep and

01:19:15.480 food... That's interesting that their gap is much narrower, because for humans, the gap is very...

01:19:20.940 You'll die way earlier from sleep deprivation than food deprivation, unless you're, you know,

01:19:25.180 the leanest person on the planet. And there is a tragic disorder that I describe a little bit in the

01:19:29.860 book. It's a very rare, genetically inherited disorder called fatal familial insomnia. And it's

01:19:34.860 really the only evidence short of the disgusting travesty that is total sleep deprivation used in

01:19:43.580 certain countries for interrogation tactics that we know of that has led to death. But withstanding

01:19:49.600 that, this disorder, fatal familial insomnia, says it in the name that essentially human beings...

01:19:55.980 How old do people typically survive?

01:19:58.540 It usually takes about 18 to 24 months before they die. But the problem is this is not,

01:20:05.040 they wake up one day and they are no longer sleeping at all. It's a gradual thing. So

01:20:09.340 it's kind of like it masquerades as just typical insomnia first, restless nights for the first two or

01:20:14.960 three months. Then it gets worse. Then you're only sleeping four hours. Then, you know, by month 12,

01:20:21.240 you're already sleeping two hours.

01:20:23.320 How much of this is genetic versus environmental versus? We don't know.

01:20:27.120 It's strongly heritable that we know. So it's within the Mendelian sort of, you know,

01:20:33.600 pool of genetic disorders. And it's testable. We know the gene. Interestingly, it's what seems to be

01:20:39.340 the cause.

01:20:39.640 Oh, it's a single gene?

01:20:40.080 Yeah. And it's a prion protein gene abnormality, which is, if you don't know about prion proteins,

01:20:46.940 you may have heard of mad cow disease, which was rife in the United Kingdom. That's another thing.

01:20:51.840 That's another prion protein disorder.

01:20:53.900 That's why you can't donate blood.

01:20:55.300 And it's also why, yeah, I'll probably forget half of what I've been saying and start repeating myself

01:21:00.020 because I've been vegetarian for about 12 years. But before I went that route, I think I definitely

01:21:04.980 was in the phase of eating beef when it was filled with these prions in the United Kingdom.

01:21:11.280 So my brain is probably Swiss cheese at this stage. But essentially, this disorder,

01:21:15.640 fatal familial insomnia, it gives us some degree of disease proof that insufficient sleep is fatal

01:21:21.520 for human beings. But we knew it in these rat studies because these rats would die as quickly

01:21:25.640 from food deprivation as they did from sleep deprivation. But the most striking thing was then

01:21:30.520 they did a second round of studies. They selectively deprived them of either just non-rapid eye movement

01:21:37.120 sleep or rapid eye movement sleep. And what they found is that the rats were dead within about

01:21:43.220 nine to 10 days with total sleep deprivation. They were dead within just about the same period of time

01:21:49.860 from REM sleep deprivation.

01:21:52.540 Duration was kept constant or also reduced?

01:21:55.380 It was still reduced somewhat.

01:21:57.540 But you could probably look at a control that had the same duration without a selective deprivation.

01:22:02.880 Yeah. And there is what's called a yoked paradigm. It's a very clever technique where you can mimic

01:22:08.240 the total amount of sleep deprivation in one other animal, but the other animal is still getting the chance

01:22:14.100 to get REM sleep. So REM sleep selectively removed. The rats died basically almost as quickly as they did

01:22:20.660 from total sleep deprivation. And then non-REM sleep deprivation was deathly. Ultimately, it just took

01:22:28.500 longer. It took about 20 days.

01:22:29.940 Total sleep deprivation, let's just say for argument's sake, you die in nine days. Sleep reduction,

01:22:35.400 but selective REM deprivation, you die in the same period of time.

01:22:39.000 Basically, yeah.

01:22:39.580 Sleep reduction, but deep deprivation, you live maybe twice as long.

01:22:44.940 Yeah. So everything deathly, everything fatal, just that one takes a little bit longer to kill

01:22:51.880 you than the other. And what I find either strange or exciting about that is if you look

01:22:58.780 from an evolutionary standpoint, those two types of sleep did not emerge at the same time. We only

01:23:05.260 have observed true rapid eye movement sleep in birds and mammals. We don't see true rapid eye movement

01:23:14.700 sleep in reptiles, amphibians, fish, or insects. Now, birds and mammals evolved separately along the

01:23:24.120 evolutionary path from amphibians, fish, and reptiles. So what this means is that when we went from the

01:23:32.060 transition from being reptilian to either avian or mammalian, that's when REM sleep emerged because

01:23:40.200 we see non-REM sleep in every species. Even in insects, we can observe something that looks like

01:23:46.080 non-REM sleep. In reptiles, fish, amphibians, birds, and mammals, every one of them has non-REM sleep.

01:23:53.040 Not every one of them has REM sleep. Only birds and mammals have REM sleep. In other words,

01:23:58.420 REM sleep is the new kid on the block in the time course of evolution. REM sleep firstly emerged later

01:24:05.120 in the course of evolution. And secondly, what's kind of wonderful is it emerged twice independently,

01:24:13.340 once in birds and once in mammals. So you could argue that, well, if deep non-REM sleep, which is

01:24:19.700 much more evolutionary from an evolutionary perspective, it's a much older form of sleep,

01:24:24.600 wouldn't it be the much more necessary fundamental one? And therefore, wouldn't you die sooner from deep non-REM sleep?

01:24:31.540 It seems the opposite because you'd say, well, for the same reason that the cockroach can survive the nuclear

01:24:36.520 disaster, the avian mammalian pathway is more evolved neuronally, presumably, and therefore probably requires

01:24:44.500 something essential in REM that when it deprives you, it takes away your superpower, which is a cortex.

01:24:51.860 And I think that's exactly what it is, which is when you look at anything that has essentially a complex

01:24:57.400 nervous system, maybe that's where the pressure for this thing called REM sleep emerged. Even though

01:25:04.060 we know that non-REM sleep actually supports many essential central nervous system functions like

01:25:09.640 synaptic pruning and some great sort of neural function, you know, the glymphatic system clearing up.

01:25:15.800 It's sort of similar to your oxygen analogy, which is you, when you go from being an anaerobe to a

01:25:21.540 faculty of anaerobe to an a-robe, who's going to be most sensitive to oxygen deprivation?

01:25:26.540 Exactly. And so, you know, it's always a dangerous scientific hypothesis where you can argue it both

01:25:30.740 ways, but I always found that interesting and it's uncommented, but I just wanted to note that for,

01:25:35.960 you know, 20 minutes ago in kind of in defense of REM sleep that, you know, if we should all be worried

01:25:42.800 about the amount of deep sleep that we're getting, absolutely. But we also need to be worried about

01:25:47.600 how much REM sleep that we're getting, because from a pure mortality risk standpoint, REM sleep

01:25:54.060 is arguably even more important than non-REM sleep.

01:25:57.160 And we'll come to this later, but my recollection of the literature is, and I think you even alluded

01:26:01.800 to this earlier, ethanol's biggest hit is going to be on REM, isn't it?

01:26:05.540 Hands down. This is where we come on to substances that are misunderstood and misused when it comes to

01:26:12.020 sleep. And now, before we go there, Matt, look, for me, I don't need any more of a case in the

01:26:16.840 importance of sleep. And I think honestly, the Alzheimer's thing alone should be sufficient for

01:26:22.960 anyone to take this seriously, because certainly my patients, that's the single most important thing

01:26:29.180 to them. Now, there are some of them that are more high risk, you know, people that have,

01:26:32.580 you know, APOE4 genotype that's in the, you know, the direction that would increase risk or family

01:26:37.060 history. But I don't think I've met a patient yet who doesn't say on some level preserving

01:26:42.000 cognition is their single highest priority in life. That said, can you spend even just a minute,

01:26:48.860 a couple of minutes, it's the drive after all, we have all the time in the world, making the case,

01:26:54.600 the similar case in the presence of cardiovascular disease and cancer?

01:26:58.440 Yeah. And I want to touch on that point before we go on about this Alzheimer's disease risk. You know,

01:27:05.160 a lot of people out there may be having this sort of like sleep machismo moment where they have this

01:27:10.980 braggadocio attitude of getting insufficient sleep. And they say, look, I'm in my fifties and, you know,

01:27:16.800 I'm fit as a fiddle and my cognition is razor sharp and I sleep five or six hours a night. So we've got,

01:27:23.060 you know, heads of state, for example, who are saying this, people who, you know, have a huge impact

01:27:28.420 on the public and they're saying these things. And it makes me sad because you can look at two other

01:27:34.360 heads of state who were very vociferous about their proclamation of the uselessness of sleep.

01:27:41.920 Margaret Thatcher and Ronald Reagan, both of them were chess beaters when it came to insufficient sleep.

01:27:48.620 They were very proud of saying that they only got four or five hours of sleep, that they were immune

01:27:53.540 to the effects of insufficient sleep, that sleep was for the weak and that you can sleep when you're

01:27:58.160 dead, which P.S. is mortally unwise advice based on the evidence. The shorter your sleep,

01:28:03.360 the shorter your life. That's what all of the studies tell us. I don't think it's coincidental

01:28:08.880 now looking back that both of them went on to develop Alzheimer's disease. And it makes me so sad

01:28:16.740 because I saw pictures of Margaret Thatcher in an enfeebled state of dementia. And I'm just thinking,

01:28:25.100 you know, if only when she was, you know, in her fifties and had that bravado about sleep, 0.55

01:28:30.280 if she could see herself in that bed, in that later state of life, in such a crippled mental 0.65

01:28:37.700 cognitive functioning state, would you be so brave about it? And I would hope that she wouldn't,

01:28:45.080 that humility would come into place. So that's the first thing about, I think,

01:28:49.640 the insufficient sleep argument and people rightly being frightened, having the daylights frightened

01:28:55.540 out of them about preserved cognition. You should have that regarding your sleep. The other one,

01:29:00.520 though, that frightens me too, is the other way that a lack of sleep can get you is not chronically

01:29:05.260 across the lifespan with Alzheimer's disease, but acutely with car crashes. And I know that we'll come

01:29:11.080 on to this at some point too, but the only thing I would say is that I get many emails from people

01:29:16.060 who've read the book. And two months ago, I got an email from a gentleman, I won't name him,

01:29:20.980 and he was in his late thirties. And he described how three months before that, he had lost his wife

01:29:30.900 and his 18 month, 18 month old child in a car crash. And it later became apparent that the person

01:29:39.120 who hit them had basically been surviving on about five and a half hours of sleep for the two months

01:29:44.700 prior that they'd had what's called a micro sleep. They actually lost control of their car at 60 miles

01:29:51.400 an hour at a moderate drift angle. You are essentially in the next lane within less than a second,

01:29:59.280 and you're in the ditch after a second. And they went straight head on. And now he was going to have

01:30:05.400 to raise his four-year-old son by himself without the wife that he loved, without the 18 month daughter

01:30:13.520 that they'd invested in. Because someone perhaps had heard from someone vocally in the public

01:30:20.140 that surviving on five hours of sleep was something that was okay to do. And to me, that just breaks

01:30:27.580 my heart. I just don't feel comfortable in, you know, having the responsibility like the head of a state

01:30:35.420 of a country. And then speaking about this thing, this kind of magic attitude of insufficient sleep,

01:30:43.520 because that's what it can lead to. It can lead to the travesty of the explosion of a nuclear

01:30:50.240 family. And my guess is, even though it's going to be very difficult to actually quantify this,

01:30:56.760 I believe that the story you just shared is an especially tragic, poignant example of something

01:31:05.200 that probably occurs much more than we were able to quantify. I sometimes say to patients, you know,

01:31:11.160 the four horsemen of death would be these big chronic diseases that kill us, but we can never

01:31:15.700 forget accidental death. And then when you look at accidental death, how do you quantify it? Well,

01:31:20.780 it turns out that three forms of accidental death form 80% of them. And the biggest among them is motor

01:31:27.880 vehicle, right? And so the question is, what can you do to reduce that risk? So if you look at it,

01:31:36.500 some form of distraction, and I've included sleep deprivation in that, but now I almost feel like

01:31:41.580 they need to be separated out. Distraction, meaning you're putzing with your phone, you're checking

01:31:46.040 ways, you're doing something you shouldn't be doing with your phone, sleep deprivation, ethanol,

01:31:52.560 failure to wear a seatbelt and reckless driving. I mean, that basically accounts for it. It wouldn't

01:31:57.900 surprise me if that sleep deprivation bucket is bigger than we think.

01:32:03.240 I mean, we've got some decent data on this. And firstly, what we know is in the hour that we've

01:32:10.160 been talking already, numerous people have lost their life in a drowsy driving related accident

01:32:15.740 based on the statistics. We know that already. And the way that we can tell that is both from

01:32:22.240 speaking to the people who are involved in the accident, but you can also pick this apart from

01:32:27.600 the dynamics of the accident as well. And here's the problem. When you are underslept,

01:32:32.440 we're not talking about a night of total sleep deprivation. That's where you fall asleep at

01:32:36.520 the wheel. And that's the obvious one. The much less obvious one, but the much more common one is

01:32:41.560 what we call micro sleeps. So when you are trying to survive on seven hours of sleep or six hours of

01:32:48.460 sleep, and it really just takes that little of a dose. By the way, I haven't forgotten our cardiovascular

01:32:52.840 question. I will come back to it. But what you essentially have are these small lapses. We call them

01:32:59.020 micro sleeps where the eyelid will partially close. And you can pick this up on a camera and they've

01:33:05.700 done these studies systematically where you dose people from getting nine hours of sleep to eight

01:33:10.820 hours of sleep, seven, six, five, four, three, four, one week, two weeks, three weeks. These studies

01:33:18.840 have been essentially looked at, sliced and diced parametrically. They're probably the best studies

01:33:25.360 done within the field of sleep. And so we know them very well, what your lapse rate is, what your

01:33:30.860 micro sleep rate is. And what you find is that the eyelid partially closes. And at that point,

01:33:38.600 you no longer react. And the reason I'm bringing this point up is that drowsy driving accounts for more

01:33:46.820 accidents on our roads than either drugs or alcohol combined, combined. Now I'm not condoning

01:33:54.500 drink driving or driving inebriated under any circumstance. What I am saying, however, is that

01:34:00.560 firstly, drowsy driving is a huge part of this problem on our roads. And the reason that it's not

01:34:07.260 just problematic is because it's usually much more deathly. When you are drunk, you typically react,

01:34:15.380 but you react too late. So you do something, but it's not enough and you crash. When you have a

01:34:22.460 micro sleep, you do nothing. That's why drowsy driving is more fatal than drunk driving typically

01:34:30.900 is because there is no reaction. There's no application of brake. There is no application

01:34:35.560 of brake. There's no course correction of steering wheel angle. There is nothing. There is simply a

01:34:41.080 car with no one in control. And it's a two ton missile on the freeway at 65 miles an hour with no

01:34:48.740 one in charge. This is why I feel really uncomfortable about the idea of public proclamations of

01:34:57.140 insufficient sleep. And I get that. And, you know, there are people, sports stars, all sorts of people

01:35:03.360 who I respect hugely. And, and I have an immense amount of sympathy for where they're coming from,

01:35:11.640 because I think there's a lot of public pressure to be this outspoken. I'm going to be crushing it on

01:35:18.300 the sports field. And then I'm going to be, you know, back in the gym the next morning and sleep

01:35:24.160 is for, you know, X, Y, and Z. Who cares? But then you get someone in a car and you hear a story like

01:35:30.980 that. And I can't help, but just have my heart broken. But I think, you know, Matt, your work

01:35:35.760 has come along at a really interesting point. I mean, I have to be honest with you. I remember when

01:35:39.660 your book came out and my initial thought was, does the world need another sleep book? Because

01:35:46.520 there's 2000 of them out there. And I think there's two things that are different. The first is your

01:35:54.880 book is exceptional. I should say three things. Your book is exceptional. The third is your understanding

01:36:01.820 of the topic and your credibility is exceptional. And those aren't necessarily synonymous. You can have

01:36:07.200 really credible, thoughtful people who write a book, but the book itself isn't exceptional or it

01:36:12.660 doesn't hit the tone. And conversely, you can have someone who puts together a good book, but they

01:36:16.980 themselves don't have the credibility. So you can take credit for the first two. I think the third

01:36:22.140 is the little bit of luck, which is, I think we are at a point in time where people are ready to hear

01:36:30.760 this. In other words, if you, the exact same professor wrote the exact same book 20 years ago,

01:36:37.940 I'm not convinced it would have the same impact it is having today. Now, I don't know. You probably

01:36:43.360 have a better sense than me. I don't know exactly what accounts for that difference today. Is it that

01:36:48.800 today we are more interested in, you know, the thin end of the wedge for this might've been our

01:36:54.800 interest in performance. Maybe that is the thing that first got people to think about this,

01:36:59.340 which is, well, wait a minute. You know, if I want to be better at X, Y, and Z, whether it be my

01:37:04.880 sport, my craft, my whatever, is sleep a tool I need to be optimizing the way people have for many

01:37:11.600 years realized, Hey, what you eat and how you exercise absolutely impacts your performance.

01:37:16.920 So I'm not, I'm not sure what it is, but I, I have a feeling, and I actually had this discussion

01:37:23.140 with my daughter a couple of weeks ago that we will look back at the way we think about

01:37:31.600 nighttime electronics, sleep deprivation, and things like that. We will look back at that in 20 years,

01:37:37.980 hopefully sooner, the way you and I would look back at, you know, a smoking mother being told by 0.88

01:37:44.320 her doctor, smoke this brand versus that brand. Like we, we, we can't even fathom that that was a 0.97

01:37:50.100 discussion that took place as recently as about 1966 in the United States. And yet we're kind of

01:37:56.380 in the, our infancy in the dark from a clinical perspective, from sort of a public health perspective

01:38:02.400 and functioning on this. So firstly, I mean, so much great stuff to unpack there. I think you're

01:38:08.880 absolutely right. I think the success of the book is largely, despite my, my ability to, to write and

01:38:15.920 convey ideas, I think what happened is that it was a perfect storm where there has, I gave you, I gave

01:38:21.640 you two thirds of the credit. Thank you. And I don't know if I can take them. I feel so ashamed, but

01:38:27.660 I think the, the middle point is the erosion of sleep time has hit an all time high. We are now sleeping

01:38:37.920 less than we have ever done in what seems to be the history of our species. Now we can't say that

01:38:45.120 for sure, but what we certainly can say is that back in the earliest record that we have of a

01:38:49.820 systematic questionnaire, which was in 1942 by Gallup, what they found was that the average American adult

01:38:55.660 was sleeping 7.9 hours a night. Now that number is down to six hours and 31 minutes. That's the average

01:39:03.140 6 hours and 31 minutes. That must mean that. How linear has that decline been, Matt? Has that

01:39:08.800 non-linear? Yeah. It's accelerating, isn't it? It's accelerating. Yeah. So I think you've, you've had

01:39:13.560 this ballistic erosion of sleep time, this decimation of sleep throughout industrialized nations. And it's

01:39:21.500 not just America, by the way, Japan is even worse. It's down to 6,021 minutes. You know, my home country,

01:39:27.160 the United Kingdom, not much better, 6 hours and 49 minutes. And these are averages. You know,

01:39:33.660 you've got to imagine that, you know, at least a third of the people are probably trying to survive

01:39:38.160 on five hours of sleep, certainly during the week they are. It's that plus this incredible explosion

01:39:45.680 of sleep science over the past 30 years. And this perfect storm has led to a collision where society

01:39:53.540 is sleeping less, disease and sickness has been escalating in a manner that, you know,

01:40:00.040 we've almost never seen before. And the knowledge of what sleep is doing, we know far more now,

01:40:06.520 sleep is not just curing sleepiness. Sleep is actually supporting, there is no physiological system

01:40:13.900 in the body and there is no operation of the brain that isn't wonderfully enhanced by sleep when you get

01:40:20.040 or demonstrably impaired when you don't get enough. That knowledge and the causal knowledge of sleep

01:40:26.740 in that, it's not just associational, it is many of them are causal links with those disease states,

01:40:32.560 I think led to this ability to write this book at this time that felt as though it was somewhat

01:40:39.700 unequivocal, that it was very difficult to argue with. You know, you can pick apart one study from

01:40:45.020 another, but when you build that blueprint manifesto, when you build the scientific, cold,

01:40:51.720 hard, rational, emotion-free case for sleep, it's very difficult to argue with. And it just happened

01:40:59.500 that we were at that time in our human history with sleep and we were at that time in our scientific

01:41:05.940 journey regarding our understanding of sleep. I think that's why the book felt like it needed to be

01:41:11.920 written and needed to be written now. For what it's worth, you had many moons on the front and

01:41:16.980 back covers, so my little IR student was delighted. He loves that book. Yeah, he's probably read from

01:41:24.460 literally from cover to back cover. Back cover, nothing in between. Yeah, exactly. That's just fine.

01:41:30.020 I ask people, please, you don't have to read the book, you just have to buy it. That's all I ask.

01:41:35.200 Okay, there's like literally a hundred more things I want to ask you about. So can you make the case

01:41:41.640 for why cardiovascular disease is worsened by or accelerated by sleep deprivation? And then same

01:41:48.600 question I'm going to ask you in a moment for cancer. You can find all of this information and

01:41:54.460 more at peteratiamd.com forward slash podcast. There you'll find the show notes, readings, and links

01:42:00.380 related to this episode. You can also find my blog at peteratiamd.com. Maybe the simplest thing to do

01:42:06.720 is to sign up for my subjectively non-lame once a week email where I'll update you on what I've been

01:42:11.420 up to, the most interesting papers I've read, and all things related to longevity, science, performance,

01:42:16.940 sleep, etc. On social, you can find me on Twitter, Instagram, and Facebook, all with the ID

01:42:22.300 peteratiamd. But usually Twitter is the best way to reach me to share your questions and comments.

01:42:27.060 Now for the obligatory disclaimer, this podcast is for general informational purposes only and does

01:42:32.040 not constitute the practice of medicine, nursing, or other professional healthcare services,

01:42:36.880 including the giving of medical advice. And note, no doctor-patient relationship is formed.

01:42:42.300 The use of this information and the materials linked to the podcast is at the user's own risk.

01:42:47.380 The content of this podcast is not intended to be a substitute for professional medical advice,

01:42:51.900 diagnoses, or treatment. Users should not disregard or delay in obtaining medical advice for any

01:42:57.040 medical condition they have, and should seek the assistance of their healthcare professionals

01:43:00.920 for any such conditions. Lastly, and perhaps most importantly, I take conflicts of interest

01:43:06.440 very seriously. For all of my disclosures, the companies I invest in and or advise,

01:43:10.920 please visit peteratiamd.com forward slash about.

The Peter Attia Drive - April 01, 2019

#47 – Matthew Walker, Ph.D., on sleep – Part I of III： Dangers of poor sleep, Alzheimer's risk, mental health, memory consolidation, and more.

Episode Stats

Summary

Transcript