The Peter Attia Drive - #48 – Matthew Walker, Ph.D., on sleep – Part II of III： Heart disease, cancer, sexual function, and the causes of sleep disruption (and tips to correct it)

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

00:00:23.600 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

00:00:33.000 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

00:00:45.980 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

00:00:56.460 most about professionally are how to live longer and how to live better. I have a complete fascination

00:01:02.780 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 quality of

00:01:12.560 their lives. Curating and sharing this knowledge is not easy, and even before starting the podcast,

00:01:18.200 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

00:03:19.840 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

00:03:30.780 deals possible on the products that I love. And as I said, we're not taking ad dollars from anyone,

00:03:35.620 but instead what I'd like to do is work with companies who make the products that I already love and would

00:03:42.900 have them pass savings on to you. Again, the podcast will remain free to all, but my hope is that many

00:03:50.860 of you will find enough value in one, the podcast itself, and two, the additional content exclusive

00:03:57.900 for members to support us at a level that makes sense for you. I want to thank you for taking a moment

00:04:02.960 to listen to this. If you learn from and find value in the content I produce, please consider supporting

00:04:08.900 us directly by signing up for a monthly subscription. Welcome back to the three-part series on sleep

00:04:14.700 with Professor Matthew Walker, Professor of Neuroscience and Psychology at the University

00:04:18.860 of California, Berkeley, and the founder and director of the Center for Human Sleep Science. Matthew earned

00:04:23.480 his undergraduate degree and PhD in neurophysiology in London and subsequently became a professor of

00:04:29.060 psychiatry at Harvard Medical School before moving to Berkeley. His research examines the impact of

00:04:33.340 sleep on human brain function in healthy and diseased populations. To date, he has published over a hundred

00:04:39.320 scientific studies. He has received numerous funding awards from the National Science Foundation, National

00:04:44.420 Institutes of Health, and he's a fellow with the National Academy of Sciences. He's the author of the

00:04:48.940 international bestseller, Why We Sleep, which also happens to be the favorite book of my not-yet-two-year-old

00:04:55.140 son. He holds many patents covering various consumer-based sleep recordings, sleep tracking, and sleep

00:05:01.320 simulation. He's a sleep scientist at Google, where he helps the scientific exploration of sleep in

00:05:05.980 health and disease. He is also an enormous fan of Formula One and my hero, Ayrton Senna. And in the third

00:05:13.660 part of this installment, we actually spend quite a bit of time discussing this. In the second part of this

00:05:19.840 series, we discuss sleep and cardiovascular disease, sleep and how it affects diet, sleep and

00:05:25.380 reproductive function, the risk of deliberate sleep deprivations, sleep and cancer, the lack of

00:05:31.420 sleep within the medical profession, school start times and sleep deprivation, sleep patterns and

00:05:37.200 light, the biochemistry of sleep and naps, the efficacy of various medications on sleep. As a reminder,

00:05:44.600 at the end of this series, we're going to take questions for a follow-up AMA with Matthew. Asking

00:05:49.880 questions on the AMA forum and listening to the AMA podcast with Matthew, along with all AMAs,

00:05:54.760 will only be available to subscribers. So if you haven't signed up yet, you can do so now at

00:05:58.780 peteratiamd.com forward slash subscribe. So without further delay, here is part two of three with

00:06:05.220 Professor Matthew Walker. Okay, there's like literally a hundred more things I want to ask you

00:06:12.100 about. So can you make the case for why cardiovascular disease is worsened by or accelerated by sleep

00:06:19.740 deprivation? And then same question I'm going to ask you in a moment for cancer. Just pick the best,

00:06:23.980 like, because again, you could write a book on each of those topics.

00:06:26.560 And we can go into sort of the end of the book. But let me just give you one example for cardiovascular

00:06:29.960 disease. There is a global experiment that is performed on 1.6 billion people across 70 countries

00:06:37.240 twice a year. And it's called daylight savings time. And what we've seen is that in the spring,

00:06:43.680 when we lose an hour of sleep, there is a subsequent 24% relative increase risk for heart attacks that

00:06:51.000 following day, 24%. In the fall, when we gain an hour of sleep opportunity, there is a 21% reduction

00:06:59.280 in heart attacks that follow. How long has that been known? I mean, I remember the first time I saw

00:07:04.240 that I was like, that can't be right. And you know, when you see it bi-directionally like that,

00:07:08.800 it's very different. You know, that seems like a very robust manipulation. That's been known for

00:07:14.480 probably about 10 years. What's also interesting, and I won't get away from cardiovascular disease,

00:07:19.880 don't worry. But just simply to mention that when you look at that same transition,

00:07:23.800 you see the same profile for car accidents. You see the same profile for suicidal attempts

00:07:29.800 and suicide completion as well. What's also interesting, by the way, is that you see it

00:07:34.940 in terms of federal judges handing out harsher sentences. They hand out harsher sentences on the

00:07:44.300 Monday after that time change in the spring because they're more moody, emotionally irrational,

00:07:50.820 and less sort of, you know, empathetically sensitive because of that one hour of lost sleep.

00:07:55.260 And in the fall, more lenient sentences. So I can-

00:07:59.600 And does this effect last for two or three days?

00:08:02.180 So there is a blast radius to it that you can see it, that it's better but still worse in the spring,

00:08:08.360 which is where you lose that hour of sleep. The effect is still worse on the second day. And

00:08:14.840 it's almost recovered by the third day and finally comes back to what looks like a baseline by about

00:08:20.340 day four. So there's a blast radius. And this is, you know-

00:08:23.500 You know why I find this interesting, by the way?

00:08:24.880 One hour of sleep.

00:08:25.400 Yeah. And here's the other thing I always found amazing about that statistic.

00:08:28.560 In the Northern Hemisphere, you would predict the opposite if sleep were irrelevant, right?

00:08:34.720 Wouldn't you think that gaining that hour of sunlight in the day, in the spring, just the

00:08:41.380 overall anticipation, I mean, there's virtually no one in the Northern Hemisphere that isn't happier

00:08:45.960 in April than they are in November, right? So it's almost like you have all of that positive stuff

00:08:52.660 working against you and you still see the signal you just described.

00:08:56.820 Yeah. So in the face-

00:08:57.620 It's like a kick in the face. 0.97

00:08:58.980 Yeah. So in the face of birds chirping, sunlight streaming, temperature warming,

00:09:04.720 that one hour of lost sleep will still put you back on your derriere rather quickly in terms of

00:09:10.480 all of these facets. So there's one example for cardiovascular disease. Another one, there was

00:09:15.540 a fascinating study where they took a group of otherwise healthy middle-aged adults who had no sign

00:09:21.780 of coronary artery disease, and then they tracked them for five years. And then they looked at how

00:09:27.860 much sleep that they were getting. So again, hands up. This is essentially an associational,

00:09:33.040 longitudinal prospect of study, and you can't derive causality from it. But what you can certainly

00:09:38.620 say is when they started, they had no signs of calcification of the coronary artery. Those

00:09:45.700 people in that study at the end of the five years who were getting five hours of sleep or less had a

00:09:51.600 two to 300% increased risk of calcification of the coronary artery, which is the main corridor of

00:09:58.180 life for your heart. If you have a massive coronary, essentially, when you hear that colloquially,

00:10:03.360 that's essentially what's happening there. So that you could have that calcium buildup on the basis of

00:10:10.820 being bucketed into insufficient sleep. That tells me that it's not that when you are insufficiently slept,

00:10:18.700 you also are someone who has calcified arteries. It is saying that if you are insufficiently sleeping,

00:10:27.700 you are increasing your risk for developing that condition. This is about the development of that.

00:10:33.440 Now, how hard is it to control that for some obvious things that would track with that? For

00:10:38.320 example, shift workers or people of lower socioeconomic status who are working three jobs and

00:10:44.120 all the other things that can work against you.

00:10:45.800 So there are exclusion, there's some great exclusion criteria in that study where they prevent those

00:10:50.980 participants from entering the city. They also then added in other factors that you could imagine

00:10:55.980 would lead to that. So they controlled for those things like exercise, BMI, neck circumference,

00:11:03.480 smoking. They even include history of snoring. So they tried to take sleep apnea out of the equation

00:11:09.400 and still that relative risk was significantly larger.

00:11:13.840 What do you think is the mechanism?

00:11:15.640 I think the mechanism is probably several fold. I think the first thing that we see when we

00:11:21.840 undersleep people is that they become much more sympathetically nervous system driven. Now that

00:11:28.120 to some people may sound, oh, that sounds like a good thing. No, you have two branches of your

00:11:33.600 automatic nervous system.

00:11:34.880 I swear for a moment, I thought you would say good knowing that the listener would understand that

00:11:40.780 they're sympathetic and parasympathetic, but thinking you were, you were going to make a comment about

00:11:45.080 being more high strung is better because, but no, you actually were making a joke. Yeah, I love it.

00:11:49.960 I mean, I think you're so well tuned to this.

00:11:52.400 I think that's the beauty. My IQ is so low. I'm such a simpleton that I can always revert to the

00:11:58.760 naive state and I don't have to make assumptions. That's the beauty of my idiocy. But no, sympathetic 1.00

00:12:04.680 nervous system being sort of cranked on your sympathetic nervous system is not a good thing.

00:12:10.040 Your sympathetic nervous system essentially is your fight or flight branch of the nervous system.

00:12:14.120 And what we find is that as soon as you start to undersleep an individual, that fight or flight

00:12:20.300 branch of the nervous system starts to ratchet up. When that increases, you start to see, or perhaps

00:12:27.060 the reason why it increases is that you get a greater amount of adrenaline release. You get a high spiking

00:12:34.780 in levels of cortisol. You get a blunting in growth hormone. And I think probably just the cortisol and

00:12:43.360 the growth hormone alone may set you on a path towards cardiovascular disease, because we know

00:12:49.260 that those are two factors that lead into that sort of some of that atherosclerotic sort of equation.

00:12:55.100 And that brings it back to Alzheimer's disease, which is even if you just look at the epidemiologic

00:12:59.580 data, the signal's quite large on the benefits of IGF and growth hormone more than any other disease

00:13:06.540 in protection from neurodegeneration. So that may be, even in addition to everything you talked about

00:13:12.360 with respect to clearance, you're simply taking away neurotrophic factors that are essential.

00:13:17.760 There's some data, and I don't know how well replicated it is. I just read it in one city, where if you look

00:13:22.780 if you're ApoE4, so this is in terms of your genes, there are some genes that can predispose you to

00:13:28.480 Alzheimer's disease. On this podcast, there are wonderful descriptions of going into all of the

00:13:33.440 details of these, but ApoE4 significantly increases your risk for the development of Alzheimer's disease,

00:13:39.660 it appears. But what's interesting is that if you are ApoE4, but you are normotensive, you don't

00:13:48.000 necessarily have an increased risk of Alzheimer's disease. If you are ApoE4, but you are hypertensive,

00:13:54.500 then your Alzheimer's disease risk is far higher. So in other words, there seems to be an interaction,

00:14:00.760 a gene by cardiovascular disease interaction that leads that genotype to predispose you to Alzheimer's disease.

00:14:09.660 And therefore, you know, if you are under sleeping, you are putting yourself on a path towards

00:14:17.500 many new different factors that we are learning regarding cardiovascular disease, you know,

00:14:23.360 calcification of the arteries is one of them. But we also just see, you know, blood pressure,

00:14:27.920 you know, spike, we can take someone who is in a lovely state of, you know, normotensive profile,

00:14:34.640 and has been, and after either one night of total sleep deprivation, or one night of short sleep,

00:14:40.420 you know, we can almost start to see it after about reducing sleep by three hours, take someone down

00:14:45.580 down to five hours for just one night, you immediately see this spike in the fight or flight

00:14:51.620 nervous system, blood pressure goes up, you start to see cortisol increase, heart rate starts to

00:14:58.580 increase as well. You know, it's almost as though you've just got a beautiful car engine, you've put

00:15:05.300 it in neutral, and you've put your foot on the gas pedal, and you're just revving the living daylights

00:15:10.820 out of that engine in a fight or flight state. Now, if you do that chronically, which is what most

00:15:16.940 people do with insufficient sleep, you know, day after day, week after week, year after year, decade

00:15:22.360 after decade, it's not surprising that, you know, just revving the daylights out of that engine, at some

00:15:28.300 point, gaskets are going to start blowing, it's not designed to operate in that high revving state.

00:15:34.320 Now, high revving occasionally, just fine. And if you have a beautiful, let's say, you know,

00:15:41.320 Mezga Porsche engine, which loves to go to nine grand, and you and I going...

00:15:45.760 Of course, as long as it's in the GT3.

00:15:47.160 I had to geek out on cars for a little bit, because I'm just so obsessed, I had to get it in there.

00:15:51.300 But even then, you just can't stay at that RPM, that high RPM for very long. You just know it from

00:15:57.480 listening to the engine, you just know that mechanical badness is happening in neutral when you've got your

00:16:02.740 foot on the gas pedal. Well, that's the same way with chronic sleep deprivation in your nervous

00:16:07.360 system. So I think many of the cardiovascular effects, in fact, I think we're writing a paper

00:16:14.480 right now, I think if there is one central common pathway through which we can understand almost

00:16:21.240 all aspects of the deleterious impact of insufficient sleep, it is through the autonomic nervous system,

00:16:28.360 and specifically an excessive leaning on the fight or flight branch of the nervous system,

00:16:33.920 which is to say that your sympathovagal balance is way off, and you are far more in that fight or

00:16:41.180 flight sort of aversive state. So I think that to me is where that disease pathology starts and

00:16:48.320 perhaps progresses from that point. And I think only adding to that, which is probably what you

00:16:54.140 described is sufficient alone. But when you look at what I think are pretty well documented

00:17:00.580 appetitive changes in people under sleep deprivation, and more importantly, the inferior fuel partitioning

00:17:08.580 that follows, so the hyperinsulinemia, the impaired glucose disposal, the tendency to probably eat more 0.85

00:17:14.800 crap. I mean, I'm at my worst when I am sleep deprived. It is, you know, I just, it's so hard to

00:17:21.980 avoid junk food. Whereas probably the single greatest tool in my arsenal to eat well is to sleep well.

00:17:31.800 I mean, it doesn't sound like that's an obvious thing, but it's amazing.

00:17:36.120 When you read the data, I mean, it's, it's striking. And, you know, we can get into that,

00:17:40.320 which is essentially what you've just described is the, the energy balance of an organism of a human

00:17:45.480 being. And both the regulation of energy, once you've taken it in, and also the input of the

00:17:53.880 energy or, and the selection of the energy, how do you eat and how much do you eat? And then secondly,

00:18:00.360 once you've eaten, what does your body do with that food in terms of a basic kind of blood sugar

00:18:05.860 control?

00:18:07.160 And then the fourth thing is where do you access it? So I'm sure, I mean, I'm sure this has been done.

00:18:11.440 I just haven't seen it. But if you look at respiratory quotient of people sleep deprived

00:18:15.740 versus not, I'm sure RQ must be going up, right? I'm sure they're absolutely going after the wrong

00:18:21.300 fuel partitioning.

00:18:22.380 That's exactly what you see.

00:18:23.520 I'll do what you do well, but then I always forget to do. Let me explain why that matters.

00:18:27.980 Respiratory quotient going up implies that under the same level of exertion, you would preferentially

00:18:33.520 go after glycogen for ATP as opposed to fat. And so not only is that not what most people want

00:18:40.120 because they'd prefer to burn their fat than burn their glucose, it signals a metabolic inefficiency

00:18:46.400 which lays at the heart of all of this stuff.

00:18:49.880 And a beautiful example of that is a great study that looked at the efficiency of dieting when you

00:18:57.260 are underslept. And effectively what they found is that your diet is all for nothing if you're not

00:19:04.120 sleeping well. Because what they found is that when you are underslept, defined as sleeping six

00:19:09.620 hours or less, 70% of the weight that you lose will come from lean muscle mass and not fat.

00:19:18.860 In other words-

00:19:19.280 Which is perfect when your cortisol is high.

00:19:21.400 I mean, that's exactly what cortisol is.

00:19:23.280 So that's exactly what you'll see with cortisol. So in other words, your body becomes immensely

00:19:28.380 stingy at giving up its fat. Your body will ruthlessly hold onto its fat when you are underslept

00:19:35.320 and not give it away. Instead it will-

00:19:38.020 So when you are underslept and you're losing weight, you're losing the thing that you want

00:19:42.660 to keep, which is beautiful muscle definition. And you're holding onto the one thing you're trying

00:19:47.360 to get rid of, which is the blubbery fat.

00:19:49.360 And it's really everything. It's the cortisol is working against you. Insulin will work against

00:19:54.320 you. You're going to have more hyperinsulinemia. You will have more catecholamines working against

00:19:58.340 you.

00:19:59.200 Yep. We see that with norepinephrine in particular.

00:20:01.860 Oh, and testosterone will go down. That's going to work against adiposity, getting rid of it.

00:20:06.880 Yeah. I should note, by the way, that men, and I think we'll come onto this later, but men who are

00:20:12.760 sleeping just five to six hours a night will have a level of testosterone, which is that of someone

00:20:17.360 10 years they're senior. So insufficient sleep will age a man by a decade in terms of that critical

00:20:24.380 aspect of wellness and virility. So if you want to sort of be very bravado about insufficient sleep,

00:20:30.100 be careful. I would also say that, by the way, we found it simply associational, but men who are

00:20:35.060 sleeping just five hours a night have significantly smaller testicles than those men who are sleeping 0.92

00:20:39.800 seven hours or more. 0.89

00:20:40.960 And I'm sure that in that you would see FSH and LH must be lower, right?

00:20:45.120 So we do. So, you know, I'm not trying to make necessarily funny, even though sometimes I'll

00:20:49.060 start talks that way just to sort of get the bravado folks out the way. But you also firstly

00:20:55.000 see that you men who are sleeping six hours or less will have fewer sperm. Those sperm will have

00:21:00.980 more deformities, but this is not limited.

00:21:03.980 You say deformities, even genetics. So even aneuploidy in the split.

00:21:06.280 That's correct. Yep. And you can see that in rats too, if you do those studies and those

00:21:10.100 studies have been done with short sleeping rats, but it also transfers to women that you see that 0.61

00:21:15.660 women who are sleeping just five to six hours a night will have about a 20% reduction in FSH, 0.90

00:21:22.200 which is called follicular stimulating hormone, which essentially is a critical part of the

00:21:29.780 reproductive pathway in terms of getting pregnant. Women who are sleeping that little too, 1.00

00:21:35.720 typically have about a 30% higher rate of abnormal menstrual cycles. So if you put together a couple

00:21:44.140 that is trying to conceive, and that couple is on six hours of sleep, you've got a man who's down on

00:21:49.760 testosterone, down on the amount of sperm that they're producing, the motility of those sperm are

00:21:55.800 reduced. And then the woman you're down on, you know, you've got erratic menstrual cycles, 0.74

00:22:00.140 you've got FSH that's down to, from a reproductive standpoint, this is devastation. 0.99

00:22:06.140 And again, evolutionarily isn't that surprising. It's basically mother nature saying,

00:22:10.840 if the conditions are not optimized for you to sleep, it's probably not optimal for us to

00:22:15.340 propagate the species at this moment. And that is probably something that rarely happens.

00:22:20.660 What happens to FGF 21? Do we know?

00:22:22.840 I was about to say, we don't know. I should say, I don't know of any studies that have looked at that.

00:22:29.280 FGF 21 has such an interesting, it's one of the few places in the brain where you'll see a difference

00:22:34.300 between men and women under a fasted state. So under a fasted state, you don't appear to see the same

00:22:40.080 hit in reproductive fitness of a man, but you do see it in the woman. There's actually an anatomic 0.98

00:22:46.920 reason for this that I, at this moment, am forgetting, but it basically has to do with this

00:22:52.000 rise of FGF 21. And you'll see basically the inhibition of FSH in the women without it in the

00:22:59.040 men, which again is like, it's this question of if, if food is scarce, we really don't want a mother 0.99

00:23:06.240 to try to nurture another body.

00:23:09.620 Yeah. Maybe that's the biological warning beacon that just says time out on, you know,

00:23:15.520 reproductive capacity right now.

00:23:18.040 I always get asked, is a ketogenic diet in any way good, bad, or indifferent towards fertility?

00:23:24.040 And my answer is I actually have no clue, but just on first principles, it would seem that a

00:23:30.260 ketogenic diet could have one negative side effect on the woman, which is exactly that because a 0.92

00:23:36.360 ketogenic diet generally does put up FGF 21. It could be offset by benefits. So if you took

00:23:41.160 someone who's metabolically ill, who is fixed by the ketogenic diet, it could offset it, but all

00:23:46.160 things equal. Yeah. You know, I certainly wouldn't recommend fasting for a woman who's trying to 1.00

00:23:51.520 conceive, but you've had the luxury of spending a lifetime thinking about this. I'm sort of playing

00:23:55.800 catch up over the past couple of years.

00:23:58.120 By the way, it's one of the things that you do brilliantly, by the way, from what I've, I listened

00:24:02.540 to on every one of your podcasts. But, but what I would say in response to your question, though,

00:24:07.320 about that evolutionary question, the reason I suspect it rarely occurs, however, is the following.

00:24:14.400 Human beings are the only species that deliberately deprive themselves of sleep for no apparent reason.

00:24:20.520 You almost never see other organisms undergo sleep deprivation. The only time that you see another

00:24:26.880 species seemingly restrict its sleep is under two conditions. The first is under certain, with certain

00:24:36.060 mammals, after they've given birth, they will deprive themselves of sleep to care for their young. The

00:24:42.900 most obvious example there is in killer whales, where the mother will break from what's called the pod, 0.97

00:24:47.900 which is the main group. They will go away and they will give birth and then they will bring that calf

00:24:53.580 back to the main pod. And during that time, they will short sleep. The second is under conditions

00:25:00.360 of starvation. That when an animal starts to become starved, they will, there is a wake promoting

00:25:08.020 signal where you will actually stay awake longer and it gives the animal the ability to forage in a

00:25:14.460 larger perimeter because presumably the current perimeter is not containing sufficient food. And I think

00:25:21.660 that that's the reason why some people will say when I undergo, you know, either time restricted

00:25:25.920 feeding or especially when I undergo fasting, the reason that I feel like I don't need to sleep as

00:25:31.460 much or, you know, I, and I feel more awake even though I'm sleeping less and it's great. I actually

00:25:37.980 would be very cautious about doing that for long periods of time and doing it sort of repetitively for

00:25:43.040 long periods of time is because there are still, you're still suffering the decrements of insufficient

00:25:47.600 sleep. But the reason that you feel awake is because there is a biological mechanism in place

00:25:53.460 that says, Oh my goodness, if you're not getting food, we need to keep you awake for a little longer.

00:25:59.180 We're going to sacrifice this thing called sleep for a little bit so that you can go and find more

00:26:04.560 food. So there's a wake promoting mechanism and we know how that works. But the more important point

00:26:10.020 here being made, I think for me is that because human beings are the only species that deliberately

00:26:15.420 deprive itself of sleep and no other species has really done that. It means that mother nature

00:26:20.900 throughout the course of evolution has never had to face the challenge of sleep deprivation.

00:26:26.060 So in other words, mother nature has never come up with any safety nets for any of our major

00:26:32.740 biological systems, metabolic, reproductive, cardiovascular, immune, mental health. None of these things

00:26:40.940 have any safe holding. They don't have any sort of crutches when sleep goes away. Why? Because

00:26:49.000 mother nature has never been asked to solve that question. Now I'll give you an example.

00:26:53.220 I use the same argument by the way, for sitting in chairs, you know, or, or, or, or so many of these

00:26:59.000 things that we just do that. We, you could argue if our species could survive another hundred million

00:27:08.140 years, maybe sitting around doing nothing, drinking soda, we'd, we'd, we'd figure out a way to make

00:27:15.160 that not so harmful, you know, assuming it interfered with, you know, our capacity to reproduce. If it

00:27:20.400 didn't, then maybe evolution wouldn't take any interest in it. But I mean, I think it's, it's the

00:27:26.440 reason for me, at least why an alternative example there in the pro would be, you know, an adipose cell

00:27:33.480 that mother nature throughout the course of our life and most species experienced times when there

00:27:40.500 was feast and times when there was famine. And so mother nature faced with that challenge of famine

00:27:46.880 came up with a solution called the fat cell so that we could store caloric credit and then spend it when

00:27:54.340 we needed to, when we went into famine, where is the adipose cell for sleep? Where is the fat cell for

00:28:00.260 sleep? You know, wouldn't it be lovely if we could store up sleep credit and then spend it? And the

00:28:06.540 answer is there is no such thing because there it's, there has never been the challenge in the course of

00:28:11.660 evolution to come up with that solution. And because hibernation is not an example, right?

00:28:15.900 No, hibernation is the opposite. Hibernation basically says it's not giving you credit to sleep less in

00:28:21.380 the summer. It's just taking stock of the fact that there really is nothing for you to eat in the winter.

00:28:26.200 Correct. And you, that's, that's really about energy expenditure and essentially energy preservation

00:28:32.360 and torpor and hibernation of different states to, to sleep itself. But that's always, I think,

00:28:39.060 being something I've thought a lot about that example of, you know, the fat cell versus sleep.

00:28:43.660 And it's the reason why people unfortunately think that sleep may be like the bank, that you can

00:28:50.220 accumulate a debt during the week and then you can just sleep it off during the weekend.

00:28:54.720 What's the name again for that phenomenon of the sort of Monday through Friday short sleeper who

00:28:59.900 binge sleeps on the weekend? What's it called? Social jet lag. Social jet lag. Yeah. Where you

00:29:04.140 sort of essentially, it's what I used to term very politically incorrectly as sleep bulimia,

00:29:10.060 which was where you would essentially, you know, purge during the week and then you just binge during

00:29:15.140 the week. It's this real binge purge kind of abstinence syndrome. The reason it's problematic is not

00:29:20.540 just because sleep doesn't work like that. And the studies show that that's very deleterious to health

00:29:25.000 is because it also is terrible torture on your circadian biology. Because what happens is that

00:29:31.820 you start to wake up, you know, you'll be going to bed at one, sort of one or, you know, midnight rather

00:29:37.540 than nine. And then waking up at 10, 11, you know, on a Saturday and a Sunday, and then come Sunday

00:29:43.720 night. You've got to drag your body clock back by three hours, you know, to get into bed and you

00:29:50.340 repeat it the next weekend. That's essentially like saying to your biology, and it's no different.

00:29:56.220 I'm going to fly back and forth from San Francisco to New York every weekend, three hour social jet lag

00:30:02.760 difference. That's a hazard ratio impact on your circadian biology.

00:30:08.620 And as I'm listening to you say this, I'm making the stronger and stronger case for my move to

00:30:14.980 Austin, which in part is driven by just shortening the trips that I have to take, you know, to be able

00:30:21.320 to only have to take a one hour time zone delta or, you know, never basically having to take more than

00:30:26.740 a two hour time zone delta.

00:30:29.180 For your lifespan, healthspan, for the impact that you can have on this planet and this society

00:30:38.300 during the time that you're here. And for the preservation of you in your children's life,

00:30:44.940 I would strongly recommend it.

00:30:46.960 I'll share with you an interesting example of a patient of mine who has an example of what happens

00:30:51.040 when that system, you know, quote unquote breaks. So this gentleman, I'll try to be as vague as

00:30:55.420 possible. So I don't identify him, even if he's listening, not that he would care if I told the

00:30:59.260 story anonymously, basically worked in the finance sector, but at some point was living on the West

00:31:05.100 Coast of the United States, but basically was exclusively involved in, you know, Eastern

00:31:10.320 European markets. So you can imagine what time he had to be waking up to deal with that.

00:31:15.960 After several years, and he in part moved from the West Coast to the East Coast to try to make that gap

00:31:22.340 smaller. But he got to the point where he could never sleep past 3 a.m. It was metaphysically

00:31:29.600 impossible, regardless of what time, if he went to bed early, if he went to bed late, he was up at 3 a.m.

00:31:34.480 And it turned out one of my colleagues, Vic Jane, who's a Stanford trained sleep doc at

00:31:41.240 Northwestern and also someone I just can't wait to interview because he's just so insightful on

00:31:46.040 sleep. You know, he basically did something that you and I had talked about earlier, which was

00:31:50.720 reverse use of melatonin in blue light to phase shift him into the correct time zone. And what humbled

00:31:58.360 me about the story was how long it took to move. And it just gives you that sense of, could you

00:32:06.460 imagine forcing that on your system once a week? And I mean, it took months to correctly.

00:32:13.680 Of course, correct.

00:32:14.020 Of course, correct. Yeah. Now, admittedly, it had been in place for a couple of years by the time,

00:32:19.200 you know, I met him. So you know, you have a little bit more to overcome. But nevertheless,

00:32:23.640 I was blown away at the lengths that Vic had to go to drag him forward by three hours.

00:32:31.740 For that biology to have been tortured for that long, it's going to take a reset of, you know,

00:32:39.320 non-trivial magnitude and help from exogenous agents like melatonin, like blue light, regular

00:32:46.640 exercise, exposure to daylight, you know, even manipulations of body temperature it could take.

00:32:52.200 So there are a whole arsenal of tools in the sleep doc kind of box that we can use to sort of

00:32:58.880 make these manipulations either be it for people with insomnia, be it people with anxiety, be it

00:33:04.520 people with sort of jet lag or what's called circadian phase delay, which is essentially what

00:33:10.620 he was suffering from, which was...

00:33:12.660 You mentioned temperature. This was something I didn't know until I learned it through this patient's

00:33:17.520 case and through Vic's learning, which was about two hours before the wake up, you see this drop

00:33:22.900 in temperature. And like that drop in this guy is occurring at 1am. He's getting that drop up. And

00:33:29.880 then... And so you... What you basically were doing was dragging that thing to, you know, 4am. You had to

00:33:36.380 drag the temperature drop to 4am. And again, I think it just speaks to the complexity of the situation.

00:33:43.680 And as a general rule, when you mess with really complex, hardwired, evolutionarily sound things,

00:33:51.300 you've got to be careful. You're going to pay the fiddler.

00:33:53.240 When you fight biology, and sleep is one of the most conserved behaviors across all living organisms

00:34:00.060 that we've observed. When you fight that kind of innate, hard-grained biology, you normally lose.

00:34:07.080 And the way that you know that you've lost is disease and sickness, either acute or chronic.

00:34:13.340 You know, at some point, a lack of sleep will get you. It may be in your late stages of life.

00:34:19.060 You know, Margaret Thatcher, Ronald Reagan examples with Alzheimer's disease could be from a heart attack

00:34:24.700 prematurely when you're 56 years old. It could be a car crash at any moment in time because of a

00:34:30.640 microsleep. You will be popped out of the gene pool pretty quickly.

00:34:33.200 So let's talk about one more thing, which I'll, we could talk about cardiovascular disease for much

00:34:39.120 longer, but let's talk about cancer. And let's talk about, when I talk about these diseases with

00:34:44.400 patients, one of the things I try to do is say, let's reverse engineer what we think is happening

00:34:48.860 when this disease goes wrong and then back things out. So when it comes to cancer, I say, look,

00:34:53.260 cancer has a bunch of things that have to go wrong. First, you have to have a genetic insult.

00:34:57.640 Two, it has to be missed by the immune system. And fortunately, most of the time,

00:35:01.920 our T cells can figure out that cancer is non-self, et cetera, et cetera. And you kind of walk through

00:35:06.920 all of these things. So let's talk about the immune system for a moment, because we know that if you're

00:35:13.580 not, well, I want to let you explain that. Yeah. I'll let you go from there because I was going to,

00:35:18.100 I was just about to go off on this topic, but tell me about how sleep impacts the immune system

00:35:21.920 and not only how that might impact getting common colds, but how it could impact cancer.

00:35:26.120 So firstly, what we know is that there are now significant links epidemiologically between sleep

00:35:32.340 and cancer of a variety of forms. Currently, that list includes cancer of the bowel, cancer of the

00:35:37.580 prostate, and cancer of the breast. And then we can step down. Which are basically three of the top four.

00:35:43.200 Exactly. Yeah. Three of the heavy hitters. Next, you can sort of say, okay, what about the causal

00:35:48.980 evidence? Well, firstly, I would say before I describe the causal evidence, that causal evidence is now

00:35:54.100 so strong that recently the World Health Organization decided to classify any form of nighttime shift

00:36:00.580 work as a probable carcinogen. Their words, not mine. And the proof of evidence that is required by

00:36:08.840 the World Health Organization to make such a statement usually has to be astronomical. And that

00:36:15.320 data is now, I think, very well put in place. I'll just mention one quick causal manipulation study done by

00:36:23.040 a colleague down at UCLA. They took a group of healthy adults, and they limited them to just

00:36:29.240 four hours of sleep for one single night. And then they looked at a set of cells called natural killer

00:36:35.900 cells. And you can think of natural killer cells like the secret service agents of the immune system,

00:36:41.680 in that they're pretty good at identifying dangerous foreign elements, one of which are malignant 1.00

00:36:47.300 cancerous cells. They'll inject some things into them and try and destroy them, essentially. 0.87

00:36:53.060 So what you want is a pretty virile set of those immune assassins circulating in your body. And what

00:36:59.620 they found is that one night of four hours of sleep reduction led to a 70% drop in natural killer cell

00:37:06.780 activity. That's quite a surprising state of immune deficiency that has happened within one night.

00:37:14.080 So you can step and repeat that and imagine what would be the state of your immune system, particularly

00:37:19.860 for those critical anti-cancer fighting immune cells, after several weeks, if not months of insufficient

00:37:26.620 sleep. That's one aspect, which is, what is your vulnerability to developing cancer? Because, you know,

00:37:33.980 many of us are, you know, will have cancer cells emerging in our body every day. So we need those aspects of

00:37:41.280 our immune system to prevent those cells from becoming the disease that we call cancer.

00:37:47.060 I mean, that's such an important point. And I'm probably biased because I trained in an immunotherapy

00:37:51.880 lab, but I don't think people necessarily appreciate that we pretty much always have cancer and our immune

00:37:59.580 system is pretty much always protecting us. It's actually the exception when the cancer develops into

00:38:05.920 something clinically. Correct. And, you know, I've debated, not debated is the wrong question. I, one of

00:38:10.500 my favorite sort of sitting around dinner discussions with, with cancer biologists is what is the greater

00:38:17.200 driving force for the obvious age association of cancer? In other words, why does cancer increase

00:38:24.980 non-linearly as you age? And I offer two hypotheses. Is it three actually, is it an increase in the rate of

00:38:34.560 mutagenesis? Do you, do you experience on a per unit of time basis, a greater insult to the genome?

00:38:40.740 Two, if you assume that that's the same, is it that over time, the accumulation and the expression,

00:38:48.360 the phenotype of that becomes more problematic? And, or three, is it that our immune system,

00:38:55.080 specifically the adaptive immune system is weakening and the balance starts tipping in favor of cancer.

00:39:00.780 And, you know, I've asked this question of Nobel laureates and future Nobel laureates. Every one

00:39:06.580 of them has said, we don't know. They suspect it's all of the above, but they all agree that the

00:39:14.600 weakening of the adaptive immune system is almost assuredly playing a role in why we get cancer more

00:39:22.560 as we age, because we know many things are working against us when we do that. And so when I hear you say

00:39:28.240 NK cells, which, you know, the CD8 cell, the NK cell, the CD4 cell, these are,

00:39:33.840 these are your green berets and your Navy seals of the war against cancer. If you put a hit on those

00:39:40.580 guys, you, you know, you could be taking a 10 year step in the wrong direction, right? Just as you

00:39:46.460 talked about the gonadal 10 year leap, right? Yeah. Testosterone. I mean, I would love to add

00:39:51.560 a 3B hypothesis there, which is that one of the most dramatic changes with age and the most

00:40:00.240 sizable and robust physiological changes in with age that we see is that your sleep gets worse.

00:40:08.460 And sleep is probably one of the most powerful regulators of your immune system. You know,

00:40:15.280 if you want a full arsenal and you want every single weapon in there to be sharp and ready to

00:40:20.700 annihilate, sleep is what you need. That data, I think is very clear in terms of the decimation of

00:40:27.420 your immune system with sleep. And that, you know, I just gave one example there too. And the funny

00:40:32.340 thing is like you give that example and it's not even that dramatic. Like there's no person listening

00:40:36.980 to this who hasn't had a four hour sleep night. That's part and parcel for just being a human being

00:40:41.860 in the civilized world, maybe even in the non-civilized world, right? And yet to think about

00:40:47.080 a lifetime of stacking those things and the compounding effect of what does it mean to get

00:40:54.180 eight hours a night versus 6.2 to five hours a night. Imagine that. I'm almost amazed it doesn't

00:41:00.720 kill us quicker because to your point, we have adapted a great system to cope with excess nutrients,

00:41:07.280 right? We had a great system and it takes a long time for that to break us down. I think in some ways

00:41:14.760 we're probably so naive in our ability to measure the short-term consequences of sleep. I mean, not

00:41:20.840 you, of course, because this is what you're thinking about day and night. But I think as a medical

00:41:24.560 community, we're really shitty at knowing how to measure short-term, like what's really happening

00:41:31.680 in sleep deprivation. Especially in, you know, in clinics and in healthcare systems. Yeah, like in

00:41:36.240 schools. Like we don't really understand how bad sleep deprivation is on a learning child. We don't,

00:41:42.640 we clearly don't understand what it's doing on the road. Like we don't, we don't understand that

00:41:46.700 this will kill you much faster than, you know, bad nutrition, right? Which is clearly going to kill 0.62

00:41:53.160 you. You know, if you look at all cause mortality and short sleep, it's not even linear, it's

00:41:57.920 exponential. You know, it really is sleep will bend that curve of your lifespan in a downward

00:42:04.740 direction with a dart into the ground when it starts to get short. I think your argument is so great

00:42:10.840 that it really comes down to the fact that at least we had a system in place to train us for

00:42:16.640 excess nutrients. Now, look, we can argue we didn't have a system to train us to consume refined

00:42:21.480 carbohydrates. We didn't have a system that trained to consume massive amounts of polyunsaturated fats

00:42:27.660 or high, high amounts of sucrose. But we still knew how to consume some of these things. And there is a

00:42:35.880 dampening effect in nutrition that the, you know, ever present adipocyte can numb and at least for a

00:42:42.960 while blunt that system. But you're right. If we didn't, if it wasn't really until what, probably

00:42:48.740 200 years ago that sleep deprivation could have become common. I mean, when do you really think

00:42:53.580 was the tipping point? I know that in the forties we could compare the forties to today, but it would

00:42:57.640 almost seem like the light bulb was a pretty big step in the wrong direction.

00:43:01.240 Natrified. Not to blame Edison.

00:43:03.720 Yeah, but you could go back and you could argue that, you know, Edison with the light bulb and his

00:43:08.760 company may have been the starting point. I think it was probably happening even earlier on the basis

00:43:15.460 of social demands. You know, the industrial era, I think then started to compound things. Once we

00:43:22.040 switch from an agrarian society to an industrial society, in my mind, that's when stuff really started

00:43:27.800 I mean, my history is horrible, but that's about 250 years. Yeah, exactly. Yeah. So I think we've been

00:43:32.500 that that curve has started to happen. So it's an evolutionary millisecond. Correct. Correct. Sounds

00:43:38.060 like a long time. Blink of an eye. The other aspect of, you know, cancer is not just that you increase

00:43:45.400 your risk for developing cancer because you weaken the immune system components that are there to combat

00:43:51.400 against all of those, you know, carcinogenic influences cellularly that you've just described.

00:43:57.440 But another study by a colleague at the University of Chicago, David Gozal, he looks at the relationship

00:44:03.320 between sleep loss and cancer in mice. And I'll just give you one example of a study he did.

00:44:07.360 Took a group of mice, inoculated them with some cancer cells on their back, and then gave that cancer

00:44:12.040 a one month period to grow. At the end of the month, he resected the skin, measured the size of the tumor.

00:44:17.160 Half of those mice were allowed to sleep normally. The other half had their sleep restricted. So they

00:44:23.320 just had their sleep kind of top and tail, not total deprivation, just limiting their sleep in the

00:44:27.660 morning and the evening a little bit. What they found is that at the end of that one month, those

00:44:34.860 underslept mice, when they looked at them, the tumor was 200% larger. I mean, it was physically

00:44:42.640 distorting the body. And if you were to see these pictures, you know, you would just think,

00:44:48.960 my goodness, you know, I can see a small, tiny little sort of dot that is the growth of the

00:44:54.460 cancer in the well-slept mice. You know, the others, it just looked like a hideous, you know,

00:44:59.980 mass on these underslept mice. Secondly, what they found was that that cancer in those underslept

00:45:06.620 mice had actually metastasized, which is just, you know, in some ways, a fancy way of saying that it

00:45:11.680 breached the original origin and started to invade other organs, bone, as well as brain.

00:45:17.400 And when cancer becomes metastatic, you know, that's when we know things can get really dark

00:45:23.840 and grim in terms of life expectancy. There are many mechanisms that you could

00:45:27.360 generate or hypothesize, could explain that. I'm curious as to how many there were. So one would,

00:45:33.540 I think, be exactly what we've described, which is this immune weakening. But you could also look at,

00:45:38.500 so hypercortisolemia is going to drive hepatic glucose output. That's going to create hyper

00:45:42.880 and sleep. Did they, when you look at those.

00:45:44.820 They adrenalectomized the mice and then they fed them prior to it. So they controlled the stress.

00:45:50.080 They controlled the metabolic response.

00:45:51.440 Correct. Yeah. And what they did find in terms of, they looked a little bit at the mechanism. They

00:45:56.160 found that macrophage M1 cells, those were actually down-regulated by a lack of sleep. And what

00:46:03.220 was up-regulated was a sort of a rogue version, which was called the M2 cells, which seemed to

00:46:08.340 have a tumor promotion component to them as well.

00:46:11.240 Wow. So, so they really, but the, the adrenalectomy is brilliant. What a, what a clever design.

00:46:16.600 So you have to take out the stress response and they limited that.

00:46:19.400 Imagine if you didn't.

00:46:20.540 Right. And that's the fear is how much worse would it be when it really happens in humans,

00:46:26.140 because you will have that plus this huge tsunami of a cortisol impact, hypercortisol impact,

00:46:34.200 which is only going to make things worse.

00:46:36.320 So bring this now back to the clinical tragedy here, right? Which is again,

00:46:39.980 the weakness of our profession. And I say ours meaning mine and not yours. You take a patient

00:46:44.640 who's got a diagnosis of cancer. Do you think anxiety is going to go up? Hell yes. Right. Do you

00:46:49.200 think their sleep is going to deteriorate on no basis other than the fact that they have this

00:46:52.800 diagnosis and they're often undergoing horrible treatment? Right. It's amazing to me that I can't

00:47:00.040 imagine how many oncologists are thinking through this problem, right? Which is as careful as we are

00:47:05.820 about crafting what the chemotherapy regimen looks like, what the exact, if you go to ASCO, right,

00:47:11.920 if you go to the largest oncology meeting, I don't know how many papers are being addressed on this

00:47:16.880 topic. Do you? No, I, and I think there are, they are few and far between with regards to sleep.

00:47:21.440 Whereas if you, if you think about some of the stuff we machinate over in oncology,

00:47:25.360 like exactly who gets the sentinel node biopsy versus this. And well, what if this is an ER PR

00:47:30.860 positive breast cancer versus an ER positive PR negative, or like we could, we could noodle and,

00:47:37.040 and machinate on the most minute details, which, you know, maybe they matter, maybe they don't.

00:47:42.440 And yet something like this seems so obvious and yet, you know, we just seem ill-equipped to deal

00:47:48.440 with it. Like it, it's almost like you'd say- And it doesn't surprise me either, by the way,

00:47:51.760 it's not your fault as a medical profession, you know, and being part of that profession,

00:47:56.900 you know, what I did was I looked at a retrospective and I looked at medical curriculums across,

00:48:02.200 you know, the top 20 first world countries practicing medicine. And what you find is that on

00:48:06.980 average, most doctors get about one to two hours of education regarding sleep. So one to two hours

00:48:14.840 regarding a third of the patient's lives, that is so anemic in terms of meeting the needs.

00:48:24.340 But the critical part is that one third of their life spent sleeping also makes such a huge difference

00:48:31.760 to their two thirds of their life awake. That is just, to me, unacceptable.

00:48:37.500 And there's a deeper problem, which is in training to be a part of that profession,

00:48:42.640 you enter a culture that is as difficult as the one you described where, you know, heads of state

00:48:50.260 are chest pounding about how little they sleep. I mean, I'll just pause for a moment to share another

00:48:55.640 sort of set of stories. So when I started my residency, we were on call every third night. I mean,

00:49:01.880 every third night, you know, I think my record, I think the most sleep I ever got on call in five

00:49:07.840 years was five hours. And that happened once. And it wasn't, of course, straight five hours, but I

00:49:12.360 had accumulated five hours of sleep on a call night. So I think you would average probably two to two

00:49:17.060 and a half hours of sleep every third night, but that's average. So guess what? Lots of those times

00:49:22.600 are 30 minutes or less. So you're on call every third night. So it's my second month on the job.

00:49:28.980 So the way it works is you're basically either on call, post call, or pre-call. Those would be

00:49:33.300 your designation. So I come in on a Monday. I am pre-call. So it's my day before my call day. So you,

00:49:41.960 you know, you come in at five in the morning, you would normally leave by about eight o'clock at night

00:49:45.340 or seven o'clock at night. And as we're finishing our rounds that evening, someone, a surgical

00:49:50.880 resident had failed to show for the ER shift. He got sick. So the chief resident said, you know,

00:49:57.960 Hey, we need somebody to stick around tonight to cover the ER. So I volunteered because I want to

00:50:02.700 show how tough I am. Like, of course I can do this. So I stick around. I do the ER shift that night.

00:50:08.380 And of course in the ER, you're absolutely not, you don't get one minute of sleep, obviously.

00:50:13.260 Now it's Tuesday. Now I'm on call. So I'm awake, of course, all day Tuesday and Tuesday night

00:50:19.180 and Wednesday. And I probably slept an hour Tuesday into Wednesday. Now it's Wednesday. I get to go

00:50:25.980 home early, which means 6 PM on Wednesday. So I've been up since Monday morning. So this is not like

00:50:31.780 within the realm of the type of sleep deprivation that's going to kill you. There are lots of people

00:50:37.240 in the life. I'll come back to that. Yeah. Yeah. No, it almost did. Actually. Yeah. But yeah.

00:50:40.340 Okay. So I'm driving home. My point in case. Yeah. So I'm driving home. So I have quite a long drive

00:50:47.160 until I have to get to the freeway. It's about two miles, but it's like stop signs and streetlights.

00:50:52.620 And then I have to get to the freeway. And then I've got kind of a hike home at this point in time.

00:50:56.160 I live kind of North of the hospital. I noticed something really odd. Every time I come to a

00:51:03.000 stoplight or stop sign, I can't keep my foot on the clutch because for some reason, I just,

00:51:08.140 I have a habit of like keeping my foot on the clutch and keeping it in first, as opposed to just

00:51:11.680 sitting in neutral, which whatever. So the point is I keep stalling and I'm thinking,

00:51:15.480 God, I've never done this in my life. Like, why can't I even sit here for 27 seconds and maintain

00:51:23.440 my motor control over this thing? So I met about the last street, maybe, maybe two streets away from

00:51:30.140 before I have to get on the freeway for a 15 minute drive. And in what I can only describe as

00:51:35.900 one of the best, dumbest decisions of my life, I decide I freaking cannot get on this freeway. 0.99

00:51:43.740 I'm going to die. So I pull over to the side of this road, Eastern Avenue, and there's this park 0.96

00:51:50.600 called Patterson Park. I have no idea what Patterson Park is like today, but at the time I

00:51:55.440 knew nothing about it other than it was grassy. It turned out at the time it was like, you know,

00:52:00.780 an open air drug market. And, and this is now speaks to the second point. So I had really good

00:52:07.280 judgment, which was don't get on the freeway. But then I exercised comically bad judgment,

00:52:13.440 which is I'm going to go take a nap in the park. So I get out of my car, probably didn't even lock

00:52:20.840 it, walked into the park, put my pager on my neck of my scrubs, set the alarm for one hour. It's like

00:52:27.960 6 p.m. Wake up at 2 a.m. in the middle of the park with rat bites on my arms. I mean, this is like

00:52:38.180 the drug infested, rat infested park. And just think to myself, what in the hell just happened? 0.65

00:52:44.440 I had about three of those in residency. The other thing that happened once in residency,

00:52:48.980 same situation, every third night call, but I had to take an extra shift in there. And to be clear,

00:52:56.280 I volunteered, right? It wasn't like no one put a gun to my head and said, you have to do this. I

00:53:00.200 volunteered for someone who couldn't take a night of call. So I basically was on the same thing,

00:53:06.420 but now I'm way further into my residency. You know, this was in my like fifth year of training.

00:53:12.740 And I think the accumulation of that was devastating. And so now it's two in the morning on that third

00:53:19.380 night. And we're operating on a patient. And the way it works in residency is when you're a senior

00:53:27.580 enough resident, you're the operating surgeon, meaning you're the one doing the case. The

00:53:31.420 attending surgeon is assisting you. And this is a very trivial operation we were doing. The only

00:53:37.260 reason, in fact, we did it in the middle of the night is just to get the OR time. Like it wasn't

00:53:40.940 even an emergent case. It certainly could have waited a day. It was just removing the gallbladder

00:53:45.320 laparoscopically. So I'm holding the two devices in the trocars and I'm doing the case and the

00:53:52.040 attending is just retracting for me. And I don't know how it happened, but I fell asleep on the

00:53:58.900 patient and face planted directly onto the patient. And again, the only solace-

00:54:06.400 Can you imagine the amount of sleep pressure that was necessary to build up for you to, I mean-

00:54:13.900 And I remembered that I had had a number of head drops prior.

00:54:17.620 Exactly. Yeah. Everyone's like jamming to good music, but no, they're not. It's just like they're

00:54:22.460 falling asleep.

00:54:23.360 Right, right. And, you know, after God knows how many of those, it just turned into bang,

00:54:29.140 your full L position face first into the patient. Again, the only fortunate thing there is in the

00:54:35.220 process, I didn't jerk my hand and tear the common bile duct or, you know, an artery that could kill a

00:54:40.780 patient. So what happened? Well, the attending surgeon said, dude, you're too tired. Let me do

00:54:46.540 the case. You come and hold this. Like it wasn't even like, it was just sort of like this happens

00:54:51.980 and you go stand over there now. Whereas you think about like if that had happened in an airplane or

00:54:57.440 in some other profession where you'd like, that's a four alarm fire. That's a, what's going on here?

00:55:03.420 How can this happen?

00:55:04.560 It is. And the data that we've got is pretty strong. You know, I've lobbied the, the medical

00:55:10.460 association on a number of times. And I just recently wrote a piece in a journal called the

00:55:14.340 Lancet, which was called a sleep prescription for medicine. And I lay out the case for why both for

00:55:20.380 patient and doctor sleep is, you know, utterly essential, but profoundly absent. And firstly,

00:55:28.580 what we know regarding, you know, your surgical experience in that, that story there, we know

00:55:33.100 that if an attending surgeon has slept only six hours in the previous 24, there is 170% higher

00:55:41.980 likelihood that they'll make a major surgical error, such as, you know, splicing some kind

00:55:49.100 of major vascular component of the patient. What's also interesting is that when, if a resident

00:55:56.580 has worked a 30 hour shift relative to when they've worked just a 16 hour shift, that resident

00:56:02.960 after 30 hours of being awake, when they get back in their car and start to drive home, just as you

00:56:07.700 did, there is a 178% increased likelihood that they'll get into a car crash, end up back in the

00:56:15.860 same emergency room from which they just came. But now as a patient requiring ER treatment, because

00:56:22.780 they got into a car crash, the irony is lost on me there. The other point is that we know that one

00:56:28.540 in five medical residents will make a serious medical error during their residency caused by

00:56:35.020 insufficient sleep. One in 20 medical residents will kill a patient because of insufficient sleep.

00:56:41.020 Now, when you realize that there's well over 20,000 medical residents active right now, and you think

00:56:48.060 about that statistic, one out of 20, and you were to then do the math and lay out the number of bodies

00:56:53.580 in front of you that were preventable deaths, preventable by way of this simple thing called

00:57:01.500 insufficient sleep. I think it is an absolute disgrace. You know, doctors working in the ER,

00:57:09.180 if the residents, if they've worked a 30 hour shift, 460%, I think it is more diagnostic errors,

00:57:16.780 which just stuns me in terms of a number. But what's also interesting is that when you

00:57:23.100 lobby the medical organization, and I've tried to do this a while back, and I lay out that compassionate

00:57:29.580 case of medical errors, impact on patients, impact on doctors themselves, there was incredible pushback

00:57:36.300 and resistance, which was my mind is made up, don't confuse me with the facts. And I want to just,

00:57:41.820 you know, have a mea culpa because there were few people that were more vocal, more vocally opposed

00:57:48.700 in our residency than me when this 80 hour work week came in. And my view was not that sleep was, you know,

00:57:57.260 good and we needed more of it. I really didn't have a point of view on that. I actually frankly

00:58:00.780 hadn't reflected on that because I obviously didn't know any of the data you shared.

00:58:03.980 My view was the conditions under which we work are horrible. And that's a selection feature.

00:58:12.140 You want people that are willing to opt into that, who are willing to take that much pain

00:58:17.100 to do this. And I had this whole thesis, which I have no idea if it turned out to be true or not,

00:58:22.300 that if you lowered the bar, lowered the pain, you would bring in a broader demographic. Now,

00:58:29.820 regardless of whether that's true or not, it doesn't really matter in the face of this data.

00:58:35.980 In other words, if it is true that back in the day, you were just training better physicians

00:58:42.620 because you were attracting people that could persevere more or had whatever, you know, make

00:58:46.940 up your thing, it still doesn't change the fact you've got to come up with a new screen for that.

00:58:51.100 Yeah, you can't get around that evidence of the deleterious impact of even if you attract,

00:58:56.380 even if it's just a selection bias that you're bringing in these hard knuckle folks who can just

00:59:02.620 jam through with wakefulness in ways that, you know, really are just horrific for everyone involved.

00:59:10.220 You still can't get around the fact that they're not going to be performing medicine at its highest.

00:59:16.380 And we know for a fact that you don't need to. So I think the only piece of evidence that

00:59:22.140 favors that type of, you know, 30 hour continuous residency type of mentality,

00:59:28.140 the only good argument I've heard is continuity of care, that once you go down to a shorter amount

00:59:34.700 of time, you lose continuity of care with the patient that you're cycling doctors between patients.

00:59:39.900 But I have to say though, even that I don't believe because when I looked at systematically

00:59:45.740 at the other medical practices, there are places that train wonderful physicians that have a maximum

00:59:53.020 of somewhere between 12 to 16 hours, places like New Zealand, Sweden, and France. They all limit their

01:00:00.300 junior doctors to training on nothing more than those amount of hours. And if you then look at the

01:00:07.180 ranking of the medical systems in terms of how well they're doing, those guys are way up there at the

01:00:13.900 top. So I find it difficult to see a case for arguing for the ability to train doctors with insufficient sleep.

01:00:25.340 I'll tell you what I think is going on because I agree. I mean, I haven't, I've been away from it for

01:00:29.980 so long, but I've spoken to a few people who have kept up with the sort of medical education literature.

01:00:36.140 And it turns out the 80 hour work week has not solved the problems that they had wanted.

01:00:40.700 I don't believe that medical errors have necessarily gone down and you could argue,

01:00:43.820 well, 80 is not low enough, but I also think a more subtle part of it is in the countries that

01:00:49.180 you mentioned, I suspect that it is the cultural norm of not just the residents, but the people who

01:00:55.500 train those residents because they were trained in that sort of more humane way that you bring a

01:01:01.580 different way of thinking about handing off a patient. It means something different because

01:01:06.940 it's just so ingrained in that's how medicine is practiced. And I think what happened in the United

01:01:11.100 States, and again, this is wild speculation. I don't follow this literature, but just based on

01:01:16.940 what I've seen since, I suspect that when you overnight created a new rule, which I think took

01:01:23.020 place like July 1st, 2003, new rule, right? Overnight. But none of the people that were

01:01:29.340 leading that had trained under those conditions. They didn't necessarily know how to teach under

01:01:35.180 those conditions. They didn't understand, as we said, like, what does it mean to hand off a patient

01:01:39.820 after 16 hours? Because in our day you were responsible for everything. I mean, you wouldn't

01:01:47.820 dream of doing something like that because you just don't know how to communicate what needs to

01:01:53.900 be done to the next person. I mean, so much of it just comes out of that. So I suspect that's a big

01:01:58.380 part of it. And of course, unfortunately, that would suggest that this is going to take time,

01:02:03.820 but it still has to change. It's time, but I fear that type of mentality is going to die one generation

01:02:10.540 at a time. Because in part, it's a bit of an old boys network, I think, some of it. But you mentioned

01:02:16.780 those things about how do we understand the informational transfer, the propagation from

01:02:21.660 one doctor to the next regarding continuity of the patient who remains. And I think that really is

01:02:27.420 an important point. Part of me, though, feels as though we've been able to put people on the moon.

01:02:34.720 And that was non-trivial, but we solved it. I think we can also probably solve continuity of care

01:02:40.900 and protect both patient and doctor alike in terms of their well-being and still do it. So I think it's,

01:02:46.520 you know, I don't mean to sort of trivialize the problem. And I think what I want to try and do

01:02:51.800 here is, you know, just raise my hands and say that, you know, I really appreciate how hard that

01:02:58.000 is. It's just like early school start times and the work that I'm doing there to try and lobby for

01:03:03.040 kids to start later. That's not a small thing either, because you've got to solve multiple problems.

01:03:08.580 You've got to solve bus unions, bus transportation. You've got parents who need to be in work at a certain

01:03:14.320 time point. And, you know, starting those times later is going to be non-trivial. All of these

01:03:20.320 things are hard problems. And it's not as though I'm either ignorant of those, and I want to really

01:03:26.820 recognize how hard they are. But what I would also love to do is to try and just make sure that we don't

01:03:34.860 lose sight of the end outcome here. The end outcome in terms of education is our children

01:03:42.440 and making sure that we are not essentially educating our children amnesic by way of their

01:03:49.000 insufficient sleep. And the same is true for our patients and our doctors. Let's keep the target

01:03:55.920 that we have in our crosshair at the end of this process in mind, rather than get perhaps distracted by

01:04:02.540 how difficult the problem is. It is difficult. And I'm so sympathetic to that. I know it's hard,

01:04:09.120 but there's so much on the line. There is too much on the line for us to be swayed by the difficulty

01:04:16.680 of the problem in terms of the importance of the solution that we must reach.

01:04:21.460 No, no, no. I don't disagree with anything you've said. I think what I'm trying to do is provide

01:04:24.940 an explanation for why, and it's not an excuse, more of an explanation for why I think the experiment

01:04:31.140 has failed here. And it probably speaks to the need for a greater buy-in to the gravity of the

01:04:39.520 problem. And therefore, a greater emphasis on solving what I agree are quite, in the grand

01:04:45.200 scheme of things, trivial problems. I mean, splitting the atom was hard. Putting a man on the moon,

01:04:49.480 that was hard. Eradicating slavery, that was hard. All of these things have been done.

01:04:54.220 But I think we will look back, you know, with shame, just like you were describing to your

01:04:58.600 family regarding sleep and insufficient sleep. As we look back now with smoking, you know,

01:05:04.460 50 years ago, we will look back in probably 20 years with shame that we were having schools start

01:05:13.080 at 7.30 in the morning. By the way, for schools that start at 7.30 in the morning, buses will start

01:05:19.320 leaving at 5.30 in the morning. That means some kids are having to wake up at 5.15,

01:05:24.700 5 o'clock or even earlier, which in my mind is lunacy. And I think the shame will be present

01:05:31.680 when we consider the impact and the impact we already know from the studies there. When we delay

01:05:36.960 school start times, what we see is that academic grades improve. We see truancy rates decrease.

01:05:43.840 We see behavioral problems and psychiatric problems also decrease. You see the number of

01:05:51.220 people who drop out of school and drop out of certain classes also decrease. And then finally,

01:05:57.800 shockingly, the life expectancy of students increases. And the reason that it increases

01:06:02.840 is because of car crashes. There's a great study in Teton County in Wyoming. They shifted their school

01:06:09.000 start times from 7.35 in the morning to 8.55 in the morning. And then they looked to see in just

01:06:14.820 this narrow age range of just 16 to 18 years old, what was the change in car accidents? And what they

01:06:21.080 found in that following year, not just that the kids reported getting one hour of extra sleep,

01:06:26.600 there was a 70% drop in car crashes that following year. Now, the advent of ABS technology in cars,

01:06:35.020 anti-lock brake systems, that dropped accident rates by 20 to 25%. And it was deemed a revolution

01:06:40.380 by some people. Well, here is a biological feature, getting enough sleep, that will drop accident rates

01:06:46.640 by up to 70%. So when sleep is abundant, minds will flourish. And if our goal as educators truly is to

01:06:58.040 educate and not risk lives in the process, then I fear that we are failing our children in a quite

01:07:04.940 spectacular manner with this incessant model of early school start times.

01:07:09.040 So say a little bit more on this. You and I have talked about this in the past, which is,

01:07:12.800 and I think I even posed the question to you this way, which was, you know, all things equal,

01:07:16.880 Matt, if you're trying to pick between three schools for your kids, like, would you weigh start time in

01:07:23.720 the matrix of decision? And you, you know, your answer was emphatically yes. Why is this the case?

01:07:29.980 Why is it that do kids just, is there something in their brain as they're developing that it's just a

01:07:35.900 question of they need more sleep and therefore you'll pick it up in the morning? Or is there

01:07:39.300 something about they need to sleep to a later time that won't be solved by putting them to bed earlier?

01:07:46.080 All of the above. So firstly, you know, kids, even when you're 16, 17, 18, you're still needing nine

01:07:55.420 to 10 hours of sleep because your brain doesn't finish developing until it's about 25.

01:08:00.200 What percentage of 18 year olds in the US do we believe are getting that sleep? Or if it's easier

01:08:06.520 to answer the question, what do we believe is the median duration of sleep for an 18 year old?

01:08:10.080 About 11%.

01:08:11.080 11% are achieving the required.

01:08:14.220 89% are not getting sufficient sleep.

01:08:18.140 And do we have a sense of what that average is? I think the, the numbers you quoted earlier,

01:08:21.360 I assume were for adults or was that for all comers?

01:08:23.720 No, that was for adults. For teenagers, it's probably, it's down to about seven hours when

01:08:29.640 it should be somewhere between nine to 10 hours. So the relative deficit is bigger than it is in

01:08:37.900 adults. What's also striking by the way is if you look at parents and you ask them, do you think

01:08:43.120 your teenager is getting sufficient sleep? 72% of them will say, yes, I think my teenager is getting

01:08:49.900 the sleep that they need. Yet only 11% of them are getting the necessary sleep. So in other words,

01:08:56.360 there is a mismatch here between the parental and child sleep equation. And what that also leads to

01:09:04.220 then because parents believe that their kids are getting enough sleep. There is a parent to child

01:09:09.520 transmission of sleep neglect. And what I mean by that is, you know, the pulling the covers off at

01:09:14.220 the weekend when these kids are sleeping in for two reasons. Firstly, they're sleeping in because

01:09:19.600 naturally their biological circadian rhythm moves forward in time. So they, they want to go to bed

01:09:26.600 later and wake up later. It's not their choice. They don't get a choice in that. It's biological.

01:09:32.540 It's hardwired. So for a parent listening, what would you say to them if they're wrestling with

01:09:38.340 their kid's sleep schedule? What do we think is the most natural time to bed, time to rise for a 16,

01:09:46.760 17, 18 year old? I think for, you know, 16, 17, you know, you're looking at a 10 hour period,

01:09:53.240 probably from, you know, somewhere between 10 to midnight, depending on their chronotype

01:09:59.020 to then sleeping in until, you know, essentially somewhere between 8.30 to 10.30 the following

01:10:05.840 morning. Now, the reason that that is actually still too early though, is because that would be

01:10:13.000 what would naturally happen if you were to let them sleep like that every single night of the seven days

01:10:18.860 of the week. But we don't because what we do is during the five days of the school week,

01:10:23.940 we are getting them up way too early. You can even put them to bed at, you know, 9 or 10 and say

01:10:30.600 sleep. They can't biologically. So they won't be sleeping as well during the week. So at the weekend,

01:10:38.160 they're trying to sleep off a chronic debt that we've saddled them with during the week due to

01:10:44.620 early school start time. So no wonder they've got such a sleep pressure that is forcing them to try

01:10:49.960 and sleep until like one or two in the afternoon. So then we chastise them for saying, you're lazy,

01:10:56.960 get out of bed, you're wasting the day, but it's not their fault. Firstly, because their biological

01:11:00.960 rhythms, their circadian rhythm wants them to be asleep late and into the morning and into the

01:11:05.900 early afternoon. And secondly, they've got this huge debt of sleep that they're having a rebound from

01:11:12.700 in terms of sleeping, trying to sleep off that debt that the school systems have given them as well.

01:11:17.900 So to me, I think we just need to, you know, reformulate our notion of how important sleep

01:11:25.440 is in that context. And you've already alluded to this. That's a hard one to solve, right? Because

01:11:30.160 you're restructuring when a parent goes to work, et cetera. It is non-trivial. So what would be a

01:11:34.740 middle ground that would be a great compromise? In other words, like if a school started at 9am,

01:11:41.040 I feel like when I was in high school, we started at 9am. Well, ironically, yeah, what's happened is that

01:11:45.640 back in the 60s, schools were starting around nine o'clock. And then evermore, as we marched on in

01:11:52.540 terms of our quote unquote development in society, in the US, that time has actually gone back and back

01:11:58.800 to earlier and earlier start times, I think, because it's been squeezed by the sort of the vice grips of,

01:12:06.180 you know, work schedules where parents are having to work longer hours. They are ultimately commuting

01:12:13.780 for longer. So they have to leave the house ever earlier. So they have to put the kids in school

01:12:19.340 ever earlier. And I think many of these kids are just sitting there unabsorbent, like a waterlogged

01:12:28.820 sponge. They're not going to be taking up information. And in fact, if you look at the data regarding

01:12:34.880 delayed school start times, and you ask, yes, overall GPA, SAT scores, all of them rocket up. They all

01:12:42.460 improve when you delay school start times. But which classes get the biggest benefit in terms of

01:12:49.440 the grade improvement? It tends to be not the classes in the afternoon, where they are finally

01:12:55.120 awake because of their circadian rhythm. It's the classes in the morning, where before those classes

01:13:00.720 would be starting at, you know, eight o'clock, but now are starting at 9.30. When you push them to 9.30,

01:13:07.060 they're doing much better in those classes. And it reaffirms the case that the hit that is going on

01:13:14.400 in terms of the amnesic impact that early school start times are having is really in those morning

01:13:20.220 hours when the brain is not designed to be awake. And it's certainly not designed to be receiving an

01:13:25.940 education. It's designed to be asleep, preparing itself for its education, which should really start

01:13:33.860 at maybe nine or 10 o'clock in the morning. I mean, the United Kingdom right now, as we speak at the

01:13:38.260 time of this podcast, there is a bill that's going through that will lobby for a 10 a.m. start time,

01:13:44.540 which I think is probably the sweet spot. And the reality of it is, because someone will say,

01:13:49.880 well, great, what do you want kids to stay in school till 6 p.m. and blah, blah, blah, blah. But

01:13:53.360 my guess is you could probably do less with more if you did it right. In other words, you wouldn't need

01:13:58.340 to spend as much time in school if you could cram. You could more efficiently learn that

01:14:04.660 information. And we've done these studies. You absolutely do learn more efficiently when you

01:14:10.040 have had sufficient sleep. So it's a forcing function. Could you compress it? It's like

01:14:14.740 zipping a file. The amount of information that can be stored is much greater. But what's also

01:14:20.520 interesting, let's say that that's not the case. And you say, look, to do all of this staggered system

01:14:25.940 with the bus unions and to make it work out with parents and work schedules, it's going to require

01:14:31.480 more money to figure this out. Well, it turns out that some of that can be cost savings at the back

01:14:38.360 end because kids get released later. There is a time during when kids get kicked out of school

01:14:45.360 if they start early, which is this kind of criminal bewitching hour, which is where the kids are out of

01:14:51.420 school, but the parents are not home yet from work. And when they get together, sometimes for

01:14:56.340 some kids, bad things happen. That's where you see a lot of juvenile crime happen, which is in that

01:15:02.440 sort of twilight hour between, you know, the 3 to 4 p.m. period when they're out of school to the 6 to

01:15:09.800 7 p.m. period where the parents are finally there and they get home. If you look at the cost of that

01:15:15.740 criminal impact of activity in those hours and then you say, if we were to start schools later,

01:15:23.020 kids get out later, we'd limit the window of criminal opportunity. The cost savings comes

01:15:30.340 back around and pays for itself for delayed school start times. So you just need to open up the aperture

01:15:36.420 of your memory, think about the problem, and ultimately you can solve these equations.

01:15:41.620 My daughter hates when I say this, but I keep saying to her that I think this whole summer

01:15:45.840 vacation thing is ridiculous because, you know, it's sort of an artifact of a world when you

01:15:51.400 needed the kids to sort of work in the farm in the summer. But it almost seems like having a shorter

01:15:56.140 day, shorter time in school, more time. In other words, uniform the situation, right? More time to do

01:16:02.180 extracurricular things, more time to sleep, but just go to school 12 months out of the year. Like go to

01:16:07.920 school like we work, right? That seems to be also culturally a very difficult solution, but

01:16:13.580 conceptually an easier way to approach this as well. I mean, I love your idea and I do think that

01:16:19.060 the data supports it empirically right now, which is that you may not necessarily need to just simply

01:16:24.300 take the school day and push it forward in time. You may just need to take the start of school and push

01:16:31.460 it forward in time and hold constant the end of the day. But by way of that sufficient sleep,

01:16:37.820 ultimately you may be able to still gain the same amount of intellectual equity within the minds of

01:16:44.540 those young individuals by way of that sort of later school start time and greater sleep as a

01:16:50.460 consequence. And it also just seems like something that is so amenable to testing. I mean, it's very easy

01:16:54.760 to do a randomized control trial of something like that. And there's lots of that stuff that's going on

01:16:58.860 right now. The data just keeps coming out and it goes in no other direction. It's so consistent.

01:17:04.640 It really is. Is there anything that we as parents can do outside of the obvious, which is choosing,

01:17:12.220 you know, if we have the luxury, right? A lot of times you don't have the luxury,

01:17:15.820 but if you have the luxury of choosing between schools that have, you know, meaning if you can choose

01:17:22.180 by where you live or if you're putting your kids into private schools or something, the later time,

01:17:27.040 let's assume you don't have that luxury and you've got this hand that's dealt to you, which is school

01:17:32.520 starts at 810 and, you know, gets out at whatever, 340. What can we do to help the kids get as much

01:17:41.600 sleep as possible, even if it means bucking against what might be their innate circadian rhythm?

01:17:47.820 I think probably right now, the best thing to do is to try and excise technology. And, you know,

01:17:56.420 I'm very... When you say that, can you, yeah, just say that again. That is such a, I mean,

01:18:02.260 that is such a hot topic right now, but it's hot. This is another discussion I had with my daughter,

01:18:07.500 which is I worry that in 20 years, we will look back at the idea of 10 year olds with iPhones as,

01:18:15.260 I worry we will look at it with even a worse lens than the smoking mother. And that's, there's two

01:18:22.680 folds, right? There's the technology piece, which we'll talk about. And then that says nothing about

01:18:26.740 the whole social piece of, you know, the influence of social media and stuff like that. But so sticking

01:18:31.800 to the first one, you take the technology away completely, you limit light after sundown.

01:18:38.060 So there's that component, which is, I think, just the basic impact of light. And this is scales

01:18:43.400 to not just children, scales to adults too, that we are a dark deprived society in this modern era.

01:18:49.480 And we need darkness at night to allow the release of a hormone called melatonin.

01:18:55.700 And that melatonin, as it rises, it will help time the onset of your sleep. And there are great

01:19:02.520 studies that have been done where if you use, let's say an iPad for an hour before bed,

01:19:08.140 first thing you get about a 50% drop in the amount of melatonin that's released. So you lose 50% of the

01:19:16.300 signal of sleep timing. Worse still, that melatonin peak, even though it's lower to 50% lower,

01:19:24.080 it will arrive three hours later at night. So it's so mistimed by the perverse impact of the screen.

01:19:34.680 So that would be like you and I living here in California, but our melatonin release is much

01:19:40.580 closer to Hawaii in terms of timing. That's one hour of iPad reading.

01:19:46.940 Is that true also, Matt, for television? Does television emit the same?

01:19:50.980 No, it doesn't. So television less so, but television has another mental impact, especially if you're

01:19:57.040 watching it in bed. That's not a good idea because then your brain associates your bedroom

01:20:03.000 as the place of being awake and watching television, not the place of sleep. And when you start to form

01:20:10.480 those maladaptive associations, it can be a trigger of insomnia and anxiety. And typically that's one of

01:20:16.460 the recommendations that we have, that you only use the bed if you're struggling with your sleep

01:20:21.100 for sleep and intimacy. That's it. But to come back to kids, that's the first impact,

01:20:27.940 which is that the light can disturb and disrupt their melatonin significantly. We also found in

01:20:34.240 those studies, by the way, that the one hour of iPad reading before bed reduced the amount of REM

01:20:39.160 sleep time. When people woke up the next day, you have them rate, how refreshed by your sleep do you

01:20:44.120 feel? Significantly lower rates of feeling restored and refreshed by their sleep. What was also interesting

01:20:50.300 is that once you stopped that iPad reading, it again had this carryover effect. It had this blast radius

01:20:56.500 where they kept sleeping poorly for another two to three days after that iPad reading.

01:21:03.300 The second component though is a mental component. I think it's the component that may be as if not

01:21:08.820 stronger than the biological component with melatonin, which is firstly that these devices that kids use

01:21:17.100 can cause sleep procrastination. What I mean by that is, and this is just for adults too, you can be there

01:21:23.380 with your phone or your iPad and you can be sleepy and you could fall asleep, no problem at all. But

01:21:28.500 because you got it there, you just think, oh, just check email one more time. Just check Facebook.

01:21:33.000 Oh, let me send that tweet out. And I forgot the detergent. Let me go onto Amazon very quickly and

01:21:38.420 I'll just order that. And then you look up and it's 40 minutes later and now you're 40 minutes short on

01:21:44.180 your sleep. Sleep procrastination is a problem. But for kids, I think the two bigger issues are waking up

01:21:52.080 in the middle of the night. There was a survey done that demonstrated that well over 80% of teenagers

01:21:58.200 admit to waking up during the night to check their phones and check social media.

01:22:04.800 I'm sorry, say that again. 80% during the week will wake up at least once to check their devices.

01:22:12.620 So you've got this dependency that is causing this alertness spike to wake you up. And that's a habit

01:22:21.360 that once it builds is quite difficult to break. The final thing, I think, even if you were to be

01:22:27.300 diligent and put your phone in airplane mode, what those technologies do is cause what's called

01:22:34.020 anticipatory anxiety. So I think many people have had that experience of having it like an early morning

01:22:39.960 flight and you've got to wake up at 530 in the morning and you know it. You set the alarm,

01:22:45.680 but you wake up at 528 and you are awake like a bolt. Matt, I'm glad you brought this up. I have

01:22:52.600 noticed this for my entire life. I sometimes will play the game of can you wake up at 528? I'm blown

01:23:01.500 away at the ability to do that. Now, you can't do it if you're really sleep deprived. At least I can't.

01:23:06.620 So it's not, I can't go to bed at 2am, three nights in a row. Yeah, you'll oversleep it if

01:23:10.860 you don't have the alarm. What is that telling us about? It's almost like I've got a CPU with a

01:23:15.980 clock in it. Correct. Like, no, an actual clock. No, correct. That's right. Because you can do it

01:23:20.880 to a wrong time. You can miss set the clock at the bed by 10 minutes and hit that time versus true

01:23:28.200 time. Right. It's crazy. And there's a great experiment that I'll tell you about. But that

01:23:33.180 anticipatory anxiety, that airplane example, that's kind of like the extreme version, but

01:23:38.680 a weaker version, but a chronic. And I think a very malign sort of version of it happens with our

01:23:45.340 phones because most people, the first thing that they do when they wake up in the morning

01:23:50.060 is swipe and they just unlock this world of anxiety that comes flooding in through their phone,

01:23:56.980 emails, texts, social media. You essentially are training your brain to anticipate that wave of

01:24:05.360 anxiety every morning. And what we've discovered is that when you embed that anticipatory expectation

01:24:13.260 in the morning, the amount of deep sleep that you get at night is less. You end up sleeping in a

01:24:20.660 shallow state and you don't get the same amount of deep sleep. And the greater the anxiety that there

01:24:28.200 is the next morning, the greater the reduction in deep sleep that you have the night before. That's

01:24:33.440 why you'll feel like you've had a pretty rough night of sleep when you're waking up for that early

01:24:38.580 morning flight because you've just had a poor quality of sleep. But the timing thing, so I think to come

01:24:44.860 back to make the point here so I don't forget, I think for kids, the best recommendation right now

01:24:51.040 is technology to try and limit it. And I know that's hard because the genie feels like it's out

01:24:56.400 the bottle and it's not going back in anytime soon. And I don't want to be puritanical. I don't want to

01:25:01.280 be puritanical about alcohol or caffeine, about CBD, THC. I don't want to be puritanical about kids

01:25:10.200 getting the sleep or timing of the sleep that they need or can't get. But what I would say is that

01:25:17.700 to try and maximize that sleep in the face of early school start times, which is what we're currently

01:25:23.160 facing right now, taking that technology out the bedroom is probably the single best thing that you

01:25:29.340 can try and do. The other thing that they recently found was a correlation between the use of or having

01:25:36.380 social media in the bedroom and fear of missing out. Because for kids so often, and I understand

01:25:42.560 this, you know, because we were both kids, you could remember that. Folks were going out, you know, in

01:25:47.500 the afternoon. You wanted to be there. You wanted to hang out. It was fear of missing out. It was FOMO.

01:25:52.540 Well, now FOMO plays out on social media and it plays out when you disengage from the parents. It plays

01:26:00.040 out, in other words, in those twilight hours. And so there is good data now showing that fear of

01:26:06.360 missing out is directly related to insufficient sleep at night in teenagers. The way that you can

01:26:12.160 try to help that is to try and limit the devices that give you the ability to have FOMO in the first

01:26:17.860 place. So, but in truth, I don't honestly have a good set of solutions right now for this teenage sleep

01:26:25.740 problem. I think it's going to have to be letting them sleep when biologically they naturally want to 0.95

01:26:30.540 sleep. Again, you fight that biology, it's not going to go your way. It's just not.

01:26:35.640 Well, Matt, that, that certainly, uh, will scare any parents senseless with respect to electronics.

01:26:41.240 You know, one other question on that, that my daughter actually asked me all the time,

01:26:44.440 if she's reading a book in bed with a, like a nightlight, is that as harmful a, from the

01:26:52.800 standpoint of light, which is actually the question she's asking. And now B would be my question more

01:26:57.740 broadly. We talked about not watching TV in bed because that's creating kind of a different

01:27:03.280 purpose. Is reading in bed a bad idea? And, or is the, you know, sort of natural white light that

01:27:10.140 not natural, but you know, the white light you get from a bulb, where does that rank on the

01:27:14.080 melatonin suppressing scale?

01:27:17.140 So if you're someone who is struggling with sleep, that you are someone who has sort of

01:27:21.960 insomnia, such that you are finding it difficult to either fall asleep or you wake up at night and

01:27:26.800 you can't get back asleep. So the first is what we call sleep onset insomnia. The other is called

01:27:32.240 sleep maintenance insomnia. You can fall asleep fine, but you can't maintain your sleep. The advice

01:27:37.100 there is even reading should be something that you should forego, that you should read in a different

01:27:42.280 room. Wait until you get really sleepy and then go to bed. That's the advice. And because otherwise,

01:27:50.120 as I mentioned, the brain is this incredibly associative device. And if you lie in bed awake,

01:27:56.360 your brain quickly learns that your bed is the place of wakefulness. I hear this from patients a

01:28:02.040 lot where they'll say, look, I'm sitting on the couch watching television and I'm falling asleep.

01:28:07.000 And then I get into bed and I'm wide awake and I don't know why. And the answer is because your brain

01:28:11.940 has learned the connection of your bedroom being a trigger for wakefulness because wakefulness is what

01:28:17.600 you do there. So what you need to do is break that association and get up after 20 minutes,

01:28:24.700 go to a different room in dim light, read a book elsewhere in dim light. And I'll come back to what

01:28:29.860 type of light in a second. And then only when you're sleepy and there's no time limit for this,

01:28:35.020 should you return to bed? And the analogy would be this, you would never sit at a dinner table waiting

01:28:40.900 to get hungry. So why do we lie in bed waiting to get sleepy? And you shouldn't. And we need to sort

01:28:50.080 of, you know, bring that back. So that's if you're struggling. In terms of light, what I would say is

01:28:55.020 be a bit mindful as to what kind of light it is. If it's an LED light, it's usually enriched in the blue

01:29:02.780 sort of low frequency of the visible light spectrum. And it's the blue light that is most harmful to

01:29:10.640 melatonin. So light in general, not great for melatonin. It will stamp the brakes on melatonin

01:29:17.100 and it will stop releasing it. So your brain is fooled into thinking it's still daytime, even though

01:29:21.900 it's actually on the clock face nighttime. But of that light, if there is a better form of light,

01:29:29.000 it's the red and the yellow, the very warm kind of colors, the cold blue light that comes from LED,

01:29:36.920 that's the more deleterious light. That's the type of light that is more detrimental. So I would say

01:29:44.320 that, you know, you can get these smart light bulbs now that can change the dynamic frequency range of

01:29:50.160 that light in the visible spectrum, make it more kind of warm and yellow in the evening, definitely favor

01:29:56.300 that and try to stay away from, you know, light that is LED light. You know, even if it's just going

01:30:02.460 back to a classic light bulb, which typically is warmer in color and low in wattage, that's your best

01:30:08.720 light for reading. Has the experiment been done when this might not even be ethical today, where subjects

01:30:14.460 are given no restriction on when to sleep, how long to sleep, but they are put in a perpetually light

01:30:20.680 environment? No, but the opposite. Well, I know of the opposite. Yeah. There's two. No, there's one

01:30:31.520 experiment that has kind of been done that is like that. And then there is the other opposite, which is

01:30:36.180 so the one experiment that's not dissimilar to that actually happened inadvertently, unfortunately,

01:30:42.020 which is with premature babies in the neonatal intensive care unit. What they used to do was they would

01:30:49.840 just leave these fluorescent, you know, light bulbs on in the NICU all of the time. And at that point,

01:30:59.620 even though the circadian rhythm isn't especially, you know, robust in infants, they're, you know, awake,

01:31:05.120 they're asleep, they're awake, they're asleep, they still need that signal of light and dark. And what

01:31:11.400 they found is that when they regularized light in the neonatal intensive care unit, in other words, when

01:31:17.140 they gave back darkness at night and gave strong light during the day, firstly, the infant started

01:31:23.520 to sleep better, you saw about a 50% improvement in oxygen saturation in those infants. And they put

01:31:31.560 more weight on within the time period because they were sleeping more regularly. And then finally, they

01:31:36.900 left the NICU somewhere between two to two and a half weeks earlier. So that's the only example I know

01:31:45.080 of where you take a condition where you were put under artificial continuous light conditions.

01:31:51.620 And then you do the experiment where you try to mimic more naturally what a 12-12 hour cycle of light

01:31:59.860 dark would be. And what you see are just biological changes, all for the better. That's the only

01:32:07.300 experiment that I know of in the positive, which is what you're describing.

01:32:10.960 The reverse experiment was done where they essentially took a group of people and they

01:32:17.480 said, you know, what time do you normally go to bed? And they said, I can't get sleepy until

01:32:21.880 about 11 p.m. And I sleep usually about six, six and a half hours. They took that group of

01:32:27.780 people, off they went to the Sierras here in America, this sort of beautiful mountain range

01:32:34.120 with no electricity whatsoever, no access to artificial light. And firstly, what was dramatic

01:32:41.580 was that these individuals started going to bed around nine o'clock in the evening. And this wasn't

01:32:48.400 necessarily just because they didn't have anything to do. They actually rated themselves as feeling

01:32:53.380 sleepier earlier. Why? Because they were getting the signal of darkness at the natural time.

01:32:59.580 Secondly, they went from being, you know, ardent about the fact that they were only six and a half

01:33:06.800 hour sleepers and that's all that they needed to then sleeping a little bit more than nine hours.

01:33:13.100 Now, I think this point about going to bed at 9 p.m. is a really important one. Have you ever

01:33:20.280 thought about what the term midnight actually means? It means, it means middle of the solar cycle. It

01:33:28.260 means middle of the night. Whereas now with the perversity of the industrialized civilization,

01:33:35.800 midnight, and especially now in the digital revolution, means it's maybe the last time to,

01:33:41.660 you know, send a few emails. Whereas if you look at hunter-gatherer tribes, whose way of life hasn't

01:33:47.520 changed for thousands of years, and you ask, how do they sleep? As if they are a good indicator. And we

01:33:53.140 think that that's probably a good indicator of how hominids should sleep. They typically go to bed 1.00

01:33:58.620 about an hour and a half to two hours after sundown. So around sort of 8.30, 9 o'clock.

01:34:04.380 They usually get about seven, seven and a half hours of sleep at night. And then they have a siesta

01:34:10.500 like nap in the afternoon to make it up to around about a total of nine hours. So I think what has

01:34:17.280 happened with modernity is that we have changed both the amount of sleep that we're getting.

01:34:22.120 We've reduced it. We've changed the timing of sleep. So we're getting less sleep and we're

01:34:28.260 getting it at a different time, quote unquote, perhaps a wrong time of night. And finally,

01:34:34.380 the nature in which we obtain our sleep has been changed. It seems as though we are designed perhaps

01:34:41.660 to be sleeping biphasically, which is a long bout at night and then a siesta like nap in the afternoon.

01:34:48.460 A lot of people will ask me about this two phases of sleep thing, but in a very different sense,

01:34:53.860 which was you sleep about four hours, then you wake up, you kind of have a social life,

01:34:58.920 and then you go back to sleep for another four hours. And it's called the two sleeps or first

01:35:03.760 sleep and second sleep. And there's a great book written on this. Now that definitely happened

01:35:08.680 sometime in human civilization. It seemed to happen around the Dickensian era,

01:35:13.100 but there is nothing in our biology that suggests that we should be sleeping in that way. It seems

01:35:19.540 to have been a social feature, a social pressure.

01:35:23.760 There was a meal that took place, a large meal that took place between those two,

01:35:28.220 sandwiched between the two four-hour sleeps, right?

01:35:30.200 Correct. Yeah. Yeah. And people would, you know, they would have social activities planned,

01:35:34.700 they would write, they would play music, make love. People, you know, it was a real thing,

01:35:39.580 but it was not a biologically defined way of sleeping. It's not a natural way of sleeping.

01:35:46.140 You'd think you'd be, I mean, I guess you could condition yourself to do anything, but

01:35:49.320 to wake up after four hours of sleep under any circumstances, generally quite miserable.

01:35:54.220 It's rough. It's absolutely terrible. So that happened. But I think to come back to the point,

01:35:59.780 it's just that how we are sleeping in modern society is not just about sleeping less. It's about

01:36:07.420 sleeping less, the timing of our sleep, and the structure of our sleep. So I think those are

01:36:13.500 the things that have changed. One of the questions I get asked a lot, and I don't know the answer,

01:36:17.480 so I'm hoping you do, is one, you sort of touched on it, but the efficacy of naps,

01:36:23.700 and two, the notion that you should or shouldn't nap if you're having sleep issues, right? So on the

01:36:32.760 second one, I have a slightly stronger point of view, and I hope it's not incorrect, which is

01:36:37.080 in that situation, the nap is wrong. You want the adenosine accumulation, and you want it to draw your

01:36:43.500 bedtime back earlier, right? If you're struggling, I would rather you push through not sleeping and

01:36:49.680 then go to bed at nine o'clock. But to the first question, more broadly speaking, I can't imagine that

01:36:56.880 one cycles through what you described at the outset of our discussion in a 20-minute or an hour-long nap,

01:37:04.660 right? I mean, what actually, do you have people nap in the lab as well? You must. Yeah, what does

01:37:08.920 that look like? So we've done a lot of these studies where we ask, you know, what are the benefits of

01:37:14.180 napping? And we certainly do find them. We find benefits for learning and memory. We find benefits

01:37:19.540 on immune function. We find benefits for things like cardiovascular health in terms of metrics of heart rate

01:37:26.320 variability. We see benefits in terms of lowered systolic blood pressure. All of these things we

01:37:33.860 found benefit by way of a nap. Now, when I say nap, you can ask, what do you mean by a nap? And we've

01:37:41.160 played around with that dose, as it were, and we've done sort of, we've tried to do a little bit of a dose

01:37:46.260 response curve with some of these things too. What we found is that you can go down as low as about 20

01:37:52.860 minutes of a nap and still see some mental benefits in terms of things like learning and memory. But

01:37:59.720 typically we do a nap that is 90 minutes because that gives the ability of the brain to go through

01:38:05.560 the full 90-minute cycle so that the brain can go through all of the stages of non-REM sleep, stages

01:38:11.480 one through four, and it gets the shot to get to REM sleep. And then we can ask by testing people in

01:38:18.540 brain scanners sort of before and after and with cognitive tests and then with peripheral body

01:38:22.620 tests. We can ask, you know, what was the benefit of that nap relative to a group that does exactly

01:38:31.060 the same thing? They lie in bed, but we don't let them fall asleep, that they may just kind of passively

01:38:37.440 watch a movie. So they're inert for the same amount of time, they're in bed, they get all of the wires on

01:38:43.520 the head. We tried to do it as controlled as possible with the only difference between being sleep

01:38:48.180 between those two groups. So you do find benefits to your first point. To your second point, you're

01:38:55.280 absolutely right in terms of your sleep prescription for your patients. The current recommendation is

01:39:02.380 this, if you are finding it difficult to sleep at night and sleep throughout the night, the

01:39:07.300 recommendation is do not nap during the day. And this helps us when we try to think about the

01:39:13.540 biological mechanisms of how wake and sleep actually work. And you mentioned a chemical called

01:39:19.140 adenosine. So from the moment that you and I woke up this morning and everyone listening, the moment

01:39:25.680 that you woke up this morning, a chemical has been building up in your brain called adenosine. The more

01:39:31.040 of it that builds up, the sleepier that you will feel. So it's the sleepiness chemical. And after about

01:39:37.780 16 hours of accumulation, you should have enough weight of healthy sleepiness, you should have

01:39:44.400 enough adenosine within the brain to have you fall asleep easily and then stay asleep soundly.

01:39:51.840 And then when you're in sleep, the brain actually will clear away that adenosine, removing the sleepiness.

01:40:00.320 And after about eight hours of sleep, you've removed 16 hours of accumulation of adenosine,

01:40:06.760 so that when you wake up in the morning, you feel alert, you feel more awake. And that's the reason

01:40:12.460 that you don't feel as sleepy anymore throughout that following day. Now, the problem with naps is

01:40:20.200 that if you nap in during the day, particularly if you nap in the late afternoon, essentially what

01:40:26.040 you're doing there is acting like a pressure valve on a steam cooker that you're building up all of this

01:40:32.940 great healthy sleepiness, this sleep pressure, and then you nap and you'll just release some of that

01:40:39.380 sleepiness. So now when it comes to your normal bedtime, you don't feel as sleepy. Why? Because

01:40:45.420 the nap removed some of that sleepiness pressure. It removed some of the adenosine and you are going to

01:40:52.280 perhaps find it more difficult to fall asleep and stay asleep. Perhaps is the important point.

01:40:58.180 There are some people who can nap. If you can nap regularly and or you don't have problems with

01:41:03.460 sleep at night, then naps are just fine. But if you can't nap regularly, and especially if you're

01:41:09.380 struggling with sleep at night, the advice is don't nap, stay awake, build that healthy sleepiness,

01:41:16.180 build lots of adenosine. That gives you the best chance to stay asleep and then fall under the spell

01:41:21.160 of sleep under that weight of sleepiness for as long as possible.

01:41:24.360 You know, it might be oversimplifying, but the way I generally describe sleep to patients is

01:41:29.040 sort of a balancing act between adenosine, cortisol, and melatonin. And you've got to like,

01:41:34.760 these three things have to make, they're like an orchestra. You have to have the crescendo of

01:41:39.880 adenosine with the decrescendo of cortisol, and then you have to let melatonin take the brakes off this

01:41:46.200 whole thing. And, you know, even when I just think about myself personally, I am convinced that

01:41:51.560 virtually all of my sleep woes are on the cortisol axis. And the reason I'm hypothesizing that is I

01:41:59.660 can't measure adenosine. It's very complicated to measure. You need a very special type of

01:42:04.280 mass spec to do that, or MRS. But we can measure melatonin in urine. We can collect urine overnight

01:42:09.820 and measure melatonin levels. So we have at least some sense of how much we're making.

01:42:13.880 But because I measure glucose 24 hours a day, and I'm pretty familiar with meal timing and response,

01:42:22.480 in many ways, nighttime glucose for me is a proxy for nighttime cortisol. And the association between

01:42:29.360 high nighttime glucose and poor sleep is overwhelming. And I suspect it's through this

01:42:37.240 cortisol axis.

01:42:37.980 There is, in some ways, a manipulation that we'll speak about in just probably a little while of

01:42:43.920 adenosine. Even though you can't measure it, you can actually manipulate it with something called

01:42:48.700 caffeine. And that can demonstrate to you how-

01:42:52.240 Matt, by the way, we're in America here. I don't think there's a human that doesn't know what

01:42:56.720 caffeine is. You get everybody's ears perked up like you're going to mention some new compound you

01:43:02.820 just discovered in the lab yesterday. It's caffeine.

01:43:06.440 Yeah, yeah, yeah. Somnus squared. No, no, it's called caffeine. But what's interesting regarding

01:43:13.720 cortisol, which really supports your hypothesis, right now, the leading underlying mechanism,

01:43:20.760 the leading theory of insomnia, essentially comes back to that fight or flight branch of the nervous

01:43:27.380 system, that you are in this profound parasympathetic state where that nervous system

01:43:33.200 is cranked too far in sort of the high strength direction. And if you essentially have a cannula

01:43:41.460 and you sample cortisol every 30 minutes across a 24-hour period, you see this, you know, this nice

01:43:49.440 rise during the day where cortisol is helping you stay awake. It's keeping you alert. It's doing all of

01:43:55.780 the things that it needs to do. And then as you start to come towards the nighttime period, cortisol

01:44:00.520 should drop and needs to drop for you to be able to initiate sleep. And in fact, it usually hits its

01:44:07.380 sort of almost its nadir at the point where most people will say, that's my typical bedtime.

01:44:14.300 And it then goes through this awesome sort of downstroke in terms of concentration. And then a few

01:44:21.100 hours before you wake up, cortisol will start to rise. In other words, it's a preparatory hormone

01:44:26.560 that starts to get you ready for wakefulness. If you look at patients with insomnia, however,

01:44:33.620 what you see is that their cortisol starts to come down nicely in the evening, sort of six or seven or

01:44:39.940 eight. So, you know, they could feel just as sleepy as a regular person, but then right before bed,

01:44:45.640 it goes back in the opposite direction, it spikes again, and then starts to come down. And then

01:44:52.400 often in the middle of the night, you will see cortisol spike back again when it should be lower

01:44:57.860 still before it starts its rise. Those two cortisol spikes, I think, are the natural biological bookmarks

01:45:06.640 for what we call sleep onset insomnia and sleep maintenance insomnia. So your cortisol spikes just

01:45:13.360 before bed, you can't fall asleep, sleep onset insomnia. Cortisol spikes in the middle of the

01:45:19.080 night, you wake up, Rolodex of anxiety in brain starts to happen. Cortisol spikes, you can't fall

01:45:25.640 back asleep. That's what we call sleep maintenance insomnia. Those are the two principal types of

01:45:30.300 insomnia. I think that's where you can see part of the biological basis. And my guess is that if we

01:45:36.560 were to measure, you know, things like your heart rate variability and we were to decompose them with

01:45:44.340 some mathematical equations that you would know of, things like a fast Fourier transform, you can look

01:45:49.780 at the contribution of the sympathetic versus parasympathetic when you break down the heart rate

01:45:54.380 variability. My guess is that you're going to see that sympathetic, that fight or flight branch of the

01:46:01.220 nervous system getting jacked just before sleep onset. And then once again, spiking in the middle

01:46:06.740 of the night, underlying the cortisol spike, which underlies the awakening, which underlies the syndrome

01:46:13.000 of insomnia. Yeah. Who do you think in that equation? I mean, I don't want to give everyone a lecture on

01:46:19.820 this topic because you've done a great job explaining what sympathetic is, right? Which is autonomic, but it's

01:46:25.600 sort of different from the glucocorticoid regulated pathway. Do you think that the sympathetic is

01:46:31.280 driving the glucocorticoid or the other way around? And the only reason I ask is in a moment, I'm going

01:46:37.400 to propose the only two things that I've ever really thought I have in the toolkit to treat that. But

01:46:44.320 the answer to this question might impact that as well. I honestly don't think we have the evidence

01:46:50.540 to argue which way round that chain of command unfolds. I just genuinely don't know of the

01:46:57.680 evidence. I think both are tenable hypotheses. Both could be independent. So they, I think they could be

01:47:03.680 non-mutually exclusive. So in truth, I don't think I know the evidence in favor. But what's nice is that

01:47:09.940 the fact that we know that they both happen and can be causal contributors means that we've got at least

01:47:15.980 two treatment targets that we can go after, which I think will then play into exactly what you're going

01:47:22.600 to say, which...

01:47:23.580 Well, I don't know that they're correct. But to me, when we see that pattern, because we do measure

01:47:28.440 nighttime cortisol and we, you know, we measure the metabolites of it and all of these other complicated

01:47:34.020 assays, when we see that high nighttime cortisol, we basically, you know, I mean, we sort of, that's a very

01:47:39.960 hard problem to treat. You know, because if you really think about it, we're just dealing with the endocrine

01:47:43.700 system, right? You know, it's a heck of a lot easier to treat thyroid dysfunction and dysfunction

01:47:49.860 of the sex hormones. And I'm not saying that those things are easy, but they are easier than treating

01:47:54.800 the upper end of the, like when you have low sympathetic function and low glucocorticoid

01:48:01.360 function, that's actually easier to treat than the opposite. The hardest problem, I think, in clinical

01:48:05.240 medicine from an endocrine perspective is high sympathetic output, high glucocorticoid output.

01:48:10.740 And really, it's amenable to behavioral therapy, right? It really comes down to a change in mindset

01:48:16.760 and a change in routine. We were talking about this earlier. It's like, the further I can separate

01:48:22.960 bed from email, the better. Because for me, email is just, I hate email so much. So anytime I'm looking

01:48:28.980 at email... Everyone has this agenda for your...

01:48:30.680 Yeah, yeah, yeah. I'm just, I'm just, I have a low level of pissed off-edness whenever I'm looking at

01:48:35.440 email. So the further I can separate my pissed off-edness from bed, the less my cortisol is.

01:48:41.700 Can I get that T-shirt, by the way?

01:48:42.980 Yeah, yeah, yeah, yeah, yeah. And then the second thing, which appears to work anecdotally,

01:48:48.140 is phosphatidylserine at a high enough dose, which generally requires about 400 to 600 milligrams.

01:48:54.020 And again, for me, the only time I find that that's really necessary is under those jet lag

01:49:00.180 circumstances. When you're trying to put yourself in the time zone of London, when you're leaving

01:49:04.480 San Francisco, and you basically have to make yourself go to bed, and it's only 2 p.m. in your

01:49:10.320 brain, but it's late enough there. Slamming yourself with 600 to phosphatidylserine really

01:49:16.520 seems to move the needle, both in terms of your glucose response and your cortisol response.

01:49:20.220 What else do you have up your sleeve on that particular problem? Because honestly,

01:49:24.780 I can't think of a more difficult clinical problem than the one you've just described.

01:49:29.840 I think the glucocorticoid issue is, I think is a really tough nut to crack. I think for the

01:49:38.620 sympathetic drive, I do think you're correct that it's behavioral. There, I would actually argue

01:49:45.100 meditation. And I looked at this data when I was researching the book. Gosh, this must have been

01:49:51.800 about four years ago now. You know, and I'm, I think I'm a bit of a hard nose, dyed-in-the-wool

01:49:57.740 scientist. So some of the, when I was reading the studies on meditation and insomnia, I was wondering

01:50:03.540 if it was a bit woo-woo and didn't know how to take it seriously. But the data was very compelling

01:50:09.320 in terms of it really does decrease the amount of time it takes someone to fall asleep. The

01:50:15.820 continuity of their sleep is improved. If you look at some of these apps that do it very well,

01:50:22.400 and you know, things like Headspace or Calm, my guess, although I don't know if they've ever released

01:50:27.140 this data, is that if you were to look at their usage statistics, you will already see that people

01:50:33.220 are self-medicating their insomnia by way of meditating right before bed. And I think, you know,

01:50:40.640 both of those apps are starting to now launch, you know, aspects of sleep and sleep health in their

01:50:46.340 portfolio of offerings, rightly so, because they've probably realized that sleep is a huge part of the

01:50:54.200 issue for which people are coming to them to solve. And the clinical data is very supportive of that.

01:51:01.660 So I would say that for driving, it's almost impossible to fall asleep if your autonomic

01:51:10.240 nervous system is pushed in a sympathetic dominant state, in a fight or flight state.

01:51:17.300 And a good example of this is, you know, when you are jet lagged, you can be there, you may not have

01:51:24.060 slept on the plane, you can feel that you're tired, you're really tired. But for some reason,

01:51:29.060 you know, your cortisol is now spiking because you're mismatched with your circadian rhythm,

01:51:34.040 you can feel your heart rate. And you can just know that your nervous system is cranked. And you

01:51:40.820 are tired as tired can be in your eyes and in your mind. But your nervous system, because it's on that

01:51:48.520 sort of upward swing, will not let you fall asleep. But meditation, I think, is a great way to drive a

01:51:56.100 parasympathetic dominance of the nervous system, which is to take you out of the fight or flight

01:52:01.760 branch and more into this quiet, sort of more introspective state. I think that's probably the

01:52:09.080 better thing. Now, I know it's not from a clinical medical perspective as desirable, because it's not

01:52:16.980 a pill that we can take. It's not a dosage. It's you just have to put in the work to sit there and

01:52:22.460 meditate. But I think that's the nature of sleep. I mean, I think CBTI has efficacy, but it requires

01:52:27.860 work. It does. And changing the way you eat requires work. It's not a pill. I mean,

01:52:32.420 and exercising, you know, same thing. And I've been asked, this is a totally unrelated topic,

01:52:36.620 but people always say, oh my God, you know, please write me a prescription for metformin because it's

01:52:40.840 going to help me lose weight and blah, blah, blah. And I said, look, let me be clear. I mean,

01:52:44.560 I think metformin is a really powerful pill. I don't think it's nearly as powerful as an exceptional diet.

01:52:51.080 In other words, metformin superimposed on the world's worst nutrition versus no metformin on

01:52:58.060 exceptional nutrition. I don't think it's any comparison. In the end, molecules that you take

01:53:04.860 intermittently aren't really nearly as potent as molecular changes that occur over a time course

01:53:11.600 behaviorally. I mean, I wish that weren't true, truthfully.

01:53:15.840 Yeah, me too. I'm not anti-pharmacology by any means. You know, if we had a good sleeping pill,

01:53:21.920 I would embrace it and I would recommend it. But right now, the sleeping pills on the markets,

01:53:27.140 they firstly don't produce naturalistic sleep. Secondly, they've been associated with a

01:53:32.080 significantly higher risk of death and also significantly higher risk of cancer.

01:53:37.160 I'm glad you mentioned this because this is a great segue to this topic. And going back to my

01:53:41.800 friend, Kirk Parsley, and I remember another thing he said, because I, this is again, going back to

01:53:46.240 2011, 2012, my go-to travel drug was Ambien. You know, fortunately, I've never, I've never had

01:53:54.660 insomnia, but I empathize greatly with those who do. I've seen enough people with it. But when I needed

01:54:00.060 to sleep, I was going to take Ambien. And I remember getting hints that this wasn't a great drug

01:54:05.240 when I would, when I worked at McKinsey, I used to have to go from San Francisco to Atlanta every

01:54:11.540 single Monday for a day. And because I was ruthless in my pursuit of not wasting time,

01:54:17.460 rather than like take a Sunday afternoon flight and come back Tuesday, I took a red eye from San

01:54:24.260 Francisco to Atlanta every single Sunday. And as one of my friends put it, that's the pink eye and

01:54:30.280 ain't the red eye. Like that's a four hour flight, right? You always get the tailwind. So you, you

01:54:36.320 get about three and a half hours of sleep on a plane. I'd get to Atlanta at five, pick up a rental car,

01:54:41.920 drive to the gym, do a two hour workout, shave, shower, get to the client, stay there all day,

01:54:49.440 you know, leave at six 30, get to the airport by, you know, eight 30, get the flight back to San

01:54:55.200 Francisco. You're back by whatever your home at minute. So it was a 26 hour door to door every

01:55:01.060 single time. And I was using Ambien to make sure I slept on that flight. I would, I would come up

01:55:06.940 with these grisly discoveries, which is I'd find emails I sent that I didn't know I'd sent.

01:55:12.500 And thank God, none of them were inappropriate, but you hear these stories of people doing really

01:55:19.520 crazy stuff on Ambien. Which is never committed to memory. And you know, it's only when people make

01:55:25.580 the phone call the next morning and they say, you know, what were you, what were you going on about

01:55:29.720 last night? Oh, I, one of the scariest things I ever saw was I remember looking at my phone and

01:55:35.160 seeing outgoing calls to people that I didn't remember making. So my friend Kirk Parsley is

01:55:40.600 saying, and by the way, that Ambien crap that you take, you know, a couple of times a month, 0.98

01:55:45.260 he said, you know, that's not sleep, right? And I said, what do you mean? It's not sleep. I'm 0.98

01:55:49.820 out. And he just, you know, again, you'll laugh at this because this is exactly the kind of

01:55:54.720 story that you tell so eloquently. He said, you're confusing consciousness and sleep or lack

01:56:00.480 of consciousness and sleep. If I took a baseball bat and hit you on the head, I could render you 0.99

01:56:06.880 completely unconscious laying on the floor for eight hours. Do you think in any way that mimics the 0.72

01:56:12.840 restorative process of sleep? I mean, not even close, right? So, you know, think of Ambien more

01:56:18.060 as a chemical baseball bat to the head than something that's actually promoting what you've

01:56:23.580 just been speaking about for the last, you know, little while. It's such a critical point. Again,

01:56:28.040 I'm not anti-pharmacology and I know some of the people in the drug industries, the scientists,

01:56:33.720 and their goal, their passion is to create drugs that better humanity. And they are genuine

01:56:41.080 and authentic about that. And I love that about them. But I also will not accept or speak in 1.00

01:56:48.880 public about a drug that clearly doesn't seem to be beneficial for a certain process. And

01:56:54.220 unfortunately, if you look at the evidence, firstly, you find absolutely that those drugs

01:56:59.980 are a class of drugs that we call the things like Ambien, Lester, sort of, you know, all of these

01:57:06.260 things that you've heard. There are a class of drugs that we call the sedative hypnotics and sedation is

01:57:11.400 not sleep, but people mistake the former with the latter. And it's not, it's literally just knocking

01:57:18.540 your cortex out. So I'm not going to argue when you take those drugs that you're awake, you're clearly

01:57:25.220 not awake. But to say that you're in naturalistic sleep is an equal fallacy. Because if I look at the

01:57:33.380 electrical signature of your sleep, when you are on or off Ambien, radically different for a start,

01:57:40.160 you don't get into the same depth of deep sleep that you get. It's not the same electrical quality

01:57:44.780 of deep sleep. The second thing that we found is that there does seem to be a higher association with

01:57:52.060 mortality risk. Now, we don't know if it's causal. And my guess is that on the strength of the evidence

01:57:57.800 right now, IRB committees, ethics committees that sort of allow you to do certain scientific studies,

01:58:03.380 may not even allow you to try and do those studies, because the evidence is already so deleterious.

01:58:10.200 The other finding was this strong association with risk of all forms of cancer, which perplexed me

01:58:19.200 because you would think, well, if I'm taking this drug that helps me sleep more, and sleep is so good

01:58:25.360 for the immune system, including things like natural killer cells and combating sleep, then this drug

01:58:31.380 induced sleep should actually make my cancer risk lower, not higher. And the fact that it is higher,

01:58:39.840 I think tells you that the type of sleep that you're getting is not of this rich, complex, restorative,

01:58:47.860 restful, biologically appropriate sleep. The other thing I would come on to, and these are very

01:58:54.040 difficult studies to get funding for, because, you know, obviously, some of the drug companies,

01:58:59.120 they may not necessarily want some of this information out there. It was the only chapter

01:59:03.600 in the book, by the way, that my publisher brought in a legal team to kind of look at when I was kind

01:59:09.180 of making all of these claims. But this is just, I'm just conveying the science. There's nothing

01:59:13.600 controversial about the science. And they ended up feeling legally comfortable with what I wrote.

01:59:18.540 There was this stunning study done by a good friend of mine called Marcus Frank at Penn. And he was

01:59:26.540 looking at basically brain plasticity. How does the brain make connections and strengthen those

01:59:35.720 connections? And those strengthening of connections is the underlying mechanism that we know leads to

01:59:42.180 the long-term formation of memories. And it's a very well understood mechanism, this idea of brain

01:59:47.800 plasticity. And there is a model where if you kind of patch the eye of an animal, and then you measure

01:59:54.220 the visual cortex, what you see is that once you patch one eye, the brain says, well, we're not going

01:59:59.840 to waste the real estate of that one eye that's no longer doing visual processing. We're going to shift

02:00:05.440 a lot of that over to the other eye's territory and give it even more kind of rich, connective fidelity,

02:00:12.600 as it were. So you've got this model of what's called monocular deprivation plasticity, which is

02:00:18.140 a classic model of how the brain rewires itself for things like learning and memory. What he found is

02:00:24.420 that he took a group of animals and he gave them naturalistic sleep and did this kind of monocular

02:00:30.620 deprivation. And sleep, it turns out, basically doubles the strengthening of plasticity. Sleep is

02:00:38.080 wonderful for shifting things and making new connections and really strengthening those neural

02:00:43.620 connections. Sleep is essential for brain plasticity. And it's almost, sleep will almost double the strength

02:00:51.220 of the connections relative to if you just keep the animal awake. So sleep, it's not just time that helps

02:00:57.580 the brain rewire itself. It's time with sleep that makes the difference. Then he dosed those animals

02:01:03.200 with Ambien. Now, firstly, the animals slept longer because they were dosed with this sedative

02:01:09.400 hypnotic. And you would think that surely that should improve the strengthening of rewiring in

02:01:16.120 the brain. It did exactly the opposite. Yeah, I was going to say, or at worse, not change from the

02:01:21.840 control. Right. From the control. If it was, quote unquote, naturalistic sleep. What he found was

02:01:27.480 actually a 50 percent unwiring of the connections that had been laid down before when the animal was

02:01:34.900 awake. In other words, not only had sleep failed to strengthen the connections, the Ambien-laced or

02:01:41.780 the Ambien-induced sleep was doing the opposite. It was weakening connections within the brains.

02:01:49.240 Now, what makes me fearful about that is, firstly, if you look at the number of people who are using

02:01:54.520 these medications, you know, it took George Lucas, I often say, I think about 40 years with the Star Wars

02:02:01.560 franchise to amass about 4 billion in profit. It took Ambien less than 24 months to amass that

02:02:08.320 in terms of profit. So you've got to imagine the number of prescriptions being written are high.

02:02:14.960 Then the fact that the prescription age of these sleep aids is starting to decrease gradually as parents

02:02:23.580 get more concerned about their kids who may have more anxiety, who are having sleep problems,

02:02:29.620 they're turning to pills. And when you bring that experiment back into the context of pediatric

02:02:37.000 prescriptions of sleep medication, you've got a brain that is developing, a brain that needs to wire

02:02:44.240 itself up, strengthen connections, build and mature. It needs to learn. It needs to consolidate.

02:02:51.060 And then you're providing a form of sleep through medication that may actually be unwiring that

02:02:57.180 nervous system and unwiring those memories rather than strengthening them. That frightens me.

02:03:03.520 You can find all of this information and more at peteratiamd.com forward slash podcast.

02:03:10.260 There you'll find the show notes, readings and links related to this episode. You can also find my blog

02:03:15.980 at peteratiamd.com. Maybe the simplest thing to do is to sign up for my subjectively non lame once a week

02:03:22.000 email where I'll update you on what I've been up to, the most interesting papers I've read, and all

02:03:26.640 things related to longevity, science, performance, sleep, etc. On social, you can find me on Twitter,

02:03:32.480 Instagram and Facebook, all with the ID peteratiamd. But usually Twitter is the best way to reach me to

02:03:38.180 share your questions and comments. Now for the obligatory disclaimer, this podcast is for general

02:03:42.860 informational purposes only and does not constitute the practice of medicine, nursing or other

02:03:47.420 professional healthcare services, including the giving of medical advice. And note, no doctor

02:03:52.920 patient relationship is formed. The use of this information and the materials linked to the podcast

02:03:58.180 is at the user's own risk. The content of this podcast is not intended to be a substitute for

02:04:02.700 professional medical advice, diagnoses or treatment. Users should not disregard or delay in obtaining

02:04:08.280 medical advice for any medical condition they have and should seek the assistance of their

02:04:12.520 healthcare professionals for any such conditions. Lastly, and perhaps most importantly, I take

02:04:18.040 conflicts of interest very seriously. For all of my disclosures, the companies I invest in and or

02:04:23.000 advise, please visit peteratiamd.com forward slash about.

The Peter Attia Drive - April 08, 2019

#48 – Matthew Walker, Ph.D., on sleep – Part II of III： Heart disease, cancer, sexual function, and the causes of sleep disruption (and tips to correct it)

Episode Stats

Summary

Transcript