Matthew Walker's "Why We Sleep" Is Riddled with Scientific and Factual Errors

created: 2019-11-15; modified: 2019-11-15

Note: I link to a bunch of paywalled studies in this essay. Please do not use sci-hub to access them for free and do not use this trick (a) to easily redirect papers to sci-hub.

Introduction

Matthew Walker (a) is a professor of neuroscience and psychology at the University of California, Berkeley, where he also leads the Center for Human Sleep Science.

His book Why We Sleep (a) was published in September 2017. Part survey of sleep research, part self-help book, it was praised by The New York Times (a), The Guardian (a), and many others. It was named one of NPR’s favorite books of 2017. After publishing the book, Walker gave a TED talk, a talk at Google, and appeared on Joe Rogan’s and Peter Attia’s podcasts. A month after the book’s publication, he became (a) a sleep scientist at Google.

On page 8 of the book, Walker writes:

[T]he real evidence that makes clear all of the dangers that befall individuals and societies when sleep becomes short have not been clearly telegraphed to the public … In response, this book is intended to serve as a scientifically accurate intervention addressing this unmet need [emphasis in this quote and in all quotes below mine]

In the process of reading the book and encountering some extraordinary claims about sleep, I decided to compare the facts it presented with the scientific literature. I found that the book consistently overstates the problem of lack of sleep, sometimes egregiously so. It misrepresents basic sleep research and contradicts its own sources.

In one instance, Walker claims that sleeping less than six or seven hours a night doubles one’s risk of cancer – this is not supported by the scientific evidence. In another instance, Walker seems to have invented a “fact” that the WHO has declared a sleep loss epidemic. In yet another instance, he falsely claims that the National Sleep Foundation recommends 8 hours of sleep per night, and then uses this “fact” to falsely claim that two-thirds of people in developed nations sleep less than the “the recommended eight hours of nightly sleep” – a myth that spread like wildfire after the book’s publication.

Walker’s book has likely wasted thousands of hours of life and worsened the health of people who read it and took its recommendations at face value.

Any book of Why We Sleep’s length is bound to contain some factual errors. Therefore, to avoid potential concerns about cherry-picking the few inaccuracies scattered throughout, in this essay, I’m going to highlight the five most egregious scientific and factual errors Walker makes in Chapter 1 of the book. This chapter contains 10 pages and constitutes less than 4% of the book by the total word count.

No, shorter sleep does not imply shorter life span

On page 4, Walker writes: If you’re concerned about me taking this quote out of context, I provide the full sentence and the two paragraphs leading up to it here.

the shorter your sleep, the shorter your life span

This is false. Walker does not cite any studies when making this claim. Most of the studies on the relationship between life span and sleep duration find a U-shaped relationship between length of sleep and longevity, i.e. both short- and long-duration sleep are associated with higher mortality [1 (a) Shen X, Wu Y, Zhang D. Nighttime sleep duration, 24-hour sleep duration and risk of all-cause mortality among adults: a meta-analysis of prospective cohort studies. Scientific Reports. 2016 Feb 22;6:21480. , 2 (a) Cappuccio FP, D’Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. 2010 May 1;33(5):585-92. , 3 (a) Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Archives of General Psychiatry. 2002 Feb 1;59(2):131-6. ]. The studies typically find that people who sleep 7 hours have the highest longevity. Here’s a graph from the first study, published in Scientific Reports in 2016, which performed a meta-analysis of thirty-five prospective cohort studies:

Figure 1. The dose-response analysis between nighttime sleep duration and risk of all-cause mortality. The solid line and the long dash line represent the estimated relative risk and its 95% confidence interval. Note: the red dashed line on the graph is mine.

Note that the lowest mortality on the graph is at just below 7 hours and that mortality at 5 hours of sleep per night is basically the same if not lower than mortality at 8 hours of sleep.

As the Encyclopedia of Sleep Kushida C. Encyclopedia of sleep. Academic Press; 2012 Dec 31. – which Walker cites two pages later – notes:

[T]he popular expectation that short sleep is correlated with short life span and long sleep with greater longevity is not supported by the existing literature.

Also, no – sleeping less than six hours a night does not double your risk of cancer

On page 3, Walker writes:

Routinely sleeping less than six or seven hours a night demolishes your immune system, more than doubling your risk of cancer.

This is false. Walker does not cite any studies that support this assertion anywhere in the book. There do not appear to exist any experimental studies or studies that would reasonably be able to establish causality, that would support this claim. Even the epidemiological evidence (which you should never use to claim causality (a) Even if you “controlled” for confounding variables (a). ) disagrees with Walker’s assertion. For example, a systematic review of sixty-five studies from 2018 (doi), Chen Y, Tan F, Wei L, Li X, Lyu Z, Feng X, Wen Y, Guo L, He J, Dai M, Li N. Sleep duration and the risk of cancer: a systematic review and meta-analysis including dose–response relationship. BMC Cancer. 2018 Dec;18(1):1149. which involved 1,550,524 participants and 86,201 cancer cases, found that neither short nor long sleep duration was associated with increased cancer risk.

Figure 2. Nonlinear dose–response analyses of sleep duration and cancer risk. The solid line and the long-dashed line represent the estimate odds ratios and their 95% confidence intervals. Seven hours of sleep per night was used as the reference Note: the red dashed line on the graph is mine.

How much confidence should we place in epidemiological sleep data?

All of the big studies that are used as inputs for meta-analyses like those I cited above use self-reported data on sleep duration, since it’s impossible to record objective sleep data for a large number of people (this will soon change with the advent of smart watches, bracelets, and rings).

Self-reported data is notoriously unreliable, so it’s not clear how meaningful those studies are, even if all we’re looking for are various correlations. See further discussion of this in Section 11, where I hypothesize that people who have the lowest mortality actually sleep just 6 hours a day.

No, a good night’s sleep is not always beneficial: sleep deprivation therapy in depression

Note: in this section, I only talk about acute sleep deprivation, i.e. being sleep deprived for one or several days. Chronic or externally imposed sleep deprivation is an entirely different matter and has no relation to sleep deprivation therapy.

On page 8, Walker writes:

[W]e are now forced to wonder whether there are any biological functions that do not benefit by a good night’s sleep. So far, the results of thousands of studies insist that no, there aren’t.

This is false. First, an enormous literature dedicated to the treatment of depression with sleep deprivation has found that people with depression frequently benefit by not getting a good night’s sleep.

Second, Walker directly contradicts himself in Chapter 7 by acknowledging that there are cases when a good night’s sleep is not helpful after all:

Approximately 30 to 40 percent of these patients will feel better after a night without sleep … the 60 to 70 percent of patients who do not respond to the sleep deprivation will actually feel worse, deepening their depression. As a result, sleep deprivation is not a realistic or comprehensive therapy option.

In this quote, not only does Walker contradict himself, but he also misrepresents the benefits and the dangers of sleep deprivation therapy. He slightly downplays the number of people who benefit from it – it’s closer to 45-50% if you believe this meta-analysis from 2017 (pdf, a) Boland EM, Rao H, Dinges DF, Smith RV, Goel N, Detre JA, et al. Meta-Analysis of the Antidepressant Effects of Acute Sleep Deprivation. The Journal of Clinical Psychiatry. 2017;78(8). or it’s 40-60% if you believe Walker himself from 2009 (doi) Walker MP, van Der Helm E. Overnight therapy? The role of sleep in emotional brain processing. Psychological Bulletin. 2009 Sep;135(5):731. – and he completely mischaracterizes the dangers of it.

In the book, Walker writes that “the 60 to 70 percent of patients who do not respond to the sleep deprivation will actually feel worse, deepening their depression”. However, a review of the literature from 2002 (a) Giedke H, Schwärzler F. Therapeutic use of sleep deprivation in depression. Sleep Medicine Reviews. 2002 Oct 1;6(5):361-77. that he cited in his study from 2009 tells us that depression worsens in less than 10% of patients:

Total sleep deprivation (TSD) for one whole night improves depressive symptoms in 40-60% of treatments. The degree of clinical change spans a continuum from complete remission to worsening (in 2-7%). Other side effects are sleepiness and (hypo-) mania. … It is still unknown how sleep deprivation works.

Here’s a review from 2010 (a): Hemmeter UM, Hemmeter-Spernal J, Krieg JC. Sleep deprivation in depression. Expert Review of Neurotherapeutics. 2010 Jul 1;10(7):1101-15.

The observation that after the recovery night a great majority of SD responders relapse into depression suggests that sleep per se may have a depressiogenic property.

The rapid effect of SD on depressive mood within hours is a fascinating experience for the patient, who may have been depressed for weeks or months …

SD is the only established antidepressant therapy that acts within hours, and therefore, can be applied in patients with treatment-resistant depression with a chance of approximately 50% of seeing an immediate, although temporary, relief from depressive symptoms without major side effects. … The experience of realizing that depression can be lifted and sleep can improve is very important for the further therapy motivation of treatment resistant depressed patients. … [Sleep deprivation] can be combined with antidepressant medication, predominantly serotonergic agents, with bright light therapy and with a phase advance of sleep cycles. All these strategies have been able to provide a chance to stabilize the SD response, at least in a subgroup of patients.

Finally, although Walker states that “sleep deprivation is not a realistic or comprehensive therapy option”, a review chapter of sleep deprivation (a) Dallaspezia S, Benedetti F. Sleep Deprivation Therapy for Depression. Sleep, Neuronal Plasticity and Brain Function Current Topics in Behavioral Neurosciences. 2014;:483–502. in the book Sleep, Neuronal Plasticity and Brain Function published in 2014 reads:

[C]onsidering its safety, this technique [sleep deprivation] can now be considered among the first-line antidepressant treatment strategies for patients affected by mood disorders. …

SD is a rapid, safe, and effective therapy for depression. In recent years, this technique has passed the experimental developmental phase and reached the status of affordable clinical intervention for everyday clinical therapy of depressed patients with an increasing literature regarding its safety and efficacy.

This is important because Walker’s scaremongering is likely to harm people with depression who decide to avoid sleep deprivation therapy as a result of reading his book.

Interlude: no, you can't randomly cite 2,000-page-long books and hope nobody will read them

On page 6, Walker writes:

[E]very species studied to date sleeps

This is false, at least, according to Walker’s own source. When making this claim, he cites:

Kushida, C. Encyclopedia of Sleep, Volume 1 (Elsever, [sic] 2013)

….which turns out to be a 2,736 page book that costs $1,995. Fortunately, Walker tells us that we should search for this information somewhere in “Volume 1” or the first 638 pages of the book.

Anyway, page 38 reads:

It now appears that many species reduce sleep for long periods of time under normal conditions and that others do not sleep at all, in the way sleep is conventionally defined.

No, lack of sleep will not outright kill you

On pages 4-5, Walker writes:

[T]here is a very rare genetic disorder that starts with a progressive insomnia, emerging in midlife [fatal familial insomnia (FFI)]. Several months into the disease course, the patient stops sleeping altogether. By this stage, they have started to lose many basic brain and body functions. No drugs that we currently have will help the patient sleep. After twelve to eighteen months of no sleep, the patient will die. Though exceedingly rare, this disorder asserts that a lack of sleep can kill a human being.

This is false. You cannot say that FFI (a) shows that a lack of sleep will kill a human being.

Here is a rough description of the disease: a genetic mutation results in the production of a misfolded protein in the brain, primarily in the thalamus. This protein is toxic to the nerve cells and, over time, it damages the thalamus, resulting in a variety of symptoms, typically including dementia, hallucinations, and insomnia. Eventually, the rest of the brain gets damaged as well, internal organs shut off, the patient’s ability to sleep gets heavily disrupted, and the patient dies.

It is reckless to claim that people with FFI die because of lack of sleep, given the amount of damage across the brain that accumulates in the course of the disease. Accordingly, FFI is considered a neurodegenerative disease. Looking at page 41 of the Encyclopedia of Sleep we discussed in the Interlude:

A disorder called fatal familial insomnia (FFI) is often presented as proof that sleep loss causes death in humans as it does in rats deprived by the forced walking method. However, FFI is a prion disease that affects all body organs and brain cells. There is little evidence that sleep induced by sedation can greatly extend life in FFI patients.

And as one paper notes (a): Schenkein J, Montagna P. Self-management of fatal familial insomnia. Part 2: case report. Medscape General Medicine. 2006;8(3):66.

[T]he prevailing belief [is] that FFI patients ultimately die of neural degeneration

Note that this was Walker’s only example of lack of sleep leading directly to death.

The seven sentences I quoted in the beginning of this section contain (at least) three more scientific errors. See discussion of them in Section 18.

No, the World Health Organization never declared a sleep loss epidemic

On page 4, Walker writes:

[T]he World Health Organization (WHO) has now declared a sleep loss epidemic throughout industrialized nations.

This is false. The WHO never declared a sleep loss epidemic.

In the footnote to this sentence, Walker cites:

Sleepless in America, National Geographic, http://channel.nationalgeographic.com/sleepless-in-america/episode/sleepless-in-america.

One might wonder why he cited a documentary film from National Geographic, rather than the WHO directly, but ok, maybe that film has a reference to the primary source. I watched the entire 88-minute long film twice (official YouTube link Can only be viewed from the US, unfortunately. (a)) to make sure I didn’t miss anything and the film never mentions the WHO or any sleep loss epidemics declared by the WHO and never features anyone from the WHO. I googled:

“world health organization” “sleep loss epidemic”

And didn’t find any documents by WHO. Further, when I restricted the search results to before September 28, 2017 (the date when the book was published) Search results page 1 (a), page 2 (a). I also googled the string ‘“world health organization” “sleep” “epidemic”’ but it didn’t return any relevant results either. , the hits either

used Walker as their source
never mentioned any sleep loss epidemics declared by the WHO

No, two-thirds of adults in developed nations do not fail to obtain the recommended amount of sleep

Suppose that you recommend that adults sleep 7-9 hours per night.

then, someone learns (a) that roughly 40% of people sleep less than 7 hours, roughly 25% sleep 7 hours, and roughly 35% sleep 8 hours or more. Linked data is for the US but it appears (a) that other developed countries have very similar sleep habits.
then they look at your recommendation and say that you recommended an average of 8 hours of sleep per night (this is technically true, under some assumptions For example, if sleep recommendations are uniformly distributed. )
then they say that you recommended 8 hours of sleep per night
then they say that two-thirds of people sleep less than the 8 hours you recommended

Would this be a fair representation of your position and of the data or would this be misleading?

This is literally what Walker does in his book. On page 3, in the very first paragraph of Chapter 1, Walker writes:

Two-thirds of adults throughout all developed nations fail to obtain the recommended eight hours of nightly sleep.

In the footnote to this sentence he writes:

The World Health Organization and the National Sleep Foundation both stipulate an average of eight hours of sleep per night for adults.

Here are the National Sleep Foundation’s sleep recommendations (a) announced in 2015:

Adults (26-64): Sleep range did not change and remains 7-9 hours

Here are the World Health Organization’s sleep recommendations:

‎

The quote is empty because WHO does not stipulate how much an adult should sleep anywhere. I don’t know where Walker got this information.

Summary

In the first chapter of Why We Sleep, Walker:

completely misrepresents the relationship between sleep and longevity and between sleep and cancer (Section 1)
erroneously states that getting a good night’s sleep is always beneficial (Section 2)
erroneously states that patients with fatal familial insomnia die because of lack of sleep (Section 3)
seems to invent a “fact” that the WHO has declared a “sleep loss epidemic” (Section 4)
misrepresents National Sleep Foundation’s sleep recommendations and uses them to misrepresent the number of adults failing to get the recommended hours of sleep (Section 5)
- also seems to invent the WHO’s sleep recommendations
calls his book “a scientifically accurate intervention”

Given the density of scientific and factual errors and an apparent invention of new “facts” by Walker, I would caution readers against taking the book’s recommendations at face value.

The potential harm done by the book

Here are some of the potential harms done by the book:

First, Walker’s misrepresentation of sleep deprivation therapy (Section 2) is likely to make many people with depression avoid this potent and largely safe treatment option.

Second, imagine that a 20-year-old who naturally needs to sleep for 7 hours a night, reads Why We Sleep, gets scared, and decides to spend the full 8 hours in bed every day. Then, assuming that they live until 75 they will waste more than 20,000 hours or more than 2 years of their life, with uncertain long-term side-effects.

Conclusion

If you take one thing away from this entire essay, remember this: as long as you feel good, sleeping anywhere between 5 and 8 hours a night seems basically fine for your health (see Section 1), regardless of whatever Big Sleep wants you to believe.

All of the evidence we have about sleep and long-term health is in the form of those essentially meaningless correlational studies, but if you’re going to use bad science to guide your sleep habits, at least use accurate bad science.

I’m planning to write more about sleep, science in general, and the interpretation of correlational studies in particular. Subscribe, if you’d like to stay updated.

If you’re interested in my personal experience with sleep, see Section 15.

Acknowledgements

I would like to thank (in reverse alphabetic order) Misha Yagudin, Brian Timar, José Luis Ricón, Ada Nguyen, Anastasia Kuptsova, Matt Kovacs-Deak, Basil Halperin, Steve Gadd, Mark Egan, and Maxim Efremov for reading drafts of this essay. They have improved it immeasurably. All remaining errors are mine.

I would like to thank Kyle Schiller and Adam Canady for the financial support of my writing and research.

Citation

In academic work, please cite this essay as:

Guzey, A. Matthew Walker’s “Why We Sleep” Is Riddled with Scientific and Factual Errors. Guzey.com. 2019 November. Available from https://guzey.com/books/why-we-sleep/

Appendix: people who sleep just 6 hours a day might have the lowest mortality

This paper by Lauderdale et al (a) Lauderdale DS, Knutson KL, Yan LL, Liu K, Rathouz PJ. Self-reported and measured sleep duration: how similar are they?. Epidemiology (Cambridge, Mass.). 2008 Nov;19(6):838-45. is the most well-cited paper on the topic of self-reported vs objectively measured sleep I found. It reads:

the correlation between reported and measured sleep duration was 0.47. Our model suggests that persons sleeping 5 hours over-reported their sleep duration by 1.2 hours, and those sleeping 7 hours over-reported by 0.4 hours.

Two other studies I found report a correlation of 0.43 (a) Cespedes EM, Hu FB, Redline S, Rosner B, Alcantara C, Cai J, Hall MH, Loredo JS, Mossavar-Rahmani Y, Ramos AR, Reid KJ. Comparison of Self-Reported Sleep Duration With Actigraphy: Results From the Hispanic Community Health Study/Study of Latinos Sueño Ancillary Study. American Journal of Epidemiology. 2016 Mar 2;183(6):561-73. and 0.4 (a). Matthews KA, Patel SR, Pantesco EJ, Buysse DJ, Kamarck TW, Lee L, Hall MH. Similarities and differences in estimates of sleep duration by polysomnography, actigraphy, diary, and self-reported habitual sleep in a community sample. Sleep Health. 2018 Feb 1;4(1):96-103. Here’s what a correlation of 0.5 might look like:

Figure 3. Source (a)

This is the kind of noisy sleep data all these big epidemiological studies are based on. Later, Lauderdale et al write:

the average difference at the mean of 6 hours measured sleep was 0.80 hours (48 minutes)

Now, looking at Figure 1, we can see that people who reported sleeping just below 7 hours a day had the lowest mortality. If we interpret this in light of the last quote, people who have the lowest mortality actually sleep 6 hours a day.

One last bit from the Lauderdale et al paper:

The correlation was very low (0.06) for persons with fair or poor self-rated health.

Appendix: “why did you only check Chapter 1?”

I spent more than 130 hours over the last 2 months researching and writing this essay (~5 hours to write the outline; ~60 hours to get to the first draft; ~65 hours to edit and fact-check), which constituted essentially all of my surplus free time over this time period. Continuing at the same pace, it would take me more than 3,000 hours to check the entire book. 3,000 hours is the equivalent of 75 weeks or 1.4 years of full-time work.

I hope that going through one full chapter, rather than cherry-picking stuff from across the book, demonstrated the density of errors in the book.

Appendix: “but this is just a popular science book, why such scrutiny?” – how academic urban legends are created

No, this is not just a popular science book. Walker and other researchers are actively citing the book in academic papers, propagating the information contained in it into the academic literature.

Google Scholar indicates that (a), in the 2 years since the book’s publication, it has been cited more than 100 times.

Three papers referring to Why We Sleep

A paper (a) Cardon JH, Eide ER, Phillips KL, Showalter MH. Interacting circadian and homeostatic processes with opportunity cost: A mathematical model of sleep with application to two mammalian species. PloS One. 2018 Dec 12;13(12):e0208043. published PloS One in 2018 reads:

All known forms of animal life must sleep ([1]).

“[1]” here refers to Why We Sleep. If you recall the Interlude, Walker miscites the Encyclopedia of Sleep while making this statement in Why We Sleep.
A paper (a) Shallcross AJ, Visvanathan PD, Sperber SH, Duberstein ZT. Waking up to the problem of sleep: Can mindfulness help? A review of theory and evidence for the effects of mindfulness for sleep. Current Opinion in Psychology. 2019 Aug 1;28:37-41. published Current Opinion in Psychology in 2019 reads:

Chronic sleep disturbance is a global pandemic with two-thirds of individuals failing to obtain the recommended 7–9 h of sleep each night [1].

“[1]” here refers to Why We Sleep. If you recall Section 4 and Section 5, both the “global pandemic” and the “two-thirds” assertions are false.
A paper (a) Lyon L. Is an epidemic of sleeplessness increasing the incidence of Alzheimer’s disease?. Brain. 2019 Apr 1;142(6):e30-. published Brain in 2019 reads:

the World Health Organization has pointed to a ‘global epidemic of sleeplessness’ with roughly two-thirds of adults sleeping less than 8 h a night.

The paper does cite any sources here and does not cite Why We Sleep anywhere, although the wording strongly suggests that this is where the information originated.

Two papers by Walker citing Why We Sleep

Since Why We Sleep was published, Walker published two academic papers that cited it.

The first one (doi) Walker MP. A sleep prescription for medicine. The Lancet. 2018 Jun 30;391(10140):2598-9. was published in The Lancet in 2018.

The second one (doi) Walker MP. A Societal Sleep Prescription. Neuron. 2019 Aug 21;103(4):559-62. was published in Neuron in 2019.

This is how academic urban legends (a) Rekdal OB. Academic urban legends. Social Studies of Science. 2014 Aug;44(4):638-54. are created.

Appendix: common objections

“But don’t many people get 8-9 hours of sleep when they don’t restrict sleep?”

Yes, they do. And when people don’t restrict their food-eating, many of them start eating much more and become obese.

Further, if we only care about longevity, we know that, for many animals, radical caloric restriction (a) makes them feel lethargic but live longer. Same could be true for sleep.

“In Chapter 1, Walker writes ‘vehicular accidents caused by drowsy driving exceed those caused by alcohol and drugs combined’. This shows how dangerous it is to not sleep and you have not refuted this part.”

I did look into this. I was not able to find any data on vehicular accidents caused by drowsy driving. However, the data by the National Highway Traffic Safety Administration on accidents that involve drowsy driving (a) and that involve drugs and alcohol (a) does not support this assertion.

According to this data, 1.2-1.4% (Table 2 in linked file) of car crashes involved drowsy driving, while 2.8% (Table 7 in linked file) of car crashes involved driving while having the alcohol blood concentration above the legal limit in the US (and this is not including accidents involving drugs).

I have not investigated Walker’s claims about sleep and learning, since he does not make any concrete statements about this in Chapter 1. However, is there any reason to expect his treatment of sleep and learning to by any more accurate than, for example, his treatment of the relationship between sleep and longevity?

“In Chapter 15, Walker writes that ‘after a thirty-hour shift without sleep, residents make a whopping 460 percent more diagnostic mistakes in the intensive care unit than when well rested after enough sleep’. This shows how dangerous it is to not sleep and you have not refuted this part.”

Is there any reason to expect that this is one statistic he decides to portray accurately? Have you considered looking at the scientific literature and seeing if this number checks out?

Appendix: my personal experience with sleep

This is not medical advice, but from my experiments in sleep deprivation and from reading the literature, it appears that consistently sleeping less than 3-4 hours during the night is definitely not enough for most people, but sleeping 4-5 hours + taking naps during the day is sustainable. I’ve been experimenting with sleep for the last month and I converged on a formula of sleeping 4-5 hours per night and then doing as many 20-30 minute naps as I need during the day (1-2 usually). If I know I won’t be able to take as many naps as I need during the next day, I just sleep the full 7-8 hours.

Here’s a report (a) from a person who I trust a lot describing how they cut their sleep from 8 hours to 6.3 hours a day and sustained it for more than a year.

Here’s a detailed story (a) with many practical tips from a friend of mine who experimented with many different ways of sleep reduction, including hardcore polyphasic sleep, and settled on the same 6.3 hours of sleep per day schedule.

Here’s the best general overview (a) of these styles of sleep that I found.

I hope to write the synthesis of everything I know about healthy and efficient sleep habits in the future.

Appendix: a strong contender for the single most absurd paragraph in the book

In Chapter 8, Walker discusses the relationship between sleep and cardiovascular health. In the first paragraph of this discussion, he mentions two studies. He seems to completely misrepresent both of them. Most notably, in the description of both of these studies he inflates their sample sizes. In one case, the study’s 474,684 people turn into “more than half a million”. In the other case, 2,282 people turn into “over 4,000”.

Here’s what Walker writes:

Unhealthy sleep, unhealthy heart. Simple and true. [1] Take the results of a 2011 study that tracked more than half a million men and women of varied ages, races, and ethnicities across eight different countries. Progressively shorter sleep was associated with a 45 percent increased risk of developing and/or dying from coronary heart disease within seven to twenty-five years from the start of the study. [2] A similar relationship was observed in a Japanese study of over 4,000 male workers. Over a fourteen-year period, those sleeping six hours or less were 400 to 500 percent more likely to suffer one or more cardiac arrests than those sleeping more than six hours. [numeration mine]

Although he does not cite the two studies he discusses, he gives enough identifying information that I believe I was able to find both of them.

The first one is Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies (a). Cappuccio FP, Cooper D, D’elia L, Strazzullo P, Miller MA. Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. European Heart Journal. 2011 Feb 7;32(12):1484-92. Here are the characteristics of the paper that match Walker’s description:

published in 2011
sample size approximately 500,000
8 different countries
increase in coronary heart disease of approximately 45% in the short sleep duration condition
follow-up between seven and twenty-five years

Here are the issues in his description of the paper:

the sample size is 474,684, not “more than half a million”
Walker writes, “Progressively shorter sleep was associated”, implying that shorter the sleep, the higher the incidence of disease. Instead, the study found that both short and long sleep duration were associated with increased risk of developing or dying from coronary heart disease
Walker fails to mention that short sleep was not associated with total cardiovascular disease (the statistic we care the most about), while long sleep was positively associated with total cardiovascular disease

The second study he appears to describe in that paragraph is The effects of sleep duration on the incidence of cardiovascular events among middle-aged male workers in Japan (a). Hamazaki Y, Morikawa Y, Nakamura K, Sakurai M, Miura K, Ishizaki M, Kido T, Naruse Y, Suwazono Y, Nakagawa H. The effects of sleep duration on the incidence of cardiovascular events among middle-aged male workers in Japan. Scandinavian Journal of Work, Environment & Health. 2011 Sep 1:411-7. Here are the characteristics of the paper that match Walker’s description:

Japanese male workers
fourteen year follow-up
4-5x figure is given for one of the outcomes that resembles cardiac arrests

Here are the issues in his description of the paper:

the sample size is 2,282 not “over 4,000”
the entire “those sleeping six hours or less were 400 to 500 percent more likely to suffer one or more cardiac arrests than those sleeping more than six hours” sentence is false (see Table 2 in the paper)
- the association Walker talks about compared those sleeping strictly less than 6 hours to those sleeping between 7 and 7.9 hours, while Walker says that the comparison was between those sleeping 6 hours or less and those sleeping more than 6 hours
- the figure was approximately 4x to 5x, which is the same as an increase of 300 to 400 percent, not 400 to 500 percent
- the paper does not have any statistics about “one or more cardiac arrests”
  - the figure Walker mentions is referring to “coronary events” which the paper describes as “myocardial infarction or angina requiring catheter or surgical intervention”, not “cardiac arrests”
  - deaths from cardiac arrests were included in “Cardiovascular events” category and were not studied separately

Appendix: sleep and testicles; sleep and testosterone

Walker begins his 2017 talk at Google with:

Men who sleep 5 hours a night have significantly smaller testicles than those who sleep 8 hours or more. In addition, men who routinely sleep 5 to 6 hours a night will have a level of testosterone which is that of someone 10 years their senior.

Walker begins his 2019 TED Talk with:

Men who sleep 5 hours a night have significantly smaller testicles than those who sleep 7 hours or more. In addition, men who routinely sleep just 4 to 5 hours a night will have a level of testosterone which is that of someone 10 years their senior.

This (a) Zhang W, Piotrowska K, Chavoshan B, Wallace J, Liu PY. Sleep duration is associated with testis size in healthy young men. Journal of Clinical Sleep Medicine. 2018 Oct 15;14(10):1757-64. appears to be the only paper that examined the relationship between sleep and testicles size. It does not support either of Walker’s claims.

Appendix: fatal familial insomnia

Once more, on pages 4-5, Walker writes:

[T]here is a very rare genetic disorder that starts with a progressive insomnia, emerging in midlife [fatal familial insomnia (FFI)]. [1] Several months into the disease course, the patient stops sleeping altogether. By this stage, they have started to lose many basic brain and body functions. [2] No drugs that we currently have will help the patient sleep. [3] After twelve to eighteen months of no sleep, the patient will die. Though exceedingly rare, [4] this disorder asserts that a lack of sleep can kill a human being. [numeration mine]

Statement [1]:

Several months into the disease course, the patient stops sleeping altogether.

This is false. Here’s a description of a case of FFI (a) from the case report paper I referenced in Section 3 (a): Schenkein J, Montagna P. Self-management of fatal familial insomnia. Part 2: case report. Medscape General Medicine. 2006;8(3):66.

Until roughly the 23rd month, DF’s sleep patterns showed a definite cycle, which may have reflected his rotating schedule of the various medications. The first night, he slept well; the second night, less well; and the third, still less, followed by 1–2 sleepless nights. Then the cycle repeated.

Statement [2]:

No drugs that we currently have will help the patient sleep.

This is false. From the same paper:

Gamma-hydroxybutyrate (GHB) was administered during the last month of DF’s life. … According to his caretaker, GHB resulted in sleep within 30 minutes of administration, but did not last long enough for DF to feel rested.

Finally, statement [3] is also false. In case you suspect that he was only writing about the typical course of the disease when writing “After twelve to eighteen months of no sleep, the patient will die”, in Chapter 12, Walker clarifies:

Every patient diagnosed with the disorder [FFI] has died within ten months, some sooner”

Setting aside the question of how all the patients who died within ten months of diagnosis were able to get “twelve to eighteen months of no sleep” – if, as Walker writes, the patient loses sleep several months into the disease course and dies within eighteen months of no sleep, then life expectancy from the onset of the disease is no more than two years. However, for example, this paper (doi) finds that: Montagna P, Cortelli P, Avoni P, Tinuper P, Plazzi G, Gallassi R, Portaluppi F, Julien J, Vital C, Delisle MB, Gambetti P. Clinical features of fatal familial insomnia: phenotypic variability in relation to a polymorphism at codon 129 of the prion protein gene. Brain Pathology. 1998 Jul;8(3):515-20.

Detailed analysis of 14 cases from 5 unrelated families showed that patients ran either a short (9.1+ 1.1 months) or a prolonged (30.8 + 21.3 months) clinical course according to whether they were homozygote met/met or heterozygote met/val at codon 129.

See discussion of statement [4] in Section 3.

Basically everything in Walker’s description of the disease is wrong, aside from the fact that people suffering from it die eventually.

Appendix

Most sleep does not serve a vital function: Evidence from Drosophila melanogaster

No relation to Walker’s book, but this paper (a): Geissmann Q, Beckwith EJ, Gilestro GF. Most sleep does not serve a vital function: Evidence from Drosophila melanogaster. Science Advances. 2019 Feb 1;5(2):eaau9253. is really fun:

Sleep appears to be a universally conserved phenomenon among the animal kingdom, but whether this notable evolutionary conservation underlies a basic vital function is still an open question. Using a machine learning–based video-tracking technology, we conducted a detailed high-throughput analysis of sleep in the fruit fly Drosophila melanogaster, coupled with a lifelong chronic and specific sleep restriction. Our results show that some wild-type flies are virtually sleepless in baseline conditions and that complete, forced sleep restriction is not necessarily a lethal treatment in wild-type D. melanogaster. We also show that circadian drive, and not homeostatic regulation, is the main contributor to sleep pressure in flies. These results offer a new perspective on the biological role of sleep in Drosophila and, potentially, in other species.

No, not every living creature generates a circadian rhythm

In the third paragraph of Chapter 2, Walker writes:

[E]very living creature on the planet with a life span of more than several days generates this natural [circadian] cycle

This is false. Brewer’s yeast (S. cerevisiae):

live for more than 20 days (Figure 1 in this paper (a) Fabrizio P, Longo VD. The chronological life span of Saccharomyces cerevisiae. Aging Cell. 2003 Apr;2(2):73-81. )
do not generate a circadian cycle [1 (a) Merrow M, Raven M. Finding time: A daily clock in yeast. Cell Cycle. 2010;9(9):1671–2. , 2 (a) Wildenberg GA, Murray AW. Evolving a 24-hr oscillator in budding yeast. Elife. 2014 Nov 10;3:e04875. ]

There are two ways to view this error:

this is a minor point and most people would think that he only means animals anyway
this is a minor but fundamental point. Walker doesn’t know even basic facts about sleep in living creatures and misleads readers about them. Counterexamples to “universal” phenomena are important

Circadian rhythm in general is overrated [1 (a) Hazlerigg DG, Tyler NJ. Activity patterns in mammals: Circadian dominance challenged. PLSS Biology. 2019 Jul 15;17(7):e3000360. , 2 (doi) Bloch G, Barnes BM, Gerkema MP, Helm B. Animal activity around the clock with no overt circadian rhythms: patterns, mechanisms and adaptive value. Proceedings of the Royal Society B: Biological Sciences. 2013 Aug 22;280(1765):20130019. ].

Extended quote about the dangers of lack sleep from Chapter 1

I doubt you are surprised by this fact, but you may be surprised by the consequences. Routinely sleeping less than six or seven hours a night demolishes your immune system, more than doubling your risk of cancer. Insufficient sleep is a key lifestyle factor determining whether or not you will develop Alzheimer’s disease. Inadequate sleep—even moderate reductions for just one week—disrupts blood sugar levels so profoundly that you would be classified as pre-diabetic. Short sleeping increases the likelihood of your coronary arteries becoming blocked and brittle, setting you on a path toward cardiovascular disease, stroke, and congestive heart failure. Fitting Charlotte Brontë’s prophetic wisdom that “a ruffled mind makes a restless pillow,” sleep disruption further contributes to all major psychiatric conditions, including depression, anxiety, and suicidality.

Perhaps you have also noticed a desire to eat more when you’re tired? This is no coincidence. Too little sleep swells concentrations of a hormone that makes you feel hungry while suppressing a companion hormone that otherwise signals food satisfaction. Despite being full, you still want to eat more. It’s a proven recipe for weight gain in sleep-deficient adults and children alike. Worse, should you attempt to diet but don’t get enough sleep while doing so, it is futile, since most of the weight you lose will come from lean body mass, not fat.

Add the above health consequences up, and a proven link becomes easier to accept: the shorter your sleep, the shorter your life span.