How to Improve Your Sleep

post by lynettebye · 2019-10-27T23:28:44.228Z · LW · GW · 19 comments

Contents

  1.  Are you fatigued?
  2.  Are you giving yourself enough time to sleep?
  3. Are you exercising regularly?
  4. Are you practicing solid sleep hygiene?
    The feeling of sleepiness
    Wake up at the same time
    Avoid stimulants and alcohol too close to bedtime
    Have a dark night and bright morning
    Take Melatonin
    Sleep cool
    Keep the noise down
  5.  Are you depressed?
  6.  Are you having trouble falling or staying asleep? Insomnia
    CBT-I
    Stimulus Control Therapy
    Phototherapy
    Sleep restriction therapy
    Cognitive therapy
    Progressive relaxation therapy
    Sleep aids
  7.  Do you have a health issue causing fatigue?
  8.  Do you have a sleep disorder?
  9.  What else can you do during the day to manage energy?
    Stimulants
    Short naps
  Sources and Notes
None
19 comments

This piece is cross-posted on my blog here.

Do you feel sleepy enough during the day that it impairs your ability to function? Do you fall asleep involuntarily during the day? Do you feel tired enough that you feel weak, have difficulty doing things, or feel unable to focus?

If you answered yes to any of these questions, here’s a set of guidelines to help you figure out what’s worth trying. I go over generally applicable advice for daytime sleepiness and fatigue, then cover signs that would indicate more rare causes you can investigate. If feeling tired is significantly impairing your productivity, the information value of trying these experiments is probably well worthwhile.

This infographic checks for the most common causes of fatigue and what you should do it for each. These common causes account for the majority of people with daytime sleepiness, so there’s a good chance they cover the most relevant information for you. If you’ve already read a bunch about sleep, you’ll find loads more detail in the eleven pages of research that follow.

_This guide was heavily influenced by UpToDate.com, an evidence-based clinical resource, and interviews with medical professionals. But, just so we’re clear, I’m not a medical professional and these suggestions aren’t medical advice. Listen to the person who actually went to medical school if they give you different advice. _

1. Are you fatigued?

If you can fall asleep during the day in less than 8 minutes, you're probably sleep-deprived and may have a sleep disorder. You might have fatigue (or excessive daytime sleepiness) if:

If you experience the above more than once a week or have some weeks where you’re just tired the whole time, it’s probably worthwhile for you to spend time trying to solve the issue. (It’s probably not fatigue if you just feel tired right after a trip when jet-lagged, or if you had to pull an all-nighter.)

What works to make you feel well-rested will probably be different from others, particularly people who always feel rested by default. So, you can approach your sleep as a hits-based experiment. It might take you some trial and error to find what works for you, but one big win justifies a lot of little experiments. So if sleep is a big problem, the information value of trying these experiments is quite high.

2. Are you giving yourself enough time to sleep?

First and foremost, are you giving yourself 7 to 9 hours in bed every night? If you’re giving yourself less than 9 hours in bed, you’re sleeping through the night, and you’re tired during the day, then we found your problem - or at least the first problem you need to address.

Adults usually need between 7 and 9 hours of sleep per night, and regularly sleeping outside the normal range may be a sign of a serious health problem or, if you’re doing it voluntarily, may compromise your health and performance. Spending more time in bed can be counterproductive if you have insomnia, but you probably don’t have insomnia if you’re sleeping soundly through the night.

If you aren’t sure spending the extra time asleep is worth it, I encourage you to read Why We Sleep. For now, we’ll content ourselves with this tongue-in-cheek summary of the new wonder drug, sleep, which “makes you live longer. It enhances your memory and makes you more creative. It makes you look more attractive. It keeps you slim and lowers food cravings. It protects you from cancer and dementia. It wards off colds and the flu. It lowers your risk of heart attacks and stroke, not to mention diabetes. You’ll even feel happier, less depressed, and less anxious.”

Start by seeing how long you sleep without an alarm when you can afford to let yourself sleep in. This experiment is good for checking how much sleep you need. In general, waking up at a consistent time is a good sleep hygiene practice. It may take you several days of sleeping more than you normally need to work off the sleep debt before you can tell how much sleep you normally need. This could completely solve the problem, and it’ll save you wasting time having a doctor tell you to try sleeping more.

3. Are you exercising regularly?

Exercise for at least two hours per week, ideally 30 minutes per day. Regular exercise has small-to-medium beneficial effects on total sleep time, sleep efficiency, sleep latency, and sleep quality, plus it can reduce fatigue in general. It can also be good for waking up midday if you’re feeling sleepy. Both cardio and strength training will help, so just pick a form of exercise you enjoy.

It seems fine to exercise even if you’ll go to bed in a couple of hours. While this meta-analysis found that evening exercise didn’t have a significant impact on sleep latency, sleep efficiency, or total sleep time, it suggests that vigorous exercise within one hour of bed might keep you awake longer and make you sleep less. My best guess is that it’s somewhat better to exercise earlier in the day, but better to do it close to bed than not at all if you can’t do it earlier.

4. Are you practicing solid sleep hygiene?

According to Doctors Bertisch and Peña, you should see noticeable improvement within four to six weeks of consistent practice if sleep hygiene is going to help you. If you’ve already spent a lot of effort to improve your sleep hygiene and it didn’t work, it might be worth skipping to the other sections.

The short summary of tips:

  1. Wake up at the same time every day, including weekends.
  2. Don’t take stimulants closer to your bedtime than your body clears them out - a rough guess is about 10 hours before bed for 1 cup of coffee, and at least 15 hours before bed for 100mg of modafinil.
  3. Don’t drink alcohol close to your bedtime - probably good to stop drinking 4 or 5 hours before bed.
  4. Avoid blue light for a couple of hours prior to your bedtime. This includes all LED lights and flat screens from digital devices. Turning off all LEDs and screens might be a challenge, but you can somewhat reduce the exposure on screens by using blue-light reducing programs such as Flux and using Warm White lights in the evening.
  5. Expose yourself to bright light shortly after waking up, ideally natural, full-spectrum light.
  6. Take 0.3mg of melatonin 30 minutes before you go to bed.
  7. Sleep in a cool room, between 60-67 degrees Fahrenheit.
  8. Take a warm shower before bed.
  9. Make your bedroom as quiet as possible at night, particularly if you notice noise is waking you up.

The long version of these tips:

The feeling of sleepiness

Before I talk about standard sleep hygiene tips, some context on how sleep works. Two factors govern the experience of sleepiness: your circadian rhythm and homeostatic sleep drive. First, the circadian rhythm aligns your desire to sleep, be awake, and even when to eat, with your inner biological clock. Now, this clock checks if it’s light out, and adjusts it’s timing a bit to match the daylight. When it sees it’s dark out, it releases melatonin to let your body know it’s time to sleep now. Then in the morning when light filters in through your eyelids, melatonin gets shut off to wake you up. The circadian rhythm can only adjust itself a bit each day, which is why you feel jetlagged for a few days when you try to suddenly change your rhythm by several hours.

Meanwhile, your homeostatic sleep drive builds pressure to sleep by dialing down alert-promoting brain regions and dialing up sleep-inducing parts of the brain. A key part of this drive, the chemical adenosine, increases each minute you’re awake. After 12 to 16 hours of being awake, adenosine will be enough to make you irresistibly sleepy. You can temporarily pause the experience of sleepiness caused by adenosine by consuming caffeine, but the full force of the built up adenosine will hit you when the caffeine wears off. So be ready to crash hard.

These two forces act independently. If you’re sleeping a normal schedule, they will line up; your circadian rhythm will signal you to go to bed around when the adenosine is getting high enough to demand sleep, and the adenosine will be low after a good night’s sleep when your circadian rhythm is signaling to rise and shine. If you pull an all-nighter or are jetlagged, then these two forces may act on you at different times.

Together, these forces trigger a host of other chemicals that move you along the spectrum of sleep and wakefulness. Along this spectrum, excessive daytime sleepiness is deficient daytime arousal, and insomnia is excessive nighttime arousal.

So, roughly in order from most to least likely to help:

Wake up at the same time

So, your circadian rhythm is pretty important. If you don’t sleep at consistent times, you increase the risk that you’re not sleeping at the times when your circadian rhythm thinks you should be and vice versa. A consistent sleep schedule can help these line up.

Ideally, you would go to bed at the same time each night, and wake up around the same time each morning. Sleeping on a consistent schedule is one of the more difficult sleep hygiene tips. Both Dr. Bertisch and stimulus control therapy emphasize waking up at the same time, so I’d focus on this one if you want to pick just one.

Also, don’t use snooze on your alarm (if you have to use an alarm); fragmenting your sleep is worse than waking once.

Avoid stimulants and alcohol too close to bedtime

Too close to bedtime will differ by person, but avoiding caffeine after lunch is a good rule of thumb unless you’ve experimented and found different results for you personally.

For caffeine, one study found 400mg 6 hours before bed still had a huge impact on sleep - the caffeine reduced average total sleep time by an hour and reduced average sleep efficiency 9% compared to the placebo. It also increased the average sleep latency by 24 minutes (from 20 to 44) (though this result only had a p-value of .13).

That makes sense based on the half-life of caffeine. It takes about 5 hours to clear half the caffeine out of your system, so that means that after 6 hours there was still close to 200mg of caffeine. It’s understandable that the equivalent of two cups of coffee would make the study participants feel more alert while they’re trying to sleep. If we assume that 25mg of caffeine won’t interfere with sleep, then you would need to leave at least 15 hours between taking 400mg and going to bed, and at least 10 hours for 100mg. For stimulants with longer half-lives, you would need to take it a correspondingly longer time before bed (e.g. for modafinil, half of it will still be in your system 13 hours or more after taking the drug).

It’s probably worth experimenting to see how long you’re impacted by stimulants - the time it takes your body to eliminate them can vary a lot. Also, according to a nurse I spoke with, your brain desensitizes to the overload of chemicals, so the acute alertness effect is probably gone after around 15 hours even if you still have a dose in your system. However, there may still be residual effects from the remaining dose, e.g. trouble sleeping.

Similarly, drinking alcohol shortly before bed has a net negative impact on sleep. Alcohol reduces sleep latency (time to fall asleep) but decreases sleep quality and delays REM sleep. High doses of alcohol (4+ standard drinks) had a significant negative impact on sleep even when taken 4 hours before bed, but drinking 5 hours before bed was significantly better than drinking just before bed.

Have a dark night and bright morning

Bright light, particularly blue light, delays melatonin production. Delaying melatonin production keeps you awake longer immediately, and can reset your circadian rhythm to a later time over a longer time scale. Electronic devices and LED bulbs emit a lot of blue light, so we usually get too much blue light late at night.

Turning off your electronics 2-3 hours before bed, using flux or other programs that reduce blue light from your devices at night, using Warm White bulbs, and wearing blue-light blocking goggles might reduce the disturbance. Taking a melatonin supplement before bed might help correct the delayed melatonin production with far less hassle, but the sleep specialists I asked were less encouraging. So probably worth at least testing flux in addition to melatonin.

Darkness while sleeping might reinforce that it’s time to sleep. In order from least to most hassle: Buy a sleep mask. Install blackout blinds. Cover all of the little lights in your room with black tape (e.g. the smoke detector light). If you don’t have blackout blinds, try moving your bedtime earlier to see if you sleep more when the sun isn’t waking you up (it takes a week or two to adjust your bedtime). This last idea didn’t work for me, but the sleep specialist thought it might help when trying to sleep more, perhaps because the extra sleep time is at night instead of during daylight.

Bright light in the morning will probably increase your alertness by reducing melatonin (small to medium effect size). If your room doesn’t have bright natural sunlight in the morning, lots of bright bulbs can simulate the effect. I have 10,000 Lumens in my light fixture (I affectionately call it the sun), and I know people with way more [LW · GW].

Take Melatonin

Melatonin isn’t a standard sleep tip but seems like a mostly safe experiment to try.

According to Slate Star Codex and Gwern, taking melatonin before bed will probably help you fall asleep more quickly and improve your sleep quality. The studies Gwern quotes found melatonin reduced the average sleep latency (the time it takes to fall asleep) by 3.9 to 16.8 minutes, a comparable amount to sleep pills. Furthermore, sleep efficiency increased by 3.1%, and sleep duration increased by 13.7 min.

However, the sleep specialists were more pessimistic that melatonin would help as a sleep aid. Dr. Cooper thought taking melatonin was fine if you also shut off blue screens. Dr. Bertisch thought melatonin probably wouldn’t help with insomnia, but it was “not unreasonable” to try. Dr. Zhou wrote that the “evidence for melatonin as a hypnotic is weak” and “there have no been long-term studies of the safety of melatonin, and there are studies showing that the non-regulated OTC industry often does not sell what it promises to when it comes to melatonin.”

I looked into the worries that melatonin isn’t regulated by the FDA, and that commercial supplements of melatonin have been found to contain more or less than the amount claimed on the label. Both are true, but it’s probably not a big deal. The study found that >80% of the supplements studied differed from the amount on the label by less than 50%. If you’re aiming for 0.3 MG, you’re going to be fine if it differs from that amount by 50%. If you notice that you sleep worse after taking even a small dose, you might want to take even less in case you’re accidentally getting too much. However, there aren’t long term studies, so use your judgment about taking melatonin regularly over periods longer than three months.

For use as a simple hypnotic (sleep aid), melatonin is most commonly taken shortly before bed to immediately help you sleep (Gwern recommends taking melatonin 30min before you want to sleep). The recommended dose is around 0.3mg or less. However most pills contain at least 10 times that amount, so check the amount before you purchase for one with 300MCG like this one.

Sleep cool

The core body temperature naturally drops at night as skin blood flow increases, drawing body heat away from the core.

This has been experimentally tested by slightly warming skin so that more blood flowed there, causing the core temperature to drop and the participants to fall asleep more quickly. One study found that slightly warming the skin (0.4°C) reduced the sleep latency (time to fall asleep) by 1.84 minutes in healthy participants and 2.85 minutes in insomniacs. Taking a warm shower or splashing warm water on your face before bed may have the same impact.

A cool bedroom is presumed to improve sleep by mimicking this temperature drop. Most of the sources I looked at suggested between 60-67 degrees Fahrenheit, apparently because some studies indicated hot rooms decreased sleep quality. Open the window or turn on an AC if it’s too warm. I’ve heard a few strong recommendations for a ChiliPad to maintain the proper sleep temperature.

Keep the noise down

Try keeping your sleeping area quiet to avoid decreasing sleep quality. This study found that traffic noises between 39 and 50 dBA (with a maximum level up to 74 dBA) reduced total sleep time by 16 minutes and decreased sleep quality compared to a 32 dBA pink noise control. 32 dBA is about a whisper, while the other sounds range from average home noise to inside a car going 60 miles per hour. This larger study found that there were no significant changes in sleep structure if the sleep disturbances “did not exceed 4×80 dB(A), 8×70 dB(A), 16×60 dB(A), 32×55 dB(A) and 64×45 dB(A) in a single night.” So a good target seems to be noise levels near whisper levels while you sleep (or at least under normal conversation levels), with few spikes in noise levels.

Try earplugs or white noise if noise bothers you. You can test white noise cheaply on your phone, and buy a white noise machine if you like it. If you want to go a step further, install soundproofing. Insulating around doors and windows (anywhere air can get through) seems particularly important.

5. Are you depressed?

Don’t skip this right away, even if you think the answer is probably no. Depression affects 1 in 15 people, many of whom are not aware that they are experiencing depression. (I’ve known several people who didn’t realize they were depressed, sometimes for many years.) So I encourage you to err on the side of taking the PHQ-9, a short screening tool.

Only a doctor can diagnose you with depression, but if you’re feeling miserable all the time or you score high on the scale, you can probably make a pretty good guess. You might want to retake the scale every week for a few weeks if you’re on the low end – even mild depression some of the time may be worth talking with a doctor about.

If you’re concerned, go see a doctor. Your doctor can prescribe medications that might help a lot. Wellbutrin (generic bupropion) might be a good one to try if fatigue is one of your main symptoms.

If you think you might have depression but aren’t ready to see a doctor, I recommend this Slate Star Codex article.

6. Are you having trouble falling or staying asleep? Insomnia

If you give yourself plenty of time in bed but you can’t fall asleep or you wake up during the night, you probably have insomnia. It might often take you 30 minutes or more to fall asleep or you might sleep six hours or less on three or more nights per week despite spending enough hours in bed. You probably also won’t be able to nap during the day, despite being tired.

If that’s you, then the best place to start is sleep therapy and sleep hygiene. Behavioral therapies can help with falling asleep and managing waking up in the night. CBT for Insomnia (CBT-I) can include relaxation therapy, stimulus control therapy, cognitive therapy, phototherapy for adjusting sleep times, and sleep restriction.

You can experiment with these yourself at home, but talk to a doctor if you aren’t seeing results after four to six weeks. According to Dr. Bertisch, “Self-help books for CBT-I are helpful, so it's something many people can do on their own.” However, she warns that patients could become discouraged if the self-therapy isn’t working and be less likely to talk to a doctor, despite the fact that a doctor can often help even when self-help failed.

At a high-level overview, therapy reduces insomnia the most, followed by sleep drugs, followed by sleep hygiene. However, it seems likely that there is a selection effect in the insomnia cases that get included in sleep studies. Dr. Bertisch said that most people who got to the point of seeing a sleep doctor were already practicing good sleep hygiene, with the most common exceptions being not having a consistent sleep schedule and using blue-light-emitting electronic devices at night. So starting with both self-therapy and sleep hygiene seems reasonable. Again, you should see results within four to six weeks of consistent use if sleep hygiene and self-therapy are going to work.

CBT-I

CBT-I is considered the gold standard for treating insomnia. Studies find it to be more effective than sleep medication for reducing insomnia. It’s delivered one on one with a therapist, so it may be bottlenecked on finding a therapist who specializes in CBT-I. However, you can try the individual therapies yourself.

One meta-analysis found that CBT for insomnia reduced sleep latency by 19 minutes, increased total sleep time by 7 minutes, and increased sleep efficiency by 9% compared to the placebo.

Stimulus Control Therapy

This one is probably a safe option to try first, either alone or in combination with others. It aims to form an association between being in bed and sleeping (rather than being awake!), and set a consistent sleep schedule. The guidelines are as follows:

  1. Lie down to go to sleep only when you are sleepy.
  2. Do not use your bed for anything except sleep; that is, do not read, watch television, eat, or worry in bed. Sexual activity is the only exception to this rule. On such occasions, the instructions are to be followed afterward when you intend to go to sleep.
  3. If you find yourself unable to fall asleep, get up and go into another room. Stay up as long as you wish and then return to the bedroom to sleep. Although we do not want you to watch the clock, we want you to get out of bed if you do not fall asleep immediately. Remember, the goal is to associate your bed with falling asleep quickly! If you are in bed for more than about 10 minutes without falling asleep and have not gotten up, you are not following this instruction.
  4. If you still cannot fall asleep, repeat step (3). Do this as often as is necessary throughout the night.
  5. Set your alarm and get up at the same time every morning irrespective of how much sleep you got during the night. This will help your body acquire a consistent sleep rhythm.
  6. Do not nap during the day.

Phototherapy

Phototherapy uses light to shift your sleep cycle if you have a circadian rhythm disorder such as sleeping and waking up too late or too early. You can read more about shifting your sleep schedule on Slate Star Codex’s melatonin article.

Sleep restriction therapy

You limit total time in bed so that you’re tired when you go to bed and will sleep well. It usually starts with limiting you to around 6 hours in bed per night, and then gradually increases as long as your sleep efficiency stays above 85%. It can work well, but if done improperly, this therapy can leave you sleep deprived.

Cognitive therapy

Cognitive therapy with a therapist can work to reduce anxiety, including anxiety about not being able to sleep which makes it harder to sleep. I can imagine that it might be worth trying other methods of releasing stress and worry before bed, such as journaling or meditation, if anxiety is a common cause of insomnia for you.

Progressive relaxation therapy

You progressively tense and relax each muscle group over about 45 minutes. It seems to be considered less effective but is easy to try on your own.

Sleep aids

If therapy and sleep hygiene isn’t working, you can try hypnotics (sleep drugs). They have a lot of side effects but are occasionally the only thing that works. If you are going to use a sleep aid, it’s best to stick with the lowest dose for the shortest time that you can manage.

According to a pharmacology book, there are new drugs that avoid most of the problems of the old drugs, which included causing daytime sleepiness or wake maintenance problems, loss of efficacy over time, and causing worse insomnia when ceasing the drug. However, hypnotics may not be much better than Melatonin, and they can have weird side effects like causing the user to eat, drive, or have sex while asleep on the drugs with no memory of it the next day.

All that said, you can talk to your doctor if you want to try them, but it shouldn’t be your first - or third - line of action.

7. Do you have a health issue causing fatigue?

You’ll need to see your doctor if you have health concerns. If you suddenly started feeling fatigued within the past three months, you should probably talk to your doctor. It may be a sign of a health issue, e.g. mononucleosis and depression are both characterized by a relatively sudden onset of fatigue.

Your doctor will probably do blood tests to check for two common causes of fatigue, anemia and hypothyroidism, and possibly vitamin deficiencies. If your doctor doesn’t do blood tests, you can ask them politely about these. Your doctor can also check your medications to see if one of them, or a combination of them, is making you excessively sleepy.

If your fatigue is caused by other health issues, things get a bit trickier. Your doctor will probably be able to help you decide which things to test for based on your other symptoms. E.g. if you also faint a couple of times a day, your fatigue might be caused by POTS.

Finally, your primary care doctor can make other recommendations as appropriate, including if you should see a specialist or try sleep aids.

8. Do you have a sleep disorder?

If you have a sleep disorder, then the other things are unlikely to fix the problem, and you’ll want to see a sleep specialist to get it diagnosed. Note: You may have a sleep disorder and not know it. One of the sleep specialists I emailed, Dr. Zhou, said he sees “more than enough people in clinic with sleep apnea who think that shutting off electronics will help them.”

Sleep apnea is the most common sleep disorder - estimates of Americans afflicted range from 3% to 7%. If you feel fatigued and snore a lot or experience waking up gasping for breath (or if your partner reports you doing either), then you should get checked for sleep apnea. While not everyone who snores has sleep apnea, almost everyone who has sleep apnea snores.

Similarly, if you score moderate or severe on the Epworth Sleepiness Scale, you have a higher chance of having narcolepsy, although it is rare. If you have an involuntary urge to move your legs while lying still, you may have restless leg syndrome. If you can’t fall asleep until too late at night or always fall asleep too early, you might have a circadian rhythm disorder.

If you have sleep apnea, narcolepsy, or a handful of other specific sleep disorders, a sleep specialist will be quite helpful. A sleep specialist may order a Polysomnogram and Multiple Sleep Latency Test, where they monitor your sleep overnight and then during 5 naps the next day to check for disorders such as sleep apnea and narcolepsy. If they are pretty sure you have moderate to severe sleep apnea, they can do the test at home instead of in a lab.

I’ve spoken to a handful of people (4-5) who didn’t have symptoms of specific disorders and spoke to a sleep specialist, and none of them found it helpful. UpToDate says that with many patients the cause of excessive sleepiness is apparent, but others should do formal testing, particularly if they show signs of sleep disorders. So basically, you might have a sleep disorder and not know it, but there’s a good chance you and your primary care doctor can predict in advance whether it will be valuable to visit a sleep specialist.

9. What else can you do during the day to manage energy?

Identifying the cause of your fatigue is valuable for managing it long term. In the meantime, however, there are things you can do during the day to manage your energy.

Stimulants

Stimulants can be helpful, particularly for short periods or cyclically. Some build up tolerance or dependence quickly, such as caffeine. So taking them when needed or cyclically might be better than every day. Again, it’s probably best to take stimulants earlier in the day. You probably also want to do your sleep experiments before you switch to stimulants, so you can more easily tell if they are working.

For caffeine, L-Theanine is supposed to balance the jitteriness, generally a 2:1 ratio of L-Theanine to caffeine. A lot of caffeine pills come as a caffeine-L-Theanine combination. You can also buy L-Theanine separately if you drink your caffeine. Caffeine basically tricks you into not feeling sleepy by blocking adenosine, a molecule that induces the feeling of sleepiness. This can lead you to crash when the caffeine wears off because you suddenly experience all the adenosine that built up.

You could talk to your doctor about a modafinil prescription if your fatigue is extreme. Although modafinil is useful for a variety of cognitive enhancements in non-sleep deprived populations, it is unlikely to be prescribed for that purpose. (You can get a large discount via GoodRX if your insurance doesn’t cover it.) Gwern has a detailed post about modafinil here.

This Slate Star Codex post has more information on potential stimulants to try.

Short naps

Short naps might help, but be careful of longer naps further confusing your biological clock. Naps can alleviate fatigue when you are sleep deprived, but they can also increase insomnia if you already struggle with it.

A 15-minute nap is probably a good baseline to start with (it has been shown to work well, and is a good length to let you rest without entering deep sleep), and then you can experiment to see what is the ideal time for you. Some recommendations suggest trying taking caffeine with the nap, so the caffeine will be kicking in as you wake up 15 minutes later. Again, I suggest being careful not taking stimulants less than two half-lives-ish before you want to sleep.

Sources and Notes

You can check out this supplemental document for my sources and notes if you want even more information.

If you want more information, I recommend UpToDate.com for detailed guides on fatigue, insomnia, and relevant treatments. Why We Sleep also contains a plethora of information about sleep and why it’s so important.

Special thanks to the medical professionals who made time to talk or email with me regarding this article, especially Dr. Suzanne Bertisch, Clinical Director of Behavioral Sleep Medicine at Brigham and Women's Hospital and Assistant Professor of Medicine at Harvard Medical School; Dr. Laura Barger, Instructor in Medicine at the Harvard Medical School Devision of Sleep Medicine; Dr. Victor Peña, Surgeon and Health Coach; Lucas Valenca, RN; Dr. Joanna Cooper, Neurologist and Sleep Specialist; Dr. Eric Zhou, Psychologist and Faculty in the Division of Sleep Medicine at Harvard Medical School.

19 comments

Comments sorted by top scores.

comment by ChristianKl · 2019-10-28T08:45:10.758Z · LW(p) · GW(p)
If you don’t sleep at consistent times, you increase the risk that you’re not sleeping at the times when your circadian rhythm thinks you should be and vice versa. A consistent sleep schedule can help these line up.

What's the actual evidence for this claim? What does consistent times mean (how much can you derivate)? If this only an issue for people who take a long time to fall asleep or for everyone?

(note that having good and clear evidence would also make it easier to self-motivate to stick to more consistent times)

Replies from: lynettebye
comment by lynettebye · 2019-10-28T15:13:56.431Z · LW(p) · GW(p)

Good question - I don't think we know as much about that as would be ideal.

In laboratory studies and among shift workers (people working during the night and sleeping during the day), there's a bunch of studies that find negative effects if your sleeping/waking schedule is misaligned with your circadian rhythm (see the sleep timing section here https://www.sleephealthjournal.org/article/S2352-7218(14)00013-8/pdf). However, I don't know of studies testing just sleep timing in real-world normal populations.

More broadly, sleep hygiene is helpful for insomnia, but much less so than CBT-I (e.g. this meta-analysis https://www.ncbi.nlm.nih.gov/pubmed/29194467). However, according to the doctors I spoke with, sleep hygiene is often used as a first-line treatment because people can easily test it at home, and they should get results within a month of consistent use if sleep hygiene will help. It sounds like most people who have insomnia have already taken care of basic sleep hygiene, though both of the sleep specialists I interviewed (Dr. Bertisch and Dr. Barger) emphasized consistent sleep schedules as one of the most important sleep hygiene tips people commonly miss.

However, that doesn't address the specific details of what counts as consistent times. For people with insomnia, Stimulus Control Therapy indicates you shouldn't go to bed until you feel sleepy but should wake up at the same time. So if that's you, I expect setting a consistent alarm time regardless of when you went to bed might be good.

Replies from: ChristianKl, iterativecode
comment by ChristianKl · 2019-10-29T16:19:35.551Z · LW(p) · GW(p)

There's a relatively high cost in making a personal policy that results in not going to parties that result in being in bed at 1AM for it being possible to wake up at 7AM everyday which is needed on some days to be at the office early enough.

I'm not convinced that the benefit is worth that cost. Being strongly convinced would like be a requirement to be able to muster the willpower to try it out for 6 weeks.

If it's the case that what multiple sleep specialists consider to be one of the most important sleep hygiene tips people commonly miss was without evidence that would seem strange to me. It might still be good advice, but a research failure if nobody investigates when that tip is useful.

Replies from: lynettebye
comment by lynettebye · 2019-11-10T22:51:56.472Z · LW(p) · GW(p)

They don’t have RTCs of just sleep timing in everyday settings; that’s not the same as not having evidence. Sleep specialists have a theory of why sleep timing is important, RTCs of sleep hygiene that show a moderate effect on insomnia, and personal experience working with patients. It’s unfortunate that we don’t have research in finer grained detail, but there is evidence supporting the doctor’s recommendations.

Does that mean it will work for you? Not necessarily. Based on the overall performance of sleep hygiene, I would expect at most a modest improvement in sleep quality. It’s your call whether poor sleep negatively impacts your life enough that a modest improvement in expectation is worth leaving parties early for a month. It’s totally fair if the information value isn’t worth that cost to you.

comment by iterativecode · 2019-10-29T12:40:48.337Z · LW(p) · GW(p)

Good answer, I think the whole topic needs to be explored more. World wide production would increase if we could understand sleeping better.

comment by philh · 2019-10-31T18:42:31.408Z · LW(p) · GW(p)

Do not use your bed for anything except sleep; that is, do not read, watch television, eat, or worry in bed. Sexual activity is the only exception to this rule. On such occasions, the instructions are to be followed afterward when you intend to go to sleep.

I have trouble parsing that last sentence. Could you clarify?

Why is sexual activity an exception, and what counts as sexual activity? E.g. does masturbation?

Replies from: lynettebye, greylag
comment by lynettebye · 2019-11-10T22:57:17.457Z · LW(p) · GW(p)

The purpose of Stimulus Control Therapy is to avoid implicitly associating the bed with being awake. Just guessing, it seems reasonable that greylag is correct about sexual activity being a concession to practicality. I suspect that most people would resist any advice to switch their sex life to the living room.

Replies from: JohnGreer
comment by JohnGreer · 2021-02-02T20:22:00.434Z · LW(p) · GW(p)

I asked a sleep doc this:
"I’ve read a lot of advice about using the bed just for sleep and sex. I’m wondering if there’s actually some special reason why sex is an exception or if it’s just to be pragmatic (you don’t think people will actually refrain from using their beds for sex)?"

He said this:
"Good question. You’re right that it’s not about sex itself. Just that sex doesn’t really seem to interfere with the bed/sleep conditioning process. It (usually) does not take hours and hours like tossing and turning. Plus, there are reasons it may enhance ability to sleep..."

comment by greylag · 2019-11-01T19:51:37.399Z · LW(p) · GW(p)

My common-sense understanding: if you have sex and then aren’t sleepy, get up.

I assume “beds are for sexual activity and sleep”, rather than just for sleep, is a concession to practicality and comfort. Similarly, prohibiting masturbation in bed would seem counterproductive.

(I’d imagine some people would be unhappy to forego reading in bed, also, but that’s different)

comment by ChristianKl · 2019-10-29T16:22:22.111Z · LW(p) · GW(p)

Is there a self-study book for CBT-I that can be recommended?

Replies from: KevinO
comment by KevinO · 2019-11-09T13:13:19.959Z · LW(p) · GW(p)

My sleep doctor gave me a copy of "Sink Into Sleep: A step-by-step workbook for reversing insomnia" by Judith R. Davidson, PhD. I had mentioned that sometimes I'm up at night for a couple of hours. This is pretty rare for me so I haven't actually practiced CBT-I. I think my doctor was just playing it safe and/or is programmed to give out this book if people mention difficulties sleeping.

comment by shanen · 2021-02-02T20:59:06.764Z · LW(p) · GW(p)

I think it would be good if the article mentioned data sources, but perhaps I'm projecting since I have a lot of experience with them. Right now I'm using three devices to assess my sleep. One is motion based and the quality of the data is limited. The other two are wristbands that combine pulse with arm motion to automatically detect and record sleep. Both of them divide sleep into deep, shallow, and REM, but they disagree quite a bit on the actual details when they measure my sleep. (And I wear both of them on the same wrist, too.) If there is interest I can provide more details (including some impressions from older activity monitors).

As I thought about my own data going back some years, I was reminded that age is an important factor, but it doesn't appear to be mentioned in the article. I actually think I may be sleeping pretty well after adjusting for that factor.

Replies from: lynettebye
comment by lynettebye · 2021-02-03T19:06:46.671Z · LW(p) · GW(p)

Interesting. None of the sleep doctors I spoke to recommended data sources. However, they seemed to consider even at-home professional sleep tests with skepticism, so this might say more about the level of accuracy they want than about the potential usefulness of personal devices. 

As for age, I tried to focus this post on actionable advice. The non-actionable factors that influence sleep are simply to numerous for me to cover properly, and, unfortunately, however impactful aging is on sleep, reversing aging isn't (yet!) in my repertoire of recommendations. 

comment by KevinO · 2019-11-09T13:08:55.756Z · LW(p) · GW(p)

Anecdote: I have mild sleep apnea. My only symptom was feeling like my sleep was not as refreshing as it should be. I got more and more tired over time while it went without treatment.

I did an at-home test for sleep apnea which didn't find anything (I may not have correctly strapped everything to myself, unsure), but a lab test found it.

Replies from: lynettebye
comment by lynettebye · 2019-11-10T23:03:03.973Z · LW(p) · GW(p)

That sounds like what I heard – one doctor told me that home sleep tests were useless but insurance companies try to push people to use them instead of in-lab sleep tests because home tests are way less expensive. From reading elsewhere, it sounds like a home test might work if one has severe sleep apnea.

Replies from: ChristianKl
comment by ChristianKl · 2019-11-11T07:53:24.232Z · LW(p) · GW(p)

Why is that the case? Sleep apnea seems to be defined as a period with lower oxygen saturation in the blood with should be possible to be measured at home as well.

I have some experiences that suggest that I'm more likely to snore when I'm not sleeping in my own comfortable bed. It might be that the in-lab sleep test induces the sleep apnea on it's own and isn't representative of normal sleep.

Replies from: lynettebye
comment by lynettebye · 2019-11-14T18:25:40.204Z · LW(p) · GW(p)

Sleep apnea is when your airway is blocked so you’re not breathing during your sleep, which results in lower oxygen in your blood and repeatedly waking up to restart breathing. This fragments your sleep and leaves you feeling tired.

For a home test, I think the patient has to set up the equipment themselves, which I expect is more likely to result in bad data. Since doctors treat sleep apnea as a chronic condition, I would be surprised if sleeping in another bed caused it.

Replies from: ChristianKl
comment by ChristianKl · 2019-11-14T19:09:43.200Z · LW(p) · GW(p)

Having the head in a bad position can put pressure on your airways. There's more pressure on the airways when the upper cervical vertebrae are in flexion then when they are in extension. Bad pillow height can also lead to pressure on the airways.

At the last LWCW I snored the first night. After a conversation with one of the room mates I focused on not flexing the top of the spine the next night and having a better pillow height and I didn't snore the second night.

Given the reports from people who sleep with me when I'm at my home, I don't snore at home. I had reports of me snoring at another event where I stayed overnight this summer.

Sleep apnea doesn't seem to always coincide with snoring but I think pressure from the top vertebrae being in flexion could affect both.

Replies from: lynettebye
comment by lynettebye · 2019-11-15T23:49:05.867Z · LW(p) · GW(p)

Sounds possible, though most people who snore don't have sleep apnea.