Association of sleep duration in middle and old age with incidence of dementia | Hacker News Reader

Association of sleep duration in middle and old age with incidence of dementia

cnn.com

301 pointsgmays169 comments

Many forms of dementia destroy dopaminergic neurons, and this process begins much earlier than symptoms of dementia itself manifest. From what I’ve gathered from reading studies across the years, less dopaminergic capacity seems to be associated with less restful sleep.

The conclusions of this study seem to have it backwards. It’s not lack of sleep that causes dementia, it’s early dementia that causes insomnia.

Edit: it’s also possible that a third factor causes both of these. For example, both people low on dopamine and the sleep-deprived are likely to turn to coffee to keep them functional. If they drink too much coffee, they might accidentally poison themselves into dementia, because some of the Harmala alkaloids in coffee are neurotoxic and like to deposit themselves in brain fatty tissues for a long time.

Coffee causing dementia is a pretty shocking claim that I haven't heard previously. Do you have more reading on this?

I've always wondered if coffee studies are just something they give bored grad students to do. I've heard about what seems like hundreds of them in my lifetime and they contradict each other in cycles over long stretches of time.

Same! One week it’s terrible for you. The next week it’s curing cancer. I just try not to have any past a certain hour so as to not interrupt my sleep.

The behavior of drinking coffee or not is also very socially mediated-- portrayal of coffee drinkers in movies, coffee as a common activity for preliminary dates, or to meet someone new at a cafe...

All of this means that coffee drinkers are going to self select for various other features that add confounding factors. Imagine you were doing a study on marriage patterns: I wouldn't be surprised if "Drinks their coffee in the cafe itself" was more highly associated with being single. And the headline will read "Dedicated coffee drinkers less likely to find a partner"

There’s something in (1), but it doesn’t address the entire spectrum of Harmala alkaloid neurotoxicity (2) (3). It has been shown that heterocyclic aromatic amines (which Harmala alkaloids belong to) indeed damage dopaminergic neurons (4).

(1): https://hammer.purdue.edu/articles/thesis/Identification_of_... (2): https://onlinelibrary.wiley.com/doi/pdf/10.1002/ijc.20703 (3): https://www.sciencedirect.com/science/article/pii/S157002320... (4): https://academic.oup.com/toxsci/article-abstract/173/1/171/5...

https://pubmed.ncbi.nlm.nih.gov/20182054/#:~:text=In%20the%2....

this says coffee is associated with a reduced risk of dementia which is pretty much what I've read over and over again

Thank you! I was freaking out about my beloved coffee addiction.

Never blame the coffee.

That seems like a cognitive bias.

That study conflates caffeine and coffee from the very first word in the abstract. I wouldn’t trust it at all — it’s like an article about CPUs but the writer doesn’t understand the difference between a processor (called CPU) and the rest of the stuff in the computer case (which, as a whole, is also called CPU).

So I read the whole paper, and not just the abstract. The authors do not conflate caffeine and coffee in the meta-analysis, in fact, they cite several studies that document the association between coffee consumption and the reduced risk of cognitive decline.

Did you check whether those studies do this conflation? Were they epidemiological studies (which cannot prove causality), or double-blind randomized studies (which can)?

I think that burden falls on you since you're the one trying to discredit the study.

Well, yeah. The studies were all longitudinal studies relying on self-reporting and some found an increased risk for >5 cups per day. Since a ”cup” is not a well defined measurement anyway, these are not very rigorous studies. It’s baffling that scientists so readily interpret these results to mean something, when it’s infinitely more likely they’ve missed some unexpected confounders.

I found no mention of MAO inhibitors, beta-carbolines or heterocyclic aromatic amines in coffee. The authors seem to be unaware of their existence, even though they could explain the results in many ways.

https://www.nature.com/articles/s41398-019-0604-5

Those studies reported higher coffee drinkers had 31–65% decrease in the risk of AD dementia, which is quite comparable to about 65% decrease of Aβ positivity rate in higher coffee drinkers (27.14%) compared to lower coffee drinkers (17.61%). Furthermore, the relationship between higher coffee intake and lower Aβ positivity was prominent for lifetime coffee intake than for current coffee intake. This suggests that the protective effects of higher coffee intake against Aβ pathology involve the chronic effects associated with prolonged exposure rather than an acute or short-term effect.

Note: "coffee drinkers" not "caffeine"

Just coffee or anything with caffeine?

I'm really curious for a source on this, considering how much coffee I drink weekly.

If caffeine does it for you and need lots of it, eating just caffeine is probably safer than drinking a lot of coffee. Coffee beans have a lot of of other stuff and it varies from bean to bean and eating a lot of some of that stuff may not be good for your gut/brain/whatever.

If you are "addicted" to coffee sometimes you will crave it if you do just caffeine, but this goes away after a while.

Be aware also that if you want the antioxidants and other stuff in coffee beans eating just caffeine won't give you that.

There’s really no evidence that antioxidants are categorically good for you. Studies have found that some benefits of exercise are mediated through oxidative stress, and get impaired by antioxidant intake.

Just coffee. You may have a MAOA VNTR that predisposes you to like coffee.

I agree you have to examine both directions, and a common cause, but the article somewhat addresses that under "chicken or egg?" The headline is misleading.

Big idea: a common cause may be jaw development and the resulting lack of breathing (during the day and during sleep) in large parts of the population.

Jaw development has a strong claim to be the fundamental, common cause of poor breathing and sleep because it's derived from 2 fundamental differences between humans and animals: agriculture and speech.

It's a long chain of causation, and took me awhile to learn about, but it may be what you're looking for.

Of course it could also be the case that some people are just staring at screens too much before bed and getting poor sleep. But bad sleep is an epidemic and there is evidence of this all over the world, and going back hundreds of years. The books I reference tell of orthodontists from the 1800's who noticed this problem, etc. Our diets drastically changed, which caused poor jaw development, which causes bad breathing, which causes bad sleep (I learned about this from my dentist).

There are many factors that restrict airways, including weight. Remember that the average American male is about 30 pounds heavier in 2020 than in 1960 (and similarly for females). It's hard to imagine that not having an effect of breathing and sleep (and thus dementia). But I think animals can be fat and still have plenty of air for breathing; the same isn't true for humans.

Previous comment I wrote on this including recent books (2018-2020): https://news.ycombinator.com/item?id=23435964

That article is "fitful nightly sleep linked to chronic inflammation and hardened arteries". That is another correlation, not a causation. So I would suggest that bad sleep, inflammation, hardened arteries, and dementia have a common cause in jaw and airway development. This is a lot harder to see because it took place over 10,000 years and it takes 30-50 years to kill you. You can have bad breathing your whole life but still be functional. It's very related to tooth grinding/bruxism, which is an epidemic as well.

Other threads on Hacker News: https://news.ycombinator.com/item?id=24247066

> it’s early dementia that causes insomnia

> If they drink too much coffee, they might accidentally poison themselves into dementia

Oh crap

I wouldn't take anyone making a comment on HN a source of accurate and truthful data. The HN bias can be at play. Which is the bias that we believe and trust a random comment on Hacker News more than a linked research study, or the body of experts around a subject (who don't comment on Hacker News).

Please don’t upvote this nonsense. Harmala alkaloids are found in tobacco, not coffee.

I'm pretty sure harmine and harmaline are found in coffee too. I thought it was common knowledge but all I found were two papers so maybe not.

1. https://www.sciencedirect.com/science/article/abs/pii/S00243...

2. https://www.sciencedirect.com/science/article/abs/pii/S03088...

They are found in both, in high concentrations. Tobacco smoke and coffee are the two most significant sources of Harmala alkaloids in the human diet. (Source: T. Herraiz)

Well, they are found in everything, meat, soy sauce, beer, right? Tobacco is a much more common source. By the logic of your previous argument, eating yakisoba could lead to dementia.

Interestingly, your source’s paper on this says that coffee protects against brain disease:

> epidemiological studies have consistently shown that coffee drinkers have an apparently lower incidence of Parkinson's disease (PD), suggesting that coffee might somehow act as a purported neuroprotectant

https://pubmed.ncbi.nlm.nih.gov/16139309/

Either way, I’m not particularly interested in this subject, the point is that making wild medical claims without sources in an online forum is best avoided.

They are indeed found in many products, but the key is the exact nature of the specific substances. Do they inhibit MAO-A, MAO-B, or both? Are they irreversible or reversible inhibitors? Do they do other things like cause neurotoxicity? The Harmala family has dozens of compounds, some good for you (for example MAO-B inhibitors indeed prevent PD, because MAO-B inhibitors are used to treat it) and some bad, like some beta-carbolines that are indeed neurotoxic.

Again, the total amounts are what matters. It’s much easier to take in too much MAOIs from tobacco and coffee than by eating meat.

Something seeming to protect against one specific disorder does not mean that it protects against all diseases.

In fairness to the study, it's only claiming an association, not a causal link.

The OP doesn’t suggest that coffee may cause dementia but was a possibility for sleep deprived and low on dopamine individuals drinking too much coffee to stay functional, exacerbating their sleep deprivation thus rising the prospects of developing dementia.

No, I was explicitly suggesting that neurotoxic amines in coffee may be building up in some, possibly genetically predisposed heavy coffee drinkers, causing loss of dopaminergic neurons and ultimately dementia, or other neurological disorders.

This exact mechanism has been shown to be the cause of a very common neurological disorder called essential tremor. Patients have a genetic predisposition for building up tremor-inducing Harmala alkaloids due to reduced metabolism.

In contrast, a quasi-causal analysis (via Mendelian randomization) found no causal effect of sleep duration on Alzheimer's risk.

> There was little evidence to support associations of increased sleep duration with ... Alzheimer's disease risk [OR = 0.89 (95% CI = 0.67-1.18); P = 0.41].

https://pubmed.ncbi.nlm.nih.gov/31062029/

There are other forms of dementia than Alzheimer's.

Are there causal analyses of sleep on those other forms of dementia?

I'm not well versed on different kinds of studies... could you explain the differences between "causal analysis" and whatever the study linked in the article is? They mention a longitudinal study and that the "finding was confirmed in analysis using sleep duration measured by an accelerometer."

I assume a causal analysis would also determine the direction of the causality, e.g. if it turns out that people predisposed to dementia have more trouble sleeping in middle age. That definitely seems like a tall order and probably not exactly what the linked study can answer.

Accelerometers tell nothing about the actual sleep duration. Is this another one of that quack sleep doctor’s ”everyone needs 8 hours of sleep every night” studies?

Correct, the study design in the original article cannot answer causal questions (although of course it can generate hypotheses). For example, if an omniscience informed us "actually, people lose sleep in middle age as a consequence of the early stages of dementia", that would make sense as well.

The best causal analysis is interventional, but many times (such as in this case) it would be both extremely challenging and likely hard to do in any ethical fashion (I can't imagine how it could be done, but perhaps someone could).

Short of that, you're left with instrumental variables. Genetics offers a very nice instrumental variable, since the causal arrow largely goes one way if you setup your study design carefully.

That is generally performed as Mendelian randomization, which has several assumptions but, if met, has nice properties. https://en.wikipedia.org/wiki/Mendelian_randomization

Unfortunately such studies are very very hard to conduct and are never well powered

The study you cite isn't really in conflict with the parent study. They note that the number of individuals with Alzheimer's Disease in the study was too small to develop a non-linear model relating sleep duration to AD risk:

Pg. 858:

"Similarly, short and long sleep duration and poor sleep quality have also been linked with an increased risk of dementia. Although a similar J-shaped association was observed in our observational analysis, we were limited to performing only the linear MR analysis, as the non-linear MR method requires a large number of cases and individual-level data. In our linear MR analysis, we found no clear evidence that an increased sleep duration was associated with a higher risk of all-cause dementia in UK Biobank or with AD in IGAP. This is unsurprising, as the true association might be non-linear and we were limited with only 1343 dementia cases in UK Biobank. Also, IGAP does not capture non-AD dementia types and comprises an older and more heterogeneous population."

"Results for risks of dementia and AD are still too imprecise to draw any definitive conclusions. Our findings suggest that, in clinical care, attention should be paid to sleep-duration patterns and improved sleep habits could represent a potential therapeutic target for cognition."

It doesn't disprove the parent study, but it certainly is not supportive of some of the causal interpretations of the observational study.

More accurately, middle aged people who get less than 6 hours of sleep a night are 30% more likely to get dementia.

I'd expect that people who get less than 6 hours of sleep are also more stressed and less health concious though.

This guy infers. This seems like a tough study given the correlation factors.

Sure, I'd agree with the "more stressed and less health conscious" bit. But, the article does give some good reasons to suggest that either midlife insomnia is a very early symptom of dementia, or that sleep deprivation itself leads to dementia.

Nonetheless, we are talking about people who have pre-existing insomnia before being diagnosed with dementia. According to the Mayo Clinic[0], the most common causes of insomnia are:

* Stress.

* Disrupted circadian rhythms due to shift work or frequent travel.

* Poor sleep habits.

* Eating too much late in the day.

Further down the list are:

* Anxiety disorders.

* Medications.

* Other medical conditions.

* Sleep disorders.

* Caffeine, nicotine, and alcohol.

* Age (older people sleep less than younger people).

Of these, anxiety disorders; sleep disorders; and caffeine, nicotine, and alcohol seem the most interesting to me. Anxiety disorders probably have an effect much like that of chronic stress, except that simple chronic stress generally goes away eventually. Sleep disorders seem interesting because we have some evidence that sleep serves a function related to brain health, but, really, we don't know exactly what sleep does for us. And, alcohol, nicotine, and caffeine are interesting because the former two carry their own, independent health risks. Caffeine is interesting because, of course, it is active in the central nervous system, and it actually shows some limited evidence for a protective effect against Alzheimer's disease. [1]

Cross referencing that with the list of things they control for ("age (timescale), sex, ethnicity, education, marital status, alcohol consumption, physical activity, smoking status, fruit and vegetable consumption, BMI, hypertension, diabetes, cardiovascular disease, GHQ depression, and CNS medications"), we can (lightly) cross age, nicotine, and alcohol off the list of interesting things.

All this makes me wonder if either insomnia is an early dementia symptom, as they suggest, or if one of the causes of insomnia is the instigating factor. And, it seems to me the most interesting causal factors for insomnia left on the board are anxiety and stress.

Oh, and guess what? Anxiety, stress, and PTSD have been independently linked to the development of dementia! [2]

To me, that's interesting because anxiety is known to impair both working memory and cognitive processes[3], things which are both impaired in dementia.

I don't mean to suggest that anxiety and stress cause dementia all by themselves, but I do think this might point to a more significant role for anxiety and stress than is typically considered these days.

Anyway, these are just some idle thoughts from someone who is neither a doctor, neuroscientist, nor biological researcher of any kind. If anyone has anything else to add, or would care to riff off this, I'd love to see it. ---

[0]: https://www.mayoclinic.org/diseases-conditions/insomnia/symp...

[1]: https://content.iospress.com/download/journal-of-alzheimers-...

[2]: https://www.alzheimers.org.uk/blog/there-link-between-stress...

[3]: https://www.psychologytoday.com/us/blog/anxiety-fear-and-hat...

That's not more accurately at all; you're still insinuating a causal relationship.

"More accurately, people with yellow-stained fingertips are more likely to get lung cancer."

The actual study's own title puts it more accurately:

"Association of sleep duration in middle and old age with incidence of dementia"

I don't think it insinuates a causal relationship.

I interpret "30% more likely" as the same as "associated with a 30% increase," which is what the article claims.

I would say that "People who do X are 30% more likely to get X" is commonly understood to imply causation.

How would you phrase it? Because I did not read that as meaning that less sleep caused dementia.

"is associated with" is a much more neutral phrasing. The phrasing did not denote causality but does imply it to a casual reader.

"However, this study cannot establish cause and effect" is the only part of this article that matters, and it demonstrates the ineptitude of the article writer.

get some sleep

I will when I finish just one more zombie night in 7 days to die. Just one more I promise.

I agree. This sort of article is partly the reason why we so often here something like "science can't make up its mind" or "x is good now but was bad before, and will be bad again soon".

Instead of sharing accurate information about the findings of these researchers, news organizations use them to write catchy, clickbait-y articles. Needless to say, researchers are often to blame as well, since they participate in interviews using vague language so that they can benefit from the exposure, well aware of the repercussions of their behavior.

This issue goes beyond just researchers and news organizations, though. There have been plenty of discussions here in HN and other places about how scientific research is somewhat broken due to how it rewards quantity versus quality for example.

It's really hard to fight this though, because it may take many years, even decades, to thoroughly answer a follow-up question. By then, everyone has forgotten about the original reason something got published.

> researchers are often to blame as well, since they participate in interviews using vague language so that they can benefit from the exposure, well aware of the repercussions of their behavior.

I would also add that like everything in life, there's a large distribution of researchers. Some do poor research. This is true for all fields. One example is when a researcher has an hypothesis in mind which they try to force even if the results don't really support it. Doing great science is hard and researchers in the end have pressure to publish papers.

> Needless to say, researchers are often to blame as well, since they participate in interviews using vague language so that they can benefit from the exposure, well aware of the repercussions of their behavior.

Universities also have PR departments that tout their faculty research. I imagine these PR groups contribute to the hype.

Yeah, the headline isn't a conclusion that can be drawn from the study, and is guaranteed to mislead most people who come across it. Shame.

As people age and that their brains change, many end up sleeping less automatically, so it could be a common cause for both

There are lots of stories of the “I can work harder and longer than others because I require very little sleep” types having serious issues later in life. This study would seem to put some hard data behind this logic being a really bad idea for ones long term health and well being. The public discourse now seems to be shifting to one where “successful people prioritize sleep.”

I think those people already have serious issues and are escaping them through working.

It's irritating how these articles always position stuff as if us insomniacs have a choice. I'd chew my own arm off to get a solid 8 hours a night but it just doesn't happen.

I am still terrible at sleeping, but a small dose of melatonin and moving my phone away from my bed has helped me a bit. Does melatonin help you any?

Never tried, but thanks, will give it a go!

So, anyone with kids is probably going to have dementia?

Usually parents can't sleep well only for a year at most, even 6 months old children are sleeping whole nights.

That's just half of the picture. My kids are older but they have a talent to get up at 6 in the morning on weekends. And need to get up at similar times for school during the week.

If you have kids getting up at 6 for school and you haven't adjusted your sleep schedule to match, isn't that kind of on you now?

As a parent I need both my own down time and time alone with my spouse to feel mentally healthy too, and that is usually after they are in bed and asleep. Maybe more than 6 hours is easy, but it's usually less than 8 if I manage to do all my normal expected activities. Often I do sleep instead now though as I quickly realized my temperament is only agreeable with enough of it as I age, and the other things I need to do suffer. Others may not make the same choice, and it's not a wrong to prioritize other important aspects of mental and physical well being.

How does one do this? Parents get up at 6 to get their kids ready for school. Then take the kids to school, shower, have breakfast and then work until 6. Picking up the kids at some point in there. Then make dinner, family time and then kids go to bed at 8 or 9. Now you have the option of going right to sleep which is fine if you desire no personal time / partner time. But if you have your own hobbies or goals then you spend a couple hours on those and bang you are sleep deprived.

It's all about priorities, time management, efficiency, and sharing the load.

If you want to have both parents working, stay late at the office, have both parents involved with kids for every activity, and also work on side projects late at night, you're going to struggle until the kids are more self sufficient.

Even without kids, waking up at 6AM, hustling through morning routines and to the office, working to 6PM and then getting home at 6:30PM (assuming commute) leaves about 3.5 hours of free time and 8 hours for sleep. If you're trying to shoe-horn involved hobbies and side projects into those weekdays, something has to give regardless of whether or not you have kids.

The standard parent learning curve usually involves: Becoming more efficient with your routines, becoming more efficient with time management at the office, learning to say no to unnecessary requests, pushing back on jobs that demand long hours (or changing jobs to a more reasonable 8-hour workday company), becoming more efficient with meals and grocery shopping, doing more tradeoff load sharing with partner instead of both people being involved in everything, and letting kids have more independent play time as they age.

That said, you simply won't have as much alone time with kids in the equation. If you're trying to do too much, you have to sacrifice somewhere. It's better to do a few things well than to try to do everything at 75% commitment or being 75% awake.

I have and it is. I'm not sleep deprived (much). I just wanted to say that having kids influences your sleep far longer than six months and probably not for the better.

Then in their teens they may sleep 10 hours a day but that's the only time you're not worried.

You say that. I too believed as you did. I looked for the joyous day when I could sleep through the night with our first and second and third.

But it turns out toddlers will get up in the night and ask you to tuck them back in, they will get up because they had a scary dream, because they can't find their stuffed animal, and don't get me started on night terrors.

The point is I don't think this tracks.

I don't think I've had a solid week of uninterrupted sleep since our first came home.

Night terrors are not remotely normal. They are a fairly uncommon and extreme sleep issue.

My experience/understanding: Very bright kids get night terrors when you make sure they are physically tired enough to sleep but fail to make sure they are mentally tired enough to sleep. You need both to sleep deeply and bright kids who can't yet read have trouble reaching a state of being adequately mentally tired.

I suggest you join some parenting lists for people suffering from having made the mistake if giving birth to bright kids. Tagfam.org is probably still a good place to begin trying to find support resources and reliable info.

Edit: Bright kids are known to be prone to a variety of sleep issues. I spent twelve years arranging every minute of every day to make sleep feasible for one of my kids.

Ha! This is not my lived experience at all nor of my friend-group. My older kid, 5, mostly sleeps fine now, but both he and his sibling had middle-of-night issues well into their 2s. Even now, if one of my kids wakes (sometimes many hours before I would choose to wake, sometimes due to things like bed wetting), I’m not going to be finishing my nights sleep, and even if I did, the quality is simply lower.

Pre-6 months, we’re talking about a zombie lifestyle where there is no “sleeping at night,” just sleeping between feedings.

I acknowledge there are definitely both cultural and hereditary elements to this, but your comment is so starkly different from my experience and the experience of my friends.

My coffee addiction says otherwise :).

YMMV but kids have their hangups and it's often not possible to negotiate rationally. They also regress (this past year has been terrible, with no school for extended periods, staying at grandma's every couple of weeks so parents can get work done...), and you're back to square one.

Ours are 2 and 5, and I haven't slept in on a Saturday morning since 2015.

That's not my experience at all. My oldest woke up screaming with nightmares every night until she was 7 years old. That was 7 solid years of my not sleeping through the night.

So mid-life in this study is 50 to 60, if you have kids at 35 by that age your kids are probably not affecting your sleep schedule. I coudn't find out if the study found a correlation with younger patients sleeping 6 vs. 7 hours.

Also they are at 30% higher risk of something that is already kinda unlikely for most of the population if you follow a healthy lifestyle. Most of the research points towards vascular disease being the most important indicator of heightened dementia risk on an otherwise healthy population, and this study only controlled for severe vascular disease like having had a stroke or infarction.

Not doubting anyone's personal experience, but to add another anecdotal data point:

Kids weren't as disruptive to sleep as I expected after reading horror stories online. The way the internet talks about child raising sometimes makes it feel like martyrdom, but the internet tends to focus on worst-case stories and people venting about uniquely bad situations. It's almost taboo to share positive stories when other parents are sharing their struggles, so most conversations veer toward discussions from parents who are struggling the most while other parents keep quiet. I've found it helpful to identify parents who are doing well and specifically ask them what they're doing, because most parents are too polite to tell others how to raise their kids.

There are always edge cases and every child is different, so of course we should all be considerate of other parents and cut them slack and/or provide assistance when they are struggling.

However, I and most of my friends have had good results with combinations of sleep training and adjusting our own sleep schedules for additional buffer. The key for me was to go to bed an hour earlier than normal. If I get interrupted in the middle of the night, I use that extra hour to sleep in. If I don't get interrupted (most common case) I simply wake up an hour earlier than normal. It doesn't cover every night and every situation, but it fits most nights.

I’m expecting my first child in June and have sort of been preparing for misery when it comes to sleep. I appreciate you sharing this.

It can be rough. But it isn’t rough for all parents all the time. And keep this in mind: A period of bad sleep could be followed by a period of good sleep. Kids aren’t static, their behavior can vary a lot during different development stages.

Best of luck!

Thank you.

Whatever happens, know that it's ultimately not only up to you. There's quite a bit of luck involved how they are able to cope with the environment and what sleep pattern they prefer.

And OT but relevant for every new parent, sorry for giving unsolicited advice. But always remember, whenever it gets tough: "They're not giving you a hard time, they're having a hard time!". That helped me better understand that it's our job to help them through whatever crazy thing they're going through at any given time.

Thanks for this, it is the first time I’ve shared this here.

This perspective is useful. We have a cat that sometimes does stuff we wish he wouldn’t and now when unexpected stuff happens I’ve come to realize it means something is bothering him, rather than be frustrated by him.

Hopefully that translates.

Good luck bud. Better to be prepared for misery just in case.

I’m trying to treat it like Burning Man. Where if it’s your first time going, you go in with no expectations and as much preparedness as possible.

3 kids in, aged 6, 4 and 2... 12 hours a night for all of them. I guess I did the "sleep training" thing, although all I did was what I thought was common sense stuff.

Some things I did off the top of my head:

- 15 min cry buffer. More than that and check diaper, or give a bottle/feeding. Never turn on the lights during these check-in times... always use a dim (soft colored light) night light that's already on to keep things "dark". Changing diapers with little light gets easier the more you do it.

- Burping a newborn/baby is an essential technique. Don't leave the feeding until you get a good burp out of them. More often than not if the diaper is clean and they're fed, and they keep crying, it's trapped air that needs to be burped out.

- I use a sound machine with white noise which drowns out almost all external noises.

- No soothers. I let me kids figure out where their thumbs are and they use that. Too many times with my first kid I was going in to pick up the soother that dropped out of their mouth which made them cry.

- Keep their room "cool". Nobody likes sweating in their sleep.

A newborn will be waking up every 2-3 hours likely, but so far for me, by 3-5 months, the kids get to 12 hours.

Side note, I'm the father of the kids. Breastfeeding was not done by me, so I'd bottle feed from pumped milk.

Good luck!

Thank you for these specific points of advice.

To further with some views on feeding:

1. breast pumping provided us appreciated flexibility. After a crappy one, we got a good breast pump that did both at once. Wife appreciated being untethered from child for her relief. I appreciated being able to feed child.

2. We weren't too worried when we switched to formula exclusively after a while. Sooo many people get hung up on "breast is best" dialogue. Best comment I heard was a Maternal Child Health nurse pointing at a school yard of kids saying "Can you tell me which was breast fed and which was formula?"

Ultimately though, you do you.

I think my kids were breast fed for 2-3 months and then another 1-2 months from frozen milk and finally a gradual move to formula by mixing the breast milk and formula so that there's no sudden transition if the child isn't used to the taste, etc.

No qualms at all about using formula within the first 4-6 months.

Thanks to you both.

Would you mind sharing the type of breast pump that ended up better for you two?

Unfortunately work automatically blocks the site due to the word "breast" in it. Also, as an Australian this may not be helpful but that's for you to decide.

https://spectra-baby.com.au

I think we have the Spectra S2 still in the cupboard for a possible next kid.

The difference between this and a cheapo one we got because we didn't know better:

- two at a time, concurrency FTW! 10 mins instead of 20 was appreciated, especially in the middle of the night! once going, milk production waits for no one.

- much quieter

- separately adjustable timing and force(pressure). Cheapo just had a single dial for "more" or "less".

- handy nightlight built in. Not a biggy but there.

Possible negative for breast pumps in general. Wife mentioned feeling like a cow in a milking shed. But would do it again.

Thanks again.

Same experience here. We were lucky and with just a few adjustments are being able to cope quite well. It does require for me and my partner to go to bed earlier to have more of our sleep overlap with baby's, otherwise it'd be hell. But at the same time we get up earlier, too.

> However, I and most of my friends have had good results with combinations of sleep training and adjusting our own sleep schedules for additional buffer.

Bingo. Also don't insist on both being awake for every nighttime wake period and use a shift system, plus offset your go-to-bed times, and then, unless you're very unlucky or your newborn has some kind of illness, you're looking at 6ish hours of sleep a night with at least one night a week of 8+ hours (when you can each give the other a "sleep make-up night" to sleep in basically indefinitely, probably on a weekend).

Sleep train right, and you're gonna be doing this for 3 months, tops. They should be down to one feeding a night around IIRC 1 month, so it gets way better then.

3 for 3 on having our kids sleeping through the night (as in, we only have to wake up and do something if there's an unusual situation, and aren't woken up at all at least 6 weeks a night) by no later than 10 weeks. I believe some people do get unlucky and their kids just won't sleep through the night that early, but it's quite rare (again, unless your kid's sick, that changes everything).

Everyone we see having trouble with it is one or more of:

1) keeping their kid in the same room or same bed (no judgement on co-sleeping for other reasons, but continuing it past the first couple weeks seems, anecdotally, strongly correlated with kids who still won't sleep through the night at 6 months and beyond, and have trouble going to sleep in the first place),

2) pouncing on their baby the second they make a sound at night, even past the first couple weeks, at which point you should stop that and gradually increase your response times to waking/crying unless you want to train your kid to wake up a lot (by rewarding waking with attention, and not letting them learn how to transition between sleep cycles),

3) unable to handle "cry it out" for the 2-3 nights it'll probably take to finish the process, if you've followed the steps you should have to correctly reduce how often they wake by ~8 weeks or so (if you haven't, it might take a lot longer, but... just don't do that)

I'd love to go to sleep an hour earlier, but my 3year old is not asleep at that time :) And he wakes up at 5, so that's great. Maybe when he start skipping his midday nap, we'll start to get some more sleep. Until then, it's dementia I guess.

The kid still sleeps more than you need over the whole day; so you "just" need to take midday naps as well.

They paradoxically define "midlife" as "50s and 60s" so for most people, no.

Yeah, that seems kind of silly to me as well. I don't claim to speak for them, but what I think the researchers were getting at was "right before Alzheimer's symptoms typically appear," including mild cognitive impairment.

If you are 5%-8% likely to get dementia as a baseline, but don't sleep enough, you're now 6.5%-10.5% likely to get dementia.

Out of 1000 people who live to be over 70 and sleep properly, ~65 of them get dementia.

Out of another 1000 people who live to be over 70 and sleep improperly, ~80 of them get dementia.

So an extra 15 people out of 1000 get dementia in the "sleep improperly" group.

Someone please check my application of natural frequencies!

We co-slept with our kids until they were 3-ish. Made sleeping a pleasant experience for everyone involved.

IIRC there are studies that parenting small kids (not to those giving birth but to those actually raising infants) cuts off 1-2 years of expected lifespan, presumably due to the associated sleep disturbances.

I always wonder about the direction of causality (if any) in these studies.

I’m fortunate to mostly live without an alarm clock. I naturally get up before any scheduled activities, only early morning flights (previously) or EU time zone meetings require an alarm. Even without an alarm, I typically wake up feeling fully refreshed after 5-6 hours, rarely more than 7.

If I end up with dementia (family history indicates likely), is the same root cause of dementia causing the 5-6 hours of sleep now, the other way around, or neither?

Not an expert. But if you're fully refreshed I think you're fine. Many people would take it, that's for sure. Research is useful. But nothing is more useful than listening to your own body. When something is wrong there are usually signs.

I'm screwed then. Due to a mental issue and my sleep paralysis I'm lucky if I sleep 3-4 hours a night as much as I would like to sleep more.

Worrying about sleeping enough is certainly more detrimental to one’s health and well-being than getting too little sleep.

That little sleep has way more possible dire consequences that just dementia. There's literature all over the place about it, I don't even know where to start. How old are you? I don't know anyone that can live with getting that little sleep unless they are also on amphetamines.

Turning 40 in August. I haven't slept this badly all my life. It's got worse over the years. Specially the last 5 or so. I'm bipolar II, I take my meds as I'm supposed to, but I still seem to frequently cycle with manic episodes that keep me up working on projects all hours of the night. I then refuse to sleep all day, so I'll go to bed at say 3-4am and get up by 9am. Nights I can go to bed, I suffer from horrible sleep paralysis and that will have me waking me up throughout the night. Some of it is pretty disturbing, at least to me. So I end up giving after it happening several times in a night and just not sleeping any longer.

Sleep issues run in the family though. My father is a logger and I have never seen him sleep longer than 5-6 hours my entire life.

*EDIT* not abusing Amphetamine's BTW. Sorry forgot to even talk about that. Just lots of mania I would guess. Like right now, I have been kinda down and depressed all afternoon. It's now 6pm where I'm at, and I just had a switch go off, and now I'm full of energy and will probably end up coding till I drop. I'm my own worst enemy, I know.

I'm really sorry you have to live with that. Yeah few people understand how a manic episode works, that can definitely be masked as being a functional person but you are still harming yourself. Maybe it's time to change your medication? The brain changes as you age, maybe you can have a better quality of life if you try new stuff.

In the process of changing my meds again. Sadly they never last longer than 6 months to maybe a year. It's a frustrating process honestly. So, I tend to lie as to how well I'm doing just due to the fact I don't want to take a dive into a deep depression for months while they work out the kinks again. I guess the mania is better than the depression, even though I know it's unhealthy. Well they both are so, not sure what I can do till the science improves.

I wonder if there is a very simple effect involving blood flow to the head. When you sleep, you lie horizontally, and so presumably it's easier for blood to circulate through your entire body. Maybe if you don't get enough horizontal time, it leads to head related problems like dementia and hair loss.

Many people who don't sleep enough at night do spend the same time laying down as one with a full night sleep. Many people would spend like 5 hours sleeping and another 2 hours laying down resting.

~6 hours or just above it in general seems to be the sweetspot. Most negative effects (especially broad ones like all-cause mortality) only really start showing up under 6 (or over 7 hours)[0].

Too bad the majority of reporting exaggerates and starts scaring people for going under 8 (and sometimes 9!) when yet again it seems like 6-7 is likely best, with very little 'loss' between 7 and 6. Just don't go much lower than that for long unless you need exceptionally less.

0. https://cdnsciencepub.com/doi/full/10.1139/apnm-2020-0034

Yeah, I naturally sleep around 7 hours. If I get < 6 I really notice it the next day. 6+ is a normal day. But it is weird just how much I notice going less than 6 though.

Another thing that my wife hates (ie, is jealous of), is my head hits the pillow and I'm out. Also, during that 6 hour sweet spot, I can sleep through almost anything.

My dad is like this. Please tell me as I am curious, do you tend to self-reflect or have a vivid inner world? I'm trying to understand how my dad does this and my current theory is that he simply does not dwell on things and runs mostly on instinct.

Fwiw, I can also usually do this and I would consider myself a world class dweller.

For me, it's about

- consistent bedtime

- doing as little as possible involving a screen in bed

- consistent bedtime routine

- consistent bedtime

Screwing any of these up can quite easily cause me to start worrying about things instead of falling asleep.

But I also think in part I'm just lucky and I can't really explain what causes it. I was always able to fall asleep in the car when I was younger, almost instantly. I can't fall asleep on public transport or planes at all though unless I'm really shattered.

I self reflect a lot, but don't dwell on things outside of my control. How an interaction with a co-worker went I might think about for days and even journal about, but I can't tell you what I read on HN yesterday unless it fit into something I'm working on.

I exercise almost every day and have for over 20 years. It definitely helps.

I do have a basic bedtime routine. I go to bed at a consistent time, but I don't avoid screens or anything. I do avoid games and any news type things when I'm winding down for bed, but will read and watch part (it's always a part b/c I'm asleep before it's over heh) of a show on my phone. Also, DnD kicks on about an hour before bed so a random work email doesn't derail me.

Finally, if there is something I'm worried about I take care of it. This goes back to not dwelling on things outside of my control, and dealing with the things that are.

That's not what your source says though. The consensus is less than 7 hours has adverse health outcomes and more than 9 hours has an unknown effect for most people and is beneficial for certain groups of people.

My bad, I copied the wrong link (edited now). I meant to link to the largest systematic review I could find[0]. Instead I had linked to the US recommendation which I was initially thinking of linking for the part where most over-recommend 8-9.

Relevant part:

>that compared with 7 h of sleep per day, a 1-h decrease in sleep duration was associated with 6% increased risk of all-cause mortality and a 1-h increase in sleep duration was associated with a 13% increased risk of all-cause mortality (N = 241 107 adults in 43 articles).

With again, the bigger effects being under 6 but even with an per-hour average the risk from 7 to 8 is higher than 7 to 6.

0. https://cdnsciencepub.com/doi/full/10.1139/apnm-2020-0034

Your linked consensus statement recommends against under 7 or over 9 hours, and says 7-8 is "most favourable".

I already edited and answered in another comment. The actual studies (especially large meta-analyses) show very good results between 6 and 7, I was just thinking of linking this for the overrecomendation of 8-9 and copied it by mistake.

Though even here they try to reccomend more in the abstract than what their results suggest, so look at the actual results e.g. all-cause mortality.

> Most negative effects (especially broad ones like all-cause mortality) only really start showing up under 6

Most - maybe. But one very important negative effect is that waking up without enough sleep is, quite literally, torturous.

nurses and doctors, the 'health' workers, and other erratic schedule-shifting shift workers... RIP

This is well known, and there is a movement to change this for residents, at least. So far nothing has changed.

More proximally: the level of care you can receive / number of errors committed is seriously affected by the loss of sleep by your carer.

We must avoid using causal words like "raises" when a statistical relationship is found. To raise something is to cause something lower to become something higher.

What the study found is that the pattern of sleeping less than six hours is seen more in people who develop dementia.

The study's own discussion states: "There is evidence of a bidirectional association between sleep dysfunction and pathophysiological changes in dementia{citations given}, highlighting the need for a longer time frame in studies."

I really hate these kinds of news stories. It's like: yes, here are some more ways that not getting enough sleep will kill you. Add dementia to weight gain, worsened immune function, etc. etc. Gee, let's all call the Magic Sleep Fairy and ask her to make it possible to sleep longer, as if jobs, family members, noisy neighbors, and all the rest of actual pieces of human existence that make adequate sleep impossible for many of us don't exist.

I think reading the entire article is important here.

My initial response was that the headline writer was sloppy about risk, correlation and causation, but towards the end it notes some studies where one or more nights of poor sleep appeared to cause measurable increases in substances that are present in elevated levels in dementia patients. If those studies did a reasonable job of showing causation, then this could be an extension of that work.

Reverse causation not ruled out. Typically useless study within this field where no real progress has been made in years.

Why assume lack of sleep is the cause? Why not reverse it, that dementia causes people to sleep less?

The study makes it clear that it's an association. It never says it's an assertion of directional causality.

Maybe a bit extreme conclusions? One day or one week of bad sleep detrimental? So people that have more than one kid and had bad sleep for years are all doomed? Like all humanity?

This article makes me question my time management ability. I should tweak it so I can't make the chances worse. I guess I'll sleep right now.

Good to know.

Five years ago I stopped setting an alarm and now I get about 8-9h of sleep every night.

Let's see how this plays out in the long run, haha.

This just stresses me out about sleeping even more and makes sleeping harder than it already is.

Alternative title: people without children have less risk of dementia.

do investment bankers have dementia later in their lifes?

i heard they work 9-5 (9am-5am next day)

Percentage or percentiles?

Typical mainstream click bait.

Unless it's an emergency, I've ALWAYS prioritized sleep. It's just too important for health regardless of these latest findings. I'd call in to work if I didn't get a full nights rest in the past, it's not worth feeling like shit for the rest of the day. The culture of glamorizing working longer and harder till early hours of the mornings on zero sleep, has always been nonsense. Musk is notorious for bragging about this, and the dude is a rough looking 49.

I think Elon Musk looks his age, and looks pretty ok for a 49 year old. Granted he has money, and he's had hair transplants, but I've seen many late 40s early 50s people in much worse shape.

He's approaching the Brimley/Cocoon Line.

Passed the Wendt/Cheers Line two years after Tesla was founded and three years before he became CEO.

He's had a lot of cosmetic...improvements. Plus he can afford personal chefs, trainers, doctors, stylists, etc.

IME, the biggest predictor for people looking younger than their age is exercising and avoiding alcohol. I'm sure sleep is in there since people who do the other 2 normally also sleep ok.

Sleep is in there, trust me. I'm 29 and look 40.

I'd add tobacco use and UV exposure to that list (anecdotal of course).

Constant stress seems the biggest ager. Intermittent stress (even at high levels), punctuated by periods of relexation, doesn't seem as bad.

See: US presidents before-and-after vs corporate CEOs

i think genes are a bigger factor

I disagree. Genes play a role by setting a min/max kind of like guardrails, but there can be a wide gradient between that min/max. Even people with arguably 'great genes' look terrible if they abuse drugs, alcohol, or their body in other ways.

Elon looked 50 when he was 30. It's definitely the money.

Probably the hair transplants.

those are alot of caveats though. excpet for blah blah blah blahb, yeah he looks good.

Compared to his earlier pictures, it looks like he 'filled out' in terms of mody mass and skull growth. That sometimes happens when you take GH for long periods.

Some people are willing to trade health for fortune and fame. There's been interviews with Musk where he describes being unable to turn his brain off. He's probably just leaning hard into the cards that life dealt him - a wildly productive brain that wants to go go go. I don't blame him for taking what he was dealt, and running with it.

He's leaning hard into the stimulants...

People with amphetamine habits typically look very skinny and unhealthy because amphetamines suppress appetite. Same for cocaine, ritalin and all other dopaminergic stimulants. They can also have facial scabs because of obsessive picking/scratching, and tend to be very twitchy and hyper. He doesn't come off that way at all during interviews.

My personal experience leads me to believe that there is no way Elon Musk could have been dosing stimulants for the last 10-20 years and look as good and as relatively calm as he does. You build a tolerance very quickly, and any edge that you thought the stimulants were giving you will fade away fast. Then, if you increase the dose, your life will spiral off the rails as you become more manic, impulsive and irrational (eg: gambling, sex, spending sprees, irritability, violence, paranoia). Just Google John McAfee if you want to see what this can look like.

IMO Elon Musk just has lucky genetics and doesn't need much sleep, always has high energy. There are people who are fortunate to be born like that.

^^ Agreed.

> IMO Elon Musk just has lucky genetics and doesn't need much sleep, always has high energy.

I remember him explaining that 6 hours was just a sweet spot (not the highest energy spot), that he would have more energy with more sleep but it wouldn't be enough to compensate for the lack of hours available in a day. Or something to this effect.

I dunno what his schedule is. Maybe he just seems busier than he actually is.

There's no way you can run all these companies, be an international celebrity, and not be busy.

>My personal experience leads me to believe that there is no way Elon Musk could have been dosing stimulants for the last 10-20 years and look as good and as relatively calm as he does.

Looks can be deceiving :P Years ago Elon was balding hard and now he has a full head of hair.

the man is overweight

Nah he's just bulking. It's all muscle under there.

Your reasoning might hold for an average user but not for a billionaire like Musk. He'll obviously have very smart people research and design a cycle for him that mitigates the usual side effects. With calorie intake monitoring, vitamins and anything you can remotely think of. A guy like Musk will also react more controlled with regard to amphetamines.

Whether he actually uses amphetamines I have no idea. But I'd not be surprised to learn that he has his own RC lab.

I know - quite literally - at least a dozen people who take amphetamines daily and display few or none of the visible descriptions you list above. None of us know what EM ingests.

Sporadic, restricted use of amphetamine tends to retain full effect over the long-term. With regular use, the subtle change in perception it causes tends to persist longer than the physical stimulation. The appetite issue can sometimes be addressed simply by actively choosing to eat more.

I do think that, in the long run, it at best averages out between the unavoidable highs and lows. Some people benefit from being able to time when those periods happen. Probably net loss in many cases. Still, the behaviour you're describing represents the end stages of escalating use and loss of control where they have fallen into dependency and/or addiction. That's not inevitable. Some people do seem to be able to self-regulate their use of strong stimulants.

But you're right. It's clear some people are just innately more driven. In the worst case they're like Elon Musk, always flitting around, perpetually alternating between chipper and enraged, and always with coffee in hand and at most four hours of sleep the night before. I find such people infuriating, but that is probably just envy.

Well, many people don't sleep enough because they cannot. Insomnia is common in midlife.

Most cases of insomnia in midlife are caused by lifestyle problems: lack of exercise, substance abuse, and obesity-induced apnea.

I wonder if Elon Musk has the "short sleep" genes?

https://www.ucsf.edu/news/2019/08/415261/after-10-year-searc...

Heck yes. When I first started in tech, I was all about the working into the wee hours. I wanted to be hardcore. I was listening to a podcast and someone mentioned what you just said, “A major culture problem in the tech industry is glamorizing lack of sleep.” Hearing that gave me the slap of reality I needed. Just because I’m present at the computer doesn’t mean I’m doing my best work. There would be times I was so tired, I’d open an application to do something then sit staring at it for a few moments because I couldn’t remember what I was going to do. Now, I evangelize sleep to all my friends and coworkers. Especially the younger ones. I want to break that cycle.

Melatonin intake (300 mcg) is useful for falling asleep faster.

Reply on news.ycombinator.com