Sleep and Mortality: A Population-Based 22-Year Follow-Up Study (2007) | Hacker News Reader

Sleep and Mortality: A Population-Based 22-Year Follow-Up Study (2007)

fermatslibrary.com

138 pointsonderkalaci37 comments

The people that want to make criticisms about how this association is not causal can start reading halfway through the first column of page 1252. The authors are careful to phrase their finding as an association, not a causal relationship.

They discuss the confounds you might think of (sleep a lot because of undetected major illness? depression? etc) and their efforts at controlling for them.

They then acknowledge that there are still enough unknowns that they can't make a causal link, and that there should be more in depth and controlled experiments to investigate this phenomenon.

I agree. I think publication of studies/experiments that don't find anything particularly interesting should be encouraged not condemned.

It's much more of a problem that there's a bias to publishing to a headline. I'm, slightly irrationally, more inclined to believe this than a lot of papers.

Agreed. What can we really say is causal? Not that much in the grand scheme of things.

I suppose one could argue that in an absolute philosophical sense there is no causation that we can have knowledge of, only differing amounts of correlation. It would be an extreme position to take but - like solipsism - a difficult one to argue against.

(Anyone with an actual background in epistemology care to help me out here?)

Hume argues this in “enquiry concerning human understanding” https://plato.stanford.edu/entries/hume/#Cau

I always considered causality less as a binary thing that was always 0 until $somethreshold was crossed, and more of a bayesian (or other) probability model.

Since in many cases we don't have the evidence for a deductive conclusion, we must often rely on inductive logic, even while still seeking the evidence until we can switch to deductive.

https://en.wikipedia.org/wiki/Granger_causality

Slightly tangential to the study here but I’ve been going down the sleep apnea rabbit hole in recent months and I strongly encourage folks to investigate their sleep. A lot of people have sleep apnea and don’t realize it. You don’t have to snore to have it. A surprising statistic I found was 20-30% of people with ADHD have sleep apnea. A lot of people may be treating symptoms of sleep apnea like ADHD and high blood pressure with medications while ignoring the root cause. My experience with this has been that doctors are surprisingly ignorant. They’ll happy prescribe you medications for anxiety, ADHD, blood pressure for years, and never think to ask you about your sleep. Do some of your own investigations or ask about it if you have any suspicions.

I'm going to add myself as a data point. After suffering years of feeling sleepy regardless of how much I slept and frequently unmotivated and withdrawn, I was diagnosed with severe sleep apnea. A few years ago I was given Dexedrine, Ritalin, Adderall, and I even tried self-medicating with modafinil to ameliorate the symptoms of what doctors thought was ADHD with varying levels of success; but the drug-free treatment of sleep apnea with a BiPAP machine got rid of all of these issues; in addition, I feel about ten years younger.

Do you have any suggestions on how can I not too expensively measure my sleep quality at home?

If you're lucky enough to have decent medical insurance I'd simply bring it up to your primary care doctor for a referral or sleep doctor in your area (you usually don't need a referral if you want to seek out a sleep doctor directly).

They will have you come in for a visit to talk about your symptoms and evaluate you, then usually they'll schedule you for an in-home test, which means you get a little device to take home and strap to your chest. It'll have a pulse-oximeter you connect to your finger, and an air flow meter that you run to your nose using a nasal canula.

If this isn't conclusive or something else seems to be going on they will bring you into the lab to sleep with even more devices connected to you measuring brain waves, muscle activity, etc. These usually look like a hotel room so it's not too uncomfortable.

I wanted to get a head start on my own so I set up a camera that has night vision to record myself since I happened to have one. (Lots of home security type cameras will have a night vision mode.) It wasn't particularly conclusive, I couldn't really make out that my breathing was stopping, but it did make me realize I seem to wake up and move a lot. Additionally I was surprised to learn I sleep with my mouth closed, since I've always gone to sleep with it open.

There are options I've seen online from various online CPAP stores to order an in-home test as described above, I don't know how reliable they are, presumably they mail you the kit and you mail it back.

> Additionally I was surprised to learn I sleep with my mouth closed, since I've always gone to sleep with it open.

Do you grind your teeth or clench your jaw when you sleep? If you grind your teeth your dentist will be able to tell. If you clench your jaw you will wake up with muscle soreness.

When I was a trader I had a lot of stress and would slam my jaw shut with enough speed to wake myself up, and in one case, break a dental appliance.

I'm not sure if you're implying there is a connection between bruxism and sleep apnea, but after a quick search, it looks like there is[0]. I've dealt with some of the issues mentioned above -- in addition to bruxism -- and never even thought to make the connection. So, whether you meant to or not, you may have just helped my tremendously. Thanks!

- 0: https://sleepfoundation.org/sleep-topics/the-link-between-sl...

Very happy to hear I've helped you!

I have also dealt with both bruxism and sleep apnea so I know the symptoms. I also thought the same thing about going to sleep with my mouth open. I figured we had something in common.

Eventually I had surgery to fix my sleep apnea. Saying the surgery was a life changing experience is an understatement. Getting proper sleep for the first time after being tired my entire life was a giant step forward for me as a human being.

If you want to confirm a severe case of apnea, it's extremely simple: record yourself during the night. Then listen to the recording and look for moments when you stop breathing completely.

Milder cases of apnea require a more thorough counting of apnea events to confirm; that ought to be done in a proper sleep lab.

Thank you, really for your insight.

I’m also diagnosed with ADHD and I currently take everything from Ritalin, to Strattera. And self medicating with modafinil.

I’m going to look into this, again, thank you for the insight.

Results: long (>8h) and short (<7h) sleep both associated with greater risk of mortality.

It is curious when you phrase it this way. Sleep that is not aligned with the majority of the population associated with death rates not aligned with the majority. ;)

As the other poster said, this is not a surprise to the authors, so I don't need this as a criticism of it. Just like the amusing rephrasing.

Huh, interesting indeed. I didn't think of it that way.

I go to sleep at 00:00 and wake up at 07:30 so that makes me rather happy :)

I read somewhere your body likes natural 1.5h increments. I've found if I hit those numbers my days and nights go much better. I always slept 9h/night until recently. Now I mix in the occasional 7.5 hour night to no detriment. Even do 4.5 and 6.0 when necessary (early flights/etc).

...and, if I understood their tables correctly, use of tranquilizers (benzodiazepines, etc.) is associated with significantly greater risk.

doesn't that have a lot of confounding factors though? e.g. such difficulties getting to sleep medication is needed, or willingness to medicate (rather than change behavior/environment) etc.

Does this take it into account of the possibility that people who sleep longer than 8hrs probably has other unhealthy habits, like lack of exercise etc..

Your question can be answered by reading the other comments here.

To everyone requesting more, different studies: Okay, you're correct in the philosophy-of-science sense. No doubt that feels good.

But, current understanding of biology is that:

- Literally no biological process is not affected by sleep

- Disrupting sleep disrupts whichever process is being studied

- Literally every type of organism sleeps[0]

It's a huge waste of time further proving that sleep is really, really, important, and this waste of time is actively harmful to everyone. What's maybe not a waste of time is researching

- How to actually get things into a shape where sleep is a priority on a population level

- Therapeutic interventions targeting poor sleep

- Otherwise preventing or mitigating poor sleep

[0] Sleep:

- Periodic

- Quiescent (less activity, less response to environmental stimuli)

- Homeostatic, e.g. sleep less now -> sleep more later

I think studies like this are important to determine the importance of sleep and the right amount. For example, this study suggests that too much sleep is harmful, which isn't a universally agreed-on point.

Additionally, sleeping isn't a binary thing. People do mini cost-benefit analyses: sleep an extra hour vs do more work? Sleep vs talk to a friend? Sleep more or spend time eating a slow, healthy breakfast?

Maybe the health benefits of sitting down and eating a healthy breakfast outweigh the benefits of sleeping by a large margin, although both are helpful. More research could help people make decisions about how much they sleep.

It's not a waste of time at all. The "why" may become apparent from these studies.

What worries me more is the link between sleep and Alzehimer’s/cognitive impairment.

I would prefer to see an experiment (even on non human mammals) rather than see a study that likely cannot disentangle correlation and causation.

I'm not sure an experiment with non-human mammals would give you much either. How would you control how much sleep the subject gets? Only thing I can think of is sedatives to put them to sleep, and stimulants or flashing lights/loud noises to keep them awake. And those all come with fairly significant side effects, to the point that I imagine you'd really just be measuring the effects of those medications on the subjects rather than the sleep. But this isn't my area, and this is based on like 10 minutes of thought, so maybe there are more reliable ways to control sleep with fewer side effects?

I would love to see the results of a slightly different study where participants (usually) fell asleep and woke naturally (and without external influences like apartment neighbors or flatmates) and those who's sleep cycle terminated artificially.

There's been a lot of studies on such sleep cycles (esp: without an external day/night cycle). But I can't really imagine a study that removed people from external influences on sleep (including their own children) for 20 years... Without having serious issues with separating out other factors...

This is a repost from yesterday: https://news.ycombinator.com/item?id=16267306

"Signiﬁcantly increased risk of mortality" sounds like the participants were immortal before..

(2007)

Thanks! Updated.

Reply on news.ycombinator.com