5 ms·

Before we admire Thatcher's sleep minimalism too much, let's not forget the onset of dementia in her 70s (http://news.bbc.co.uk/2/hi/uk_news/politics/7579352.st

Before we admire Thatcher's sleep minimalism too much, let's not forget the onset of dementia in her 70s (http://news.bbc.co.uk/2/hi/uk_news/politics/7579352.stm http://news.bbc.co.uk/2/hi/uk_news/politics/7579352.stm).

brador 14y ago
Is there any research on correlation of dementia with lack of sleep? Would be an interesting angle.
gwern 14y ago
Dementia in one's 70s... Surely that hardly ever happens to normal women who sleep 8 hours a night...
- woodchuck64 14y ago
  I'd say roughly the normal rate for dementia is about 6 in 100 at around age 70. Clearly there are some unusual things going on with those unlucky 6 and you don't think sleep disorder is a possible risk?
  - gwern 14y ago
    > Clearly there are some unusual things going on with those unlucky 6 and you don't think sleep disorder is a possible risk? As far as I know, the underlying causes of dementia are not very well understood. Besides the fact that 6% is not very rare, there's an issue of the Texas sharpshooter fallacy (http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy): taking whatever problem exists and claiming that may be linked to whatever one disapprove of. "She has Parkinson's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has Huntington's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has Alzheimer's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has MLS, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has brain cancer, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She had a crippling stroke, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." You see my point. There are an awful lot of things that can go wrong with a brain over time, the rate at which things go wrong goes up with time (we call it "aging"), and sooner or later something will go wrong (and you will "die"). It is inevitable that some short sleepers will suffer awful brain problems. Just like everyone else. I saw another example yesterday in a discussion of famous drug chemist & inventor Shulgin; apparently, upon getting cancer at age 80 and asking for help, his friends had to deny to people that it was related to any of the drugs he took! I was like, "are you serious? A large minority or a majority of the entire population will have cancer at some point in their lifetime, many will not even make it to their 80s at all much less only develop cancer at that advanced age, and people seriously think that this might be more than-microscopically-small evidence that some of the drugs cause cancer, when if anything it may be evidence against?"
    - woodchuck64 14y ago
      Points well taken. While my comment falls short of claiming that dementia is linked to lack of sleep, I was primed by the following research in the media recently: "Sleep duration and all-cause mortality": http://www.ncbi.nlm.nih.gov/pubmed/20469800 http://www.ncbi.nlm.nih.gov/pubmed/20469800 "Disturbances to the sleep-wake cycle predict future diagnoses of Alzheimer’s disease. ": http://www.abstractsonline.com/Plan/ViewAbstract.aspx?sKey=fd313178-35c6-4098-9231-19b31db40132&cKey=6c052583-7bb1-46d4-82db-cc98608e4e30&mKey={70007181-01C9-4DE9-A0A2-EEBFA14CD9F1} http://www.abstractsonline.com/Plan/ViewAbstract.aspx?sKey=f... The first, a survey, seems to show that too little sleep increases all-cause mortality (i.e. that "aging" you mention), the second suggests that sleep disturbances (unknown, admittedly how that relates to sleep duration) may be linked (cause or effect unknown) with Alzheimer's.
      - gwern 14y ago
        Eh. Correlations are notoriously weak, and the causation could just as plausibly be reversed (sleep is, after all, a brain activity); this may just be picking up conditions causing sleep abnormalities, which would explain in your first link why they also found an association with too much sleep as well ("Long duration of sleep was also associated with a greater risk of death").
        woodchuck64 14y ago
        A 2009 article that hints the causation could be due to sleep, at least in mice: http://www.sciencemag.org/content/326/5955/1005 http://www.sciencemag.org/content/326/5955/1005 (see also http://en.wikipedia.org/wiki/Beta_amyloid#Circadian_rhythm_of_amyloid_beta http://en.wikipedia.org/wiki/Beta_amyloid#Circadian_rhythm_o...).

brador 14y ago

Is there any research on correlation of dementia with lack of sleep? Would be an interesting angle.

gwern 14y ago

Dementia in one's 70s... Surely that hardly ever happens to normal women who sleep 8 hours a night...

woodchuck64 14y ago
I'd say roughly the normal rate for dementia is about 6 in 100 at around age 70. Clearly there are some unusual things going on with those unlucky 6 and you don't think sleep disorder is a possible risk?
- gwern 14y ago
  > Clearly there are some unusual things going on with those unlucky 6 and you don't think sleep disorder is a possible risk? As far as I know, the underlying causes of dementia are not very well understood. Besides the fact that 6% is not very rare, there's an issue of the Texas sharpshooter fallacy (http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy): taking whatever problem exists and claiming that may be linked to whatever one disapprove of. "She has Parkinson's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has Huntington's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has Alzheimer's, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has MLS, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She has brain cancer, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." "She had a crippling stroke, which X in 100 have by age 70. Clearly there are some unusual things going on with those unlucky X..." You see my point. There are an awful lot of things that can go wrong with a brain over time, the rate at which things go wrong goes up with time (we call it "aging"), and sooner or later something will go wrong (and you will "die"). It is inevitable that some short sleepers will suffer awful brain problems. Just like everyone else. I saw another example yesterday in a discussion of famous drug chemist & inventor Shulgin; apparently, upon getting cancer at age 80 and asking for help, his friends had to deny to people that it was related to any of the drugs he took! I was like, "are you serious? A large minority or a majority of the entire population will have cancer at some point in their lifetime, many will not even make it to their 80s at all much less only develop cancer at that advanced age, and people seriously think that this might be more than-microscopically-small evidence that some of the drugs cause cancer, when if anything it may be evidence against?"
  - woodchuck64 14y ago
    Points well taken. While my comment falls short of claiming that dementia is linked to lack of sleep, I was primed by the following research in the media recently: "Sleep duration and all-cause mortality": http://www.ncbi.nlm.nih.gov/pubmed/20469800 http://www.ncbi.nlm.nih.gov/pubmed/20469800 "Disturbances to the sleep-wake cycle predict future diagnoses of Alzheimer’s disease. ": http://www.abstractsonline.com/Plan/ViewAbstract.aspx?sKey=fd313178-35c6-4098-9231-19b31db40132&cKey=6c052583-7bb1-46d4-82db-cc98608e4e30&mKey={70007181-01C9-4DE9-A0A2-EEBFA14CD9F1} http://www.abstractsonline.com/Plan/ViewAbstract.aspx?sKey=f... The first, a survey, seems to show that too little sleep increases all-cause mortality (i.e. that "aging" you mention), the second suggests that sleep disturbances (unknown, admittedly how that relates to sleep duration) may be linked (cause or effect unknown) with Alzheimer's.
    - gwern 14y ago
      Eh. Correlations are notoriously weak, and the causation could just as plausibly be reversed (sleep is, after all, a brain activity); this may just be picking up conditions causing sleep abnormalities, which would explain in your first link why they also found an association with too much sleep as well ("Long duration of sleep was also associated with a greater risk of death").
      - woodchuck64 14y ago
        A 2009 article that hints the causation could be due to sleep, at least in mice: http://www.sciencemag.org/content/326/5955/1005 http://www.sciencemag.org/content/326/5955/1005 (see also http://en.wikipedia.org/wiki/Beta_amyloid#Circadian_rhythm_of_amyloid_beta http://en.wikipedia.org/wiki/Beta_amyloid#Circadian_rhythm_o...).