a) people who can afford alcohol have min x threshold of income, income correlated with life outcomes
b) people who drink socially may have more friends, and friends do have relationship with longetivity
c) people who drink may be more open to experience, and more open to experience is correlated with IQ, IQ is correlated to better bodily integrity
d) type of people who drink alcohol are part of genetically endowed groups
d) above and people who drink like this understand the consequences of their life choices better, they have co-linearity with consumption patterns; but it has nothing to do with alcohol itself
I agree with Andrew Huberman's read of the alcohol literature. He did an entire podcast episode on it. There is the link to the podcast and some of the studies. He does mention hormesis as well. But he came to a different conclusion. No amount of alcohol is good for you.
The side effects simply outweigh "fun" factor. This is cultural in nature and most drinkers will not accept it. Just like leftists won't accept the very concept of IQ. Oh well.
Since I quit, I'm biased toward wanting to view it as poison, but I'll accept the evidence if anyone can ever sort it out definitively. I was pretty convinced by the Huberman podcast on alcohol, but I obviously haven't evaluated the supporting research with this level of scrutiny.
It seems like a lot of people love to hear that "it's good in moderation" to justify their drinking habits that go well beyond moderation. It seems as if to most people, "moderate" just not so much that you embarrass yourself. What most people call drinking "socially" is technically binging.
The amount of drinking that presumably might be good for you...about one drink per day max, maybe 2 for men? And I've heard some studies suggest even that reduces brain volume? Any validity to that?
Either way... 4~6 seems like a reasonable moderate amount to me, but I know it can't possibly be healthy long-term. I think I could limit myself to 1-2 if I were really, really determined to, but then I don't even see the point of drinking at all. Unless I'm just taking it as a health supplement or something, which I'm still skeptical it would be. But I guess that's because I come from a long line of alcoholics. Better to risk my health by drinking too little than too much. Also it absolutely interferes with sleep, and it interacts with the ADD meds I'm on.
Alcohol consumption also correlates strongly positively with IQ and well-being when looking at aggregate data (as with US states). iirc, alcohol consumption was the second best predictor of state well-being, losing out only to State FICO scores.
the populations which have the longest history with Alcohol in Europe: Mediterraneans, seem to have biologically selected for moderate consumption and not for teetotalism.
If alcohol had no health benefits at all, why did they select for moderate consumption and not for complete teetotalism? So this could indicate that Med pattern for alcohol consumption is probably the most optimal
What do you think about this study, notwithstanding the ideologically driven source and clearly ideological tone, which reached essentially the same conclusions as you but for concluding "The level of consumption that minimises an individual's risk is 0 g of ethanol per week, largely driven by the fact that the estimated protective effects for ischaemic heart disease and diabetes in women are offset by monotonic associations with cancer."
Does he stretch too far with the cancer risk assessment?
What about the antioxidants in alcohol as the source of the hormesis curve? Wine yes, but whiskey is high in antioxidants and beer has some as well. A prominent American diet doctor claims that beer is the 4th or 5th largest source of antioxidants in the American diet which is an indication of now bad our diet is.
It's hard not to see how selection effects might be doing all the work here. Someone has already mentioned the social aspect, which probably explains a good portion of the variation.
I think the other part of it is pre-existing health conditions. Bear in mind that for most gen X-ers and boomers born in the West, drinking has been the norm. It seems likely to me, then, that a lot of the non-drinkers amongst those generations are not drinking for health reasons.
Consequently, you end up with the people who drink the least dying the most, or at least at an equal rate to heavier drinkers, because they were already sick in the first place.
In my mid-20s my alcohol intake was about 8gm per day and once a month there was 20gm in an evening binge.
Two decades on I'm at about 3gm per day and am never drunk. My tolerance is just much reduced and I can't deal with hangovers anymore.
I think my current intake is objectively healthier but at the same time my peak alcohol period was when I made some life-long friendships, in many cases over sessions of binge drinking! I don't think there are other drugs that are quite as sociable as alcohol, and the long-term benefits of social contact are of course well known.
If you were to posit the existence of a substance that, while circulating in your blood, is very effective at dissolving fat, it would be reasonably to assume that - assuming you were able to get that substance into your blood stream on a regular basis - then doing so would have significant health benefits. And that those benefits might well outweigh any associated health risks.
Given what we know about the history of alcohol intake in humans, and about the chemical effect of alcohol on one of the most significant causes of death in humans, the J curve is the boring, expected outcome. The medical profession's desperate resistance to the tedious reality of alcohol's J curve is far more interesting than the J curve itself.
I'm not entirely sure why I got sent this - I assume I must have signed up for it recently, but I don't remember (too much alcohol, perhaps!). But I'm glad I did, as I found this interesting. Unfortunately, not being a scientist, much of it went over my head, but I found the idea that we can evolve for cultural reasons as well as physical/biological ones very interesting (logical enough, I suppose, but not something I'd really thought about before). We have efectively developed a (slight) use for something that is technically a poison. Is the same true for societies who regularly use things other than alcohol? (Again, logically, yes.) If so, this must have hightened the impact of European settlers giving natives alcohol, and might explain the extreme impact on some indiginous societies.
Anyway, I do have one (minor) comment - as a layperson, I found the measurement of alcohol in grams confusing - surely milliliters makes more sense?
A while ago I was watching a old interview with Robin Hanson on health economics, he mentioned that it seems to be the case that a moderate amount of alcohol was beneficial which is something I didn't think was true, so I spent the next couple days researching the topic. I turns out the J curve seems to be a very robust finding with both very old and very recent papers finding the same result, the standard objection seems to be the "sick quitters hypothesis" but papers that seemed to correct for this such as https://www.nature.com/articles/s41598-022-11427-x still found a J curve. Interesting enough there was actually a RCT planned https://clinicaltrials.gov/ct2/show/NCT03169530 but It was cancelled https://www.nih.gov/news-events/news-releases/nih-end-funding-moderate-alcohol-cardiovascular-health-trial.
My suspicion given the vast literature on the issue and how robust the findings are, is that various health authorities either believe that moderate alcohol consumption is good or at least find it fairly plausible, yet know that if they found strong evidence such as a high quality RCT for such a effect and or started recommending moderate alcohol consumption then the general public would use such advice to rationalize bad behaviour etc. and overall public health would decrease.
I think a part of the dispute is surrounding rationalistic vs empiricist schools of thought, with rationalists claiming various biological mechanisms as to why alcohol is bad, whilst empiricists point to evidence looking at the aggregate effects of alcohol. Interestingly enough from the rationalist types I've spoken to, they seem to find wine to be a plausible exception pointing to Resveratrol, but from what I can tell most of the Resveratrol claims seem mostly fake despite the hype by various hacks such as David Sinclair.
a) people who can afford alcohol have min x threshold of income, income correlated with life outcomes
b) people who drink socially may have more friends, and friends do have relationship with longetivity
c) people who drink may be more open to experience, and more open to experience is correlated with IQ, IQ is correlated to better bodily integrity
d) type of people who drink alcohol are part of genetically endowed groups
d) above and people who drink like this understand the consequences of their life choices better, they have co-linearity with consumption patterns; but it has nothing to do with alcohol itself
I've definitely made the half joke "binge drinking is cardio for your liver" before and it's been surprisingly challenging for people to refute
I agree with Andrew Huberman's read of the alcohol literature. He did an entire podcast episode on it. There is the link to the podcast and some of the studies. He does mention hormesis as well. But he came to a different conclusion. No amount of alcohol is good for you.
https://hubermanlab.com/what-alcohol-does-to-your-body-brain-health/
The side effects simply outweigh "fun" factor. This is cultural in nature and most drinkers will not accept it. Just like leftists won't accept the very concept of IQ. Oh well.
Since I quit, I'm biased toward wanting to view it as poison, but I'll accept the evidence if anyone can ever sort it out definitively. I was pretty convinced by the Huberman podcast on alcohol, but I obviously haven't evaluated the supporting research with this level of scrutiny.
It seems like a lot of people love to hear that "it's good in moderation" to justify their drinking habits that go well beyond moderation. It seems as if to most people, "moderate" just not so much that you embarrass yourself. What most people call drinking "socially" is technically binging.
The amount of drinking that presumably might be good for you...about one drink per day max, maybe 2 for men? And I've heard some studies suggest even that reduces brain volume? Any validity to that?
Either way... 4~6 seems like a reasonable moderate amount to me, but I know it can't possibly be healthy long-term. I think I could limit myself to 1-2 if I were really, really determined to, but then I don't even see the point of drinking at all. Unless I'm just taking it as a health supplement or something, which I'm still skeptical it would be. But I guess that's because I come from a long line of alcoholics. Better to risk my health by drinking too little than too much. Also it absolutely interferes with sleep, and it interacts with the ADD meds I'm on.
Alcohol consumption also correlates strongly positively with IQ and well-being when looking at aggregate data (as with US states). iirc, alcohol consumption was the second best predictor of state well-being, losing out only to State FICO scores.
the populations which have the longest history with Alcohol in Europe: Mediterraneans, seem to have biologically selected for moderate consumption and not for teetotalism.
If alcohol had no health benefits at all, why did they select for moderate consumption and not for complete teetotalism? So this could indicate that Med pattern for alcohol consumption is probably the most optimal
What do you think about this study, notwithstanding the ideologically driven source and clearly ideological tone, which reached essentially the same conclusions as you but for concluding "The level of consumption that minimises an individual's risk is 0 g of ethanol per week, largely driven by the fact that the estimated protective effects for ischaemic heart disease and diabetes in women are offset by monotonic associations with cancer."
Does he stretch too far with the cancer risk assessment?
https://www.thelancet.com/article/S0140-6736(18)31571-X/fulltext?utm_source=baytoday.ca&utm_campaign=baytoday.ca&utm_medium=referral
?
Interesting -- useful info. Btw I looked up what the 20 grams in the first study equals, and it’s roughly one drink (1.33 drinks.)
Emil, why do they use eerie fitting functions of k1*x^a + k2*x^b ? I recall no area where this is used. Why not 4-degree polynomial?
What about the antioxidants in alcohol as the source of the hormesis curve? Wine yes, but whiskey is high in antioxidants and beer has some as well. A prominent American diet doctor claims that beer is the 4th or 5th largest source of antioxidants in the American diet which is an indication of now bad our diet is.
It's hard not to see how selection effects might be doing all the work here. Someone has already mentioned the social aspect, which probably explains a good portion of the variation.
I think the other part of it is pre-existing health conditions. Bear in mind that for most gen X-ers and boomers born in the West, drinking has been the norm. It seems likely to me, then, that a lot of the non-drinkers amongst those generations are not drinking for health reasons.
Consequently, you end up with the people who drink the least dying the most, or at least at an equal rate to heavier drinkers, because they were already sick in the first place.
An rct to study alcohol and cardiovascular health was in fact supposed to happen (mach15) but it was cancelled over fear of conflict of interests,see : https://www1.bio.ku.dk/english/staff/?pure=da%2Fpublications%2Fthe-moderate-alcohol-and-cardiovascular-health-trial-mach15-design-and-methods-for-a-randomized-trial-of-moderate-alcohol-consumption-and-cardiometabolic-risk(27818dc1-77ed-49c1-b382-cfc9e40ec0d3)%2Fexport.html
https://academic.oup.com/eurjpc/article/28/15/e22/6145669
There are also a couple of mandelian study that don't find any cardioprotective effects but they have been contested : https://www.bmj.com/content/349/bmj.g4164
In my mid-20s my alcohol intake was about 8gm per day and once a month there was 20gm in an evening binge.
Two decades on I'm at about 3gm per day and am never drunk. My tolerance is just much reduced and I can't deal with hangovers anymore.
I think my current intake is objectively healthier but at the same time my peak alcohol period was when I made some life-long friendships, in many cases over sessions of binge drinking! I don't think there are other drugs that are quite as sociable as alcohol, and the long-term benefits of social contact are of course well known.
If you were to posit the existence of a substance that, while circulating in your blood, is very effective at dissolving fat, it would be reasonably to assume that - assuming you were able to get that substance into your blood stream on a regular basis - then doing so would have significant health benefits. And that those benefits might well outweigh any associated health risks.
Given what we know about the history of alcohol intake in humans, and about the chemical effect of alcohol on one of the most significant causes of death in humans, the J curve is the boring, expected outcome. The medical profession's desperate resistance to the tedious reality of alcohol's J curve is far more interesting than the J curve itself.
I'm not entirely sure why I got sent this - I assume I must have signed up for it recently, but I don't remember (too much alcohol, perhaps!). But I'm glad I did, as I found this interesting. Unfortunately, not being a scientist, much of it went over my head, but I found the idea that we can evolve for cultural reasons as well as physical/biological ones very interesting (logical enough, I suppose, but not something I'd really thought about before). We have efectively developed a (slight) use for something that is technically a poison. Is the same true for societies who regularly use things other than alcohol? (Again, logically, yes.) If so, this must have hightened the impact of European settlers giving natives alcohol, and might explain the extreme impact on some indiginous societies.
Anyway, I do have one (minor) comment - as a layperson, I found the measurement of alcohol in grams confusing - surely milliliters makes more sense?
A while ago I was watching a old interview with Robin Hanson on health economics, he mentioned that it seems to be the case that a moderate amount of alcohol was beneficial which is something I didn't think was true, so I spent the next couple days researching the topic. I turns out the J curve seems to be a very robust finding with both very old and very recent papers finding the same result, the standard objection seems to be the "sick quitters hypothesis" but papers that seemed to correct for this such as https://www.nature.com/articles/s41598-022-11427-x still found a J curve. Interesting enough there was actually a RCT planned https://clinicaltrials.gov/ct2/show/NCT03169530 but It was cancelled https://www.nih.gov/news-events/news-releases/nih-end-funding-moderate-alcohol-cardiovascular-health-trial.
My suspicion given the vast literature on the issue and how robust the findings are, is that various health authorities either believe that moderate alcohol consumption is good or at least find it fairly plausible, yet know that if they found strong evidence such as a high quality RCT for such a effect and or started recommending moderate alcohol consumption then the general public would use such advice to rationalize bad behaviour etc. and overall public health would decrease.
I think a part of the dispute is surrounding rationalistic vs empiricist schools of thought, with rationalists claiming various biological mechanisms as to why alcohol is bad, whilst empiricists point to evidence looking at the aggregate effects of alcohol. Interestingly enough from the rationalist types I've spoken to, they seem to find wine to be a plausible exception pointing to Resveratrol, but from what I can tell most of the Resveratrol claims seem mostly fake despite the hype by various hacks such as David Sinclair.