Relation between alcohol consumption in midlife and dementia in late life | Hacker News Reader

Relation between alcohol consumption in midlife and dementia in late life

bmj.com

139 pointsbarking105 comments

>Abstainers were mainly women, had lower education and physical activity, were obese, and had a higher prevalence of cardiometabolic risk factors, all associated with an increased risk of dementia,11,12 which could explain the differences; however, adjustment for confounding factors did not alter the findings.

This is why interactive data visualization is so important. People need to be able to adjust the results of data to fit them, otherwise the margin of error and deviation from the "average" will be so great, findings are virtually useless to an individual.

The moment I read 1-14 units a week that must be wine I thought: “well, that’s every intellectual that I know”.

I haven’t done any research, and all my knowledgeable is anecdotal. But I did work in a dementia ward as a caretaker for 5 years as a student job, and keeping your mind busy throughout life seems like a good idea.

I was thinking that 1 unit = 1 standard drink but if the study is using the UK definition of units, the 1-14/week range is actually much lower than the average alcohol consumption among the academic/intellectual types I know. To put the 1-14 figure into real world terms, 10 UK units is about equal to a single bottle of 14% ABV wine. And to put that in perspective, on multiple occasions I've witnessed a table of Oxford dons drink the equivalent of 10-20 units over lunch!

Granted, academics and intellectuals in the UK are more likely (in my admittedly anecdotal experience) to be functional alcoholics than elsewhere. But still, if I were trying to estimate the median consumption among the few dozen people I know with advanced degrees, I'd say it'd fall more in the range of 2-3 bottles of wine a week or equivalent amounts of beer, which is well into the "excessive" zone in this study. Unfortunately, I seem to fall into that group as well :(

> Granted, academics and intellectuals in the UK are more likely (in my admittedly anecdotal experience) to be functional alcoholics than elsewhere.

Anecdata: Back when I was a barkeeper, I've seen so. much. drinking. with Brits - and ordinary soccer fans at that, not (at least not visibly) academics. They regularly drank all other guests, including Slivovitz-trained Croats and Serbs, under the tables... and were still standing firmly, not stumbling around.

Going into study abroad in Wales, as an American, I had this notion that Europeans were more sophisticated and responsible drinkers.

Naaaah.

I hear you. I thought American university students drank a lot, legally and illegally. Then I went to study abroad in London for a semester, and made a lot of English friends.

Well, now. Compared to my friends in the UK, I was a consumption amateur. Even though for a time it seemed like I was subsisting on lager, I had to work to keep up during pub crawls. And my entire ancestry originates in the UK, so it was all cultural not genetic!

We're more trained, that's all. In Germany you can legally start at 16 with beer and wine, and the hard stuff is available at 18. Both in ordinary supermarkets, gas stations, wherever - the only thing that's forbidden is automated vending machines.

Compared to the US where the age limit is 21 in many places, well that's 5 years of drinking practice aka alcohol tolerance...

Only for the people who actually respect drinking age laws, which afaict is not that many.

It seems to me your comment applies to both sides of the atlantic.

Well, ok. I can only speak for my side of the ocean, and not very authoritatively at that. :) Is it really widespread, though, for people in Europe to start seriously drinking before 16?

I agree, but I don’t think it’s true for the majority in the mid part of life. Almost all my friends are 40-50, have long educations, many are still building on them, and the common consumption is two bottles a wine s week.

One Friday, one Saturday, shared with the wife/husband, which is 10-14 units a week.

Most of the functional alcoholic academic types don’t go that high until later in life, probably because alcohol and having young children don’t really mix too well.

Still completely anecdotal though. It’d be really interesting to see a comparison of this group of people and alcohol consumption. Because other research has already shown that a long education postpones dementia.

Relevant, from The Independent: https://www.independent.co.uk/life-style/health-and-families...

OTOH, the link is weird enough that my guess is that it's hiding other factors... So mapping to your own experience isn't likely to tell you much. (Unless you are a large aggregate of other people.)

Personally, I would prefer that an expert in statistical and scientific methods provides some analysis and conclusions.

There’s a good reason why “data mining” is often described as “data dredging”. While browsing through data, it’s all too easy to find the conclusions that you want to find.

https://en.m.wikipedia.org/wiki/Data_dredging

> however, adjustment for confounding factors did not alter the findings.

Let me guess: adjusting for confounding factors means throwing away most of the data, so that all that remains is a small group of abstainers and non-abstainers who are similar in all regards except for the abstinence.

Problem is, there may be more than one way to do that, with different results; then it's possible to choose a way of reducing the data to fit the desired hypothesis.

The only honest way to fix a confounding bias in the data is to re-do the study with a properly randomized sample of subjects.

> Let me guess: adjusting for confounding factors means throwing away most of the data

The usual method is statistical adjustment for the contributions attributable to other causes, not throwing out data.

Are you seriously suggesting a study in which people are randomly assigned people to drinking conditions for decades? That is perposterous.

Edit: To be clear, assignment to drinking conditions and following them over decades is unrealistic, would have problems with participant compliance, and is probably unethical.

I recently saw a study where adults were assigned to drink 1 drink per day or 0 drinks per day, for years, with the drinks subsidized.

Can anyone tell if this study avoids the "sick quitter" problem? (https://www.vox.com/2018/4/24/17242720/alcohol-health-risks-...)

I don't think it does. Quoting from the study:

"The most intriguing finding from this study was the significantly increased risk of dementia among abstainers, including long term abstainers and participants who became abstainers, and that association was only present in those who abstained from wine.

Abstainers were mainly women, had lower education and physical activity, were obese, and had a higher prevalence of cardiometabolic risk factors, all associated with an increased risk of dementia,11,12 which could explain the differences; however, adjustment for confounding factors did not alter the findings."

It's not clear what the intersection between "long term abstainers" and "became abstainers" is.

See also: “sick non-exerciser”. There’s never enough discussion of correlation vs causation in these types of studies.

Also makes you wonder if some of those wine/none-wine-alcohol effects are really female/male effects.

There are also class and ethnic biases - wine drinkers are disproportionately college-educated and caucasian.

...reading the fulltext now. They were selected from among middle-aged civil servants in London in the late 80s/early 90s so the sample is way more homogeneous than the total population, certainly if the U.S. is what you're thinking of.

In London...

Polluted London, Leaded car fuel London

Is that anecdotal or are you citing something? Boxed wine is a favourite of derelict alcoholics because it's so cheap.

"Abstinence from alcohol in midlife, long term abstinence, and decrease in consumption were associated with a significantly higher risk of dementia of 45%, 67%, and 50%, respectively, compared with consumption of 1-14 units/week, but only among participants who reported abstinence from wine."

I don't get that sentence. Doesn't abstinence from alcohol always imply abstinence from wine? Or is it saying that drinking 1-14 units/wk only lowers your risk compared to being abstinent if some of that consumption comes in the form of wine?

The next paragraph follows up a bit, the phrasing is odd but it helps clear up that the tenuous link is wine/wine components.

>The most intriguing finding from this study was the significantly increased risk of dementia among abstainers, including long term abstainers and participants who became abstainers, and that association was only present in those who abstained from wine.

Doesn't clear up anything. A person who does not abstain from wine is not an alcohol abstainer.

Among alcohol abstainers, there is no special category of those who abstain from wine; logically, they all must.

A: "I don't play sports"

B: "Which ones don't you play in particular?"

A: "Tennis."

B: "Ah too bad; the one I particularly don't play is soccer."

A: "I guess we won't be enjoying our non-play of sports together then. Your avoidance of soccer requires the lack of a big field and complete absence of half a dozen other players so that a team is not accidentally formed; whereas my aversion toward tennis requires that there be no court or racquets."

Flipping "increased" to "decreased" because it's easier to understand, it says wine significantly decreases risk of dementia compared to abstainers, but that other alcohols do not decrease the risk of dementia.

They probably chose the odd phrasing because the baseline is the same as wine-drinkers, so it's not actually a decrease, but an increase in the other groups.

The paper was likely written for a target reader interested in understanding the experiment (and therefore willing to approach it in good faith), rather than minutely parse it to identify grammatical inconsistencies.

Jean-Paul Sartre is sitting at a French cafe, revising his draft of Being and Nothingness. He says to the waitress, "I'd like a cup of coffee, please, with no cream." The waitress replies, "I'm sorry, Monsieur, but we're out of cream. How about with no milk?”

Descartes is sitting across the table. The waitress asks him if he would be having some wine.

He replies. "I think, therefore I am."

I am only guessing here, but they may refer to phrasing of a questionnaire. Remember, responses are not scientific facts that necessarily follow your assumed ontology.

Some people may conversationally use "alcohol" as others might use "hard liquor" and answer differently than others who think "alcohol" includes yeasty breads...

That interaction reads like some pretty good British comedy.

I wonder if the decrease in consumption was really just a side effect of the onsetting dementia that was as yet undetectable?

That’s an interesting idea, could the onset of dementia make someone less sociable and hence drink less.

yeah, that confuses me too.

i was also surprised by the size of the effect. being an abstainer was associated with relatively large increased risks, as you highlight. and later it says "with every 7 unit/week increase, there was a significant 17% increase in dementia risk," which is quite a lot smaller than the effect from abstinence.

> Abstainers were mainly women, had lower education and physical activity, were obese, and had a higher prevalence of cardiometabolic risk factors, all associated with an increased risk of dementia

Well, pretty hard to draw conclusions from this study then.

Zero and heavy alcohol consumption could both be side-effects of isolation. Isolation is associated with dementia as a cause or an effect

My first thought is a relationship with glutamate excitotoxicity. Fronto-temporal dementia is known to be related to glutamate. If that’s lumped in as part of the dementia cohort that may help explain the result.

Probably not a large factor because the prevalence of frontiers-temporal dementia is very small compared to vascular, Alzheimer's, Parkinson's. Anecdotally I see it in less than 1% of dementia patients.

Incidence of FTD is around 2.7-4.1 per 100,000 people.

Incidence for Alzheimer's is estimated to be roughly 10.5 per 1000

So although it's a good thought to have about the correlation, even if 100% of FTD cases were caused by that mechanism I don't think you'd see the difference they saw in the study.

Sources: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3932112/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3405821/

If I remember correctly, the mechanism underlying the glutamate toxicity is mutant SOD, which is made as a response to oxidative stress. The glutamate is simply an excitory stimulus that makes free radicals which need to be mopped up by something. Could be that other proteinopathies are a similar reaction to oxidative stress, and something that promotes GABA, such as alcohol, has a beneficial effect. But of course I’m only speculating.

I'm not knowledgeable enough about the biology to really add much to what you said there either, even though I deal with the end results.

Systems biology is one of those areas where I think AI will make a huge difference in the future. If you are interested in some thoughts on this listen to the first hour and a half talk by Craig Mundie (you can skip first 20 minutes of intro by Collins) of this recent NIH videocast on AI in medicine:

https://videocast.nih.gov/summary.asp?live=28053&bhcp=1

I post it again:

The Truth We Won’t Admit: Drinking Is Healthy

https://psmag.com/social-justice/truth-wont-admit-drinking-h...

Super smart guy by the way.

I am not sure why this gets downvoted. You may look up who this guy actually is: https://en.wikipedia.org/wiki/Stanton_Peele

In light of this piece I'm curious about how countries with large sober populations (majority Muslim countries?) compare to others while accounting for other factors

> countries with large sober populations (majority Muslim countries?)

Closet drinkers. Pay enough money and you can get alcohol everywhere, including booze parties in the midst of Teheran. It has been covered e.g. here: https://www.independent.co.uk/news/world/middle-east/drunk-i...

Reminds me of an Iranian friend I had. We were talking once about weddings in different cultures, and I asked her about Iranian weddings. She laughed and told me that despite the ban on alcohol, every man at an Iranian wedding ended up drunk "somehow".

One more thing that comes to mind: So they're comparing people who drink 14 units/week (one beer a day) versus people who drink more than one beer a day. Are the differences in the effects observed really dominated by the presence of people in that second group who drink a second beer and then stop their drinking for the day, or by the presence in that group of really excessive drinkers.

14 units per week is two drinks a day.

To find the units in a drink you multiply the ABV by the serving size in litres.

One 200 ml glass of wine at 13% ABV would be 2.6 units.

One 330ml bottle of beer at 5% ABV is 1.65 units.

Most people struggle to estimate how many units of alcohol they drink. The "one unit == one drink" thing comes from when units were introduced, when a glass of wine would be 125 ml and it would only be 8%. Those days are long gone.

Maybe we should go back to those days. When I drink wine at home, I haven't measured it, but I'm quite sure I don't drink more than 125 mL at a time as I can easily get more than 6 glasses out of a standard 750mL bottle.

Oh, sorry, just saw explanation below. In the US these studies equate 1 unit = 1 glass of wine or 1 beer, I could swear.

That's what a lot of sources in the US say, but, when they say, "drink", they mean "standard drink", which is this quasi-mythological beverage that one never encounters in real life.

I honestly think that it's negligent on the part of public health folks to continue the practice. Saying "Drinking more than 2 drinks in X time will cause you to be legally impaired" is actively encouraging people to unwittingly drive drunk in a culture when a typical ABV for beer is about twice what it was when those standard were originally calculated, and non-highball cocktails typically contain at least two shots of liquor.

"Unit" is so much better a term.

> when they say, "drink", they mean "standard drink", which is this quasi-mythological beverage that one never encounters in real life.

I encounter 12oz bottles and cans of 5% abv beer literally every day

A 12 oz is .34 L, which at 5% would be 1.7 units of alcohol — rather different than a single unit.

In the US, "units of alcohol" are not a thing or part of the government recommendations.

One standard drink in the US is tied to 12oz of 5% beer (or it's equivalents in wine/spirits), not the UK definition of a "unit" of alcohol.

When there are US health recommendations made, they are referring to standard drinks.

Haha, well in Bavaria what we think of as "a beer" is 500ml of beer at 4%, which is 20ml of pure alcohol, which is "2 units" in the UK.

If we're talking about UK, a pint of a stout is 568 ml at 5%+, which is 3 units.

A single drink generally is more than one "unit". E.g., a (British) pint of beer is 2-3 units.

One drink per day (skipping some days) is 14 units per week.

Two drinks a day is easily some 30-40 units per week.

Is this all related to sleep quality? Increased alcohol consumption decreases sleep quality, decreased sleep quality increases dementia risk, therefore increased alcohol consumption increases dementia risk.

The study is suggesting the opposite relationship - that decreased wine consumption _increases_ the risk of developing dementia

It's suggesting both. Low and high consumption increase the risk. For the high consumers, impaired sleep might be a thing.

Small consumption may with sleep, but large consumption hinders quality sleep.

Who funded this?

I'd like to see the probability changes,not just as percent change but as decimal values. The problem with reporting 17% this and 67% that is that even with a 67% increase in one group it could still be less than the 17% of another. This could help distinguishing things that are less bad vs being positive depending on where it falls in the p-range.

So what corresponds to one unit of alcohol? A glass of wine?

It's a british thing. Drink manufacturers in the UK are obliged to declare the total amount of alcohol that's in the bottle using a "units" system, where half a litre of beer at 4% equates to roughly 2 units. So 14 units per week is a bottle of beer a day.

https://www.drinkaware.co.uk/alcohol-facts/alcoholic-drinks-...

Thank for the link!

https://en.m.wikipedia.org/wiki/Unit_of_alcohol

Amusingly, and unrelated to the topic, I was visiting Scotland, and at one historical site, there was a description of the rations given to each soldier. I further learned that a Scottish pint is 1.7 liters. I could only think to myself: What a nice country!

You choose the unit, just keep it at 14 per week. 2 shelves a day for me.

10ml of alcohol. You multiply the serving size in litres by the ABV number.

200 ml of wine at 12% = about 2.4 units.

330 ml of beer at 5% = about 1.6 units.

10ml of pure ethanol.

Or 57% of an American standard drink.

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

Forfeiting beer and whiskey? No thanks. Total digression, but does anyone here know of any drugs that cause neurogenesis? I speculate shrooms and acid might, but maybe I'm just delusional..

> does anyone here know of any drugs that cause neurogenesis?

Tianeptine is fairly well known[1][2] for being able to prevent and/or reverse changes like hippocampal volume loss, which is commonly seen in untreated depression[3], chronic stress[4] and several other conditions.

7,8-Dihydroxyflavone[5] derivatives and/or prodrugs also look promising[6].

[1] https://www.ncbi.nlm.nih.gov/pubmed/22047727

[2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4034085/

[3] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1557684/

[4] https://www.nature.com/articles/srep29127

[5] https://en.wikipedia.org/wiki/7,8-Dihydroxyflavone

[6] https://pubs.acs.org/doi/abs/10.1021/jm101206p

Not a drug, of course, but don't leave physical exercise off your list.

Jon Hopkins university is conducting research on psychedelic. There is a whole universe of medical applications from what I saw but they need more funding to safely prove those.

I wonder if these differences has something to do with the personality of drinkers? People who are drinking might be more aggressive and might also be phisically active or exercising more. Do we actually have any fundamental reason to believe if moderate alcohol might be okay?

As a non-drinker myself, I think this might have a lot to do with lack of sleep.

“moderation in all things”, as usual, seems like sage advice.

maybe not all things... but most things :)

My favorite phrasing of that is a quote that I think is from Oscar Wilde: "moderation in all things, including moderation."

and it’s even got the virtue of being true.

I'd like to see a study about the relation between Zen meditation and dementia.

TL;DR: Get polyphenols in your diet. Wine is an option for this but there are others.

So basically vegetables, the stuff we already should be eating. Though sure is tempting to think some mysterious chemical in wine is a replacement.

Oops. I don't think the study showed that at all. You have jumped to a conclusion. Maybe it's valid, maybe not.

My grandfather, a wine drinker all his life, suffered severe dementia in his final years. There is no magic bullet.

Anecdotal data is not evidence.

My grandfather smoked 2 packs a day starting when he was 11 till he was 97 and didn't get lung cancer.

It is evidence of N=1.

> consuming large amounts (>21-28 units/week) was associated with an increased risk of dementia compared with light to moderate (7-21 units/week) consumption.4

The idea that 21 units of alcohol per week is moderate and also heavy drinking is.... preposterous.

Consider: drinking alcohol (ethanol) is a Group 1 carcinogen.[1] There is no level of alcohol consumption that is safe.

21 units of alcohol per week should always be considered heavy drinking.

1. https://www.cancer.org/cancer/cancer-causes/general-info/kno...

Just finishing reading "Why We Sleep" and it talks about a lack of sufficient sleep being chronically linked with dementia. Alcohol is a major inhibitor of REM sleep, so if you go to sleep intoxicated, you may be technically asleep but you loose a lot of its benefits. So, I wonder if it's the alcohol -> limits effects of sleep -> induces dementia.

a class 1 carcinogen only means the substance is known to cause cancer. this is not the same thing as saying "there is no level of consumption that is safe".

from your own reference,

"for any particular person, the risk of developing cancer depends on many factors, including how they are exposed to a carcinogen, the length and intensity of the exposure, and the person's genetic makeup."

to be clear, a unit of alcohol is defined as 10ml, or the equivalent of 1/3 of a pint of beer.

so 21 units a week is only equal to a single beer per day. i would not say its "preposterous" to call that moderate.

See DanBC's comment...a unit is based on volume and alcohol content, so 1 unit != 1 drink. Roughly speaking, 7 units is 1 serving of beer, so 21 units is just 3 beers a week, which does appear to be light to moderate alcohol consumption.

Consider: "drinking alcohol (ethanol) is a Group 1 carcinogen.[1] There is no level of alcohol consumption that is safe." Bullshit!

https://psmag.com/social-justice/truth-wont-admit-drinking-h...

Actually, there have been numerous studies that have concluded that moderate alcohol consumption reduced mortality rates. ref: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4803651/

Alcohol is a solvent. It dissolves stuff. Sometimes that is helpful in the workshop where you can see what you are doing. The same can't be said about inside your own head even though our ancestors have been experimenting for millennia.

Anyone done comparison studies with the Muslim world?

Fats dissolve in alcohol so one reason alcohol can be cardio protective is that it literally clears out your arteries. It also raises blood pressure, which effects kidneys and other organs, and can cause other problems possibly stroke. A bigger problem is that alcohol is metabolized into acetaldehyde which if you get too much of causes problems long term.

Water is a solvent as well. In fact, it dissolves more substances than any other liquid on earth. What is your point?

Not only that, but water has the highest pH of any acid!

I just wanted to be down-voted.

I was wondering that too. Mormons would be a good subject group as well.

Reply on news.ycombinator.com