Hacker News Reader

Comment by Mathnerd314 | Hacker News Reader

The overall conclusion of the study is "We observe negative relationships between alcohol intake and global gray and white matter measures". But if you look at Figure 3, it is consistent with a J-shaped risk curve with risk equal to nondrinkers around 16 grams ethanol per day, like the 16.9 number found for 40-64 year old males in https://www.thelancet.com/journals/lancet/article/PIIS0140-6.... So I would say this study adds another reason to not overdrink, and confirms that current drinking expectations are too high, but isn't a reason to abstain entirely.

View on HN

Did they correct for individuals who currently intake no alcohol but have a past history of alcohol consumption? My understanding is this is one of the ways past studies have skewed the results, because some people quit cold turkey but then live with the consequences of moderate or heavy drinking for years (or an entire lifetime) after that.

(I skimmed to the introductory paragraph where they talk about confounders, but don't see alcohol history listed as a confounder.)

Some version of their code is listed here[1] and it doesn't look like former alcohol consumption was accounted for. (Interestingly they have code to toggle inclusion/exclusion of people based on former alcohol consumption.) I didn't see it mentioned in the main text or the supplement either.

1 = https://osf.io/chgbv

In the summary section they talk briefly about this:

"It is reasonable to expect that the relationship we observe would differ in younger individuals who have not experienced the chronic effects of alcohol on the brain. An additional limitation stems from the self-reported alcohol intake measures in the UK Biobank, which cover only the year prior to participation. Such estimates may not adequately reflect drinking prior to the past year and are susceptible to reporting and recall bias.

Further, our analyses do not account for individuals with a past diagnosis of AUD. Earlier studies have shown that the brain shows some recovery with the cessation of drinking in individuals with AUD, but this varies with age and sex, and recovery might be incomplete. Thus, a past diagnosis of AUD would likely influence our results. We hope future studies will shed light on how a history of AUD with prolonged recovery is associated with brain structure in middle-aged and older adults."

>> Alcohol intake explains 1% of the variance in global GMV and 0.3% of the variance in global WMV across individuals beyond all other control variables (both p < 10−16).

Definitely shows there is some grey and white matter reduction... but the impact doesn't seem drastic? Do I use that 1%? Do other activities reduce my brain of 1%? I don't know the scale of volume they use. I could be wrong.

And looking at figure 2[1], age is also a major factor for change in white/grey volume (which they normalize).

[1] https://www.nature.com/articles/s41467-022-28735-5/figures/2

You have to admit the same argument in other contexts is a little funny. Like if you were exposed to an industrial pollutant that was shown to reduce your brain mass by 1% I don't think many people would be saying 'do I really need that brain material?' Obviously this is more than a little influenced by enjoying the effects of a drug.

I mean a pollutant has no upside. Being intoxicated has a lot of upsides including a good time, stress relief, community etc.

The pollutant isn't being dumped into the environment for fun. It's almost always done in service of some sort of economic activity (eg. farming, power generation, resource extraction). In this context the upside is the associated economic activity.

Taking alcohol is my choice. I make that choice because I think the trade off is worth it, but I respect other people who think otherwise and do not drink. It is nobody’s choice to live in a polluted atmosphere and you cannot opt out.

That argument assumes high efficiency in society when we both know that society is anything but.

Well, on one hand, I drink alcohol on my own choice, it’s not much my choice if some pollutant is dumped into the air.

> it’s not much my choice if some pollutant is dumped into the air

Aren't you enjoying the benefits of those pollutants ? The sheer fact that you are posting on HN means that you have access to a device that was produced thousands of miles away and got delivered to you. No one forced you to own one, that was your choice.

Just because you are not the one physically emitting the stuff does not mean you share no responsibility for it.

The reception your comment got says a lot about the HN community's ability to have adult discussion about subjects involving bodily harm.

The supply chain for our basic necessities of life maims and kills people but we consider putting food on plates to be well worth the tradeoffs. Having a blood diamond on your ring is not something we consider to be worth the tradeoffs. The tradeoffs made to get a smartphone or computer into a consumer's hand, along with the tradeoffs of consuming booze, exist somewhere in the nuanced continuum between those fairly extreme example points.

You are assuming that there is an explicit decision made wrt trade-offs.

A simpler explanation is - a "blood" diamond has a fairly explicit consequence, manufacturing consequences aren't so obvious or necessary.

I'd also note that it should be possible to police these things - diamonds without blood, tech without pollution - consumer boycotts aren't often so successful vs legal policy changes.

You certainly see arguments like “lowering the speed limit by X% would reduce road fatalities by 1%” and people saying “no thanks, I’ll take the risk”.

Pulling in something we all agree is bad feels like you are trying to trick us into agreeing with you.

What about exposure to television programming? Does that have an impact on brain mass? Soft drinks? Carbohydrates? Music?

Other example: more than 3.5 million people die each year from outdoor air pollution. Yet nobody seems to be alarmed by it to the point that anything changes.

It's only when there is a direct short-term life-threat that people seem bothered to take action (see COVID).

People genuinely don’t really care about air pollution in neighborhood’s broadly, actually. I mean, consider the fact brake pads can still be made with asbestos and are aerosolized every time a car brakes. But it genuinely hasn’t stopped cars.

> But it genuinely hasn’t stopped cars.

I disagree; in fact, brake pads have stopped cars. ;)

And my crazy neighbors insist on burning leaves instead of just mulching/composting. It's atrocious outside, and its all so pointless.

That says more about the US than it does 'people' - quite a few countries banned asbestos brakes decades ago (UK 1999, Aus 2003, Germany 1993).

Driving 1% of the variance is not the same as reducing your brain mass by 1%. Without knowing anything about the brain, Reducing your brain mass by 1% sounds pretty crazy to me, and probably has extreme negative effects very far before hitting 100% loss.

It’s not absurd to say the negative effects of air pollution on this specific outcome are mostly negligible but still not want air pollution for other reasons. Which is a hypothetical statement of course because we don’t have any data here.

Outdoor air pollution doesn't taste good and it's not my choice. There's a huge difference between "This is slightly dangerous" and "This is slightly dangerous but a lot of fun"

Well, a pollutant has no use. That’s the difference.

I’d like to see the same comparison between people living in cities or countryside.

>Well, a pollutant has no use. That’s the difference.

Every piece of technology that carried your commentary from your fingertips to my eyeballs has a nonzero environmental impact. Some watershed in Asia had to be fractionally (after all, the mine likely supplied material for far more devices than yours and mine) poisoned so you could say that.

Is that sufficient purpose?

Keep in mind modern society exists at the end of a very, very, very long and interconnected supply chain so the raw material from the mine probably (fractionally, of course) finds its way into smoke alarms and pacemakers. Trying to define "purpose" is an unbelievably complex can of worms.

The old mantra of people only use X% of their brain has been found to be hogwash. We use all of our brain, so I’d be very worried about a poison that even takes 1%.

> The notion that a person uses only 10 percent of their brain is a myth. fMRI scans show that even simple activities require almost all of the brain to be active.

https://www.medicalnewstoday.com/articles/321060#takeaway

But the parent isn't asking whether you use 10% (or even 99%) of your brain. He's asking what type of impact 1% reduction in brain mass results in.

What would happen if you disabled 1% of your cpu? People might say what’s a loss of 1% but I’m reminded of data from sports science where 1% more performance is an absolutely astounding amount that separates world class from the pack. I don’t want to be in the mind pack!

Treating patients with morbid obesity led to a 2% increase in total brain matter (https://www.sciencedirect.com/science/article/pii/S1871403X2...), and depression causes differences around 2% in brain areas (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6306287/#!po=29...). The study says one additional drink/day is a change of 0.034 SD or around 0.3% change. So around 6 drinks/day is enough to cause changes similar to these.

I'm considered by many folk to be a fairly heavy drinker, but 6 drinks a day, every day, is seriously indicative of what I'd call a real fucking drinking problem that one should get professional help for.

I didn't see https://www.nature.com/articles/s41467-022-28735-5/tables/2. It looks like going from 0 to 3 drinks would be a 2% decrease in brain matter.

I'm trying to understand. The impact reported by the study seems to compound with higher amounts (I would imagine as liver and kidney throughput is exceeded and the poison hits you more directly).

      Standard Control:

         GMV     WMV

0–1 AU −0.030 −0.020

1–2 AU −0.127 −0.074

2–3 AU −0.223 −0.129

So cumulatively, 0-3 drinks is is GMV: -0.380, WMV: -0.223

What I don't know and can't seem to find with a quick search is the conversion method for IDP standard deviations in brain matter relate to absolute percentages of brain matter (which I assume is not linear?). If your comment is correct, then these numbers correlate to an absolute reduction of 2% of brain matter?

I also see that the third drink seems to do about 50% more standard deviation movement than the former 2 put together.

EDIT: Never mind it's been a while since I took statistics, I remembered that standard deviations _are_ a linear amount. I'd been thinking of percentile placement on a normal distribution.

I tried to go a little further since the depression study seemed to be focused on total brain matter. I might make a mistake here, but...

Grey matter is about 40% of the brain, white matter is 60%, so if we're talking about absolute matter, we should factor that in.

For 0 to 3 drinks:

GMV: -0.380 x 0.4 = -0.152 SD WMV: -0.223 x 0.6 = -0.1338 SD

Should be sum of 0.2858 standard deviations of total brain matter, or about 2.5% reduction.

For 0 to 2 drinks:

GMV: -0.157 x 0.4 = -0.0628 SD WMV: -0.094 x 0.6 = -0.0564 SD

0.1192 SD or about 1.05% reduction

In my experience of knowing alcoholics, I'd say 6 a day really isn't that bad; it's the bottle of liquor a day types that I find impressive. Also, if they were the kind of people who could get access to professional help, they probably wouldn't drink so much to begin with.

There is a such thing as a high-functioning alcoholic.

High-functioning might contradict calling it a problem? From the stories I read about Christopher Hitchens, he seemed like the definition of an unrepentant alcoholic and seemed to function quite well, but of course he died a bit young. I don't think many people are Christopher Hitchens though.

A friend of mine is a very talented and successful salesman. If he doesn't get some alcohol by 7 PM, his skin starts to crawl and he can barely keep it together. What do you call that?

The disease isn't defined by any outcome other than alcohol consumption and dependence.

I don't know. Many of us live with disease of one sort or another; I was just saying that in my personal experience, while a 6-pack a day might seem like a lot to some, it's really not the depraved, want-to-die-alcoholic consumption level.

I knew a functioning alcoholic too, until he died. At 35.

From what?

He killed himself.

To be exact: his was high functionning until he wasn't. From my understanding, he had a bad week (health troubles), which spiraled into a worse one at work. He took a month off, but was so nervous and on the edge that he beat his kid, and his ex-wife told him to put himself together or she would ask for full custody (basically you only see your child two weekend a week instead of half/half).

Don't know what happened then (i had taken my distance since the divorce). The day before he had to go back to work (or the day after, not sure), he killed himself.

[edit] This is my flawed understanding of the events. It probably did not happen like this exactly, most of it was discussions at the church and after the incineration (we usually have dinner with friends and family during those times where i live).

I'm sorry to hear that. It seems to be a theme in this thread that it's hard to make a causal association with alcohol, and your friend would seem to illustrate this. I would think death via alcohol would mean alcohol poisoning or cirrhosis (apparently this took a friend of a friend of mine last month who worked as a bartender at a popular venue in town) or other health condition more easily related to alcohol abuse.

Mostly I think unbinding basic health care in the US from employment would help with many issues. I don't know how people are supposed to be able to heal from illness while also putting in their efforts to a full time job.

You can be alcohol dependant without it causing you a great deal of professional/personal issues. Your healths fucked though.

Same goes for other drugs - I know a few people who are dependant on opiates or cocaine, but function just fine in society.

6 cans/bottles/pints a day isn't terrible in the grand scheme of things, but its not good either, and it almost always escalates over time.

You can sustain a pretty serious drinking problem that will cause long term harm to your health while also remaining a functional member of society, even getting up to the bottle of scotch a day or more tier.

A fair amount of my professional peers are basically highly functional alcoholics, who have not yet reached a point where it causes sufficient impact on them to recognise the problem.

Most of those people may well have access to professional help, they just don't use it because they don't see a problem. Yet.

This is in a different measurement from a casual understanding of 6 drinks. It’s 6 daily alcohol units. UK’s NHS alcohol calculator says 2 pints of 5.2% beer is 6 units.

> UK’s NHS alcohol calculator says 2 pints of 5.2% beer is 6 units.

It's actually quite easy to calculate fwiw, a 'unit' is 10ml ethanol:

    2pints * 0.052 / 10
    = 2(568ml) * 0.052 / 10
    = 5.9072 units

... I'm not sure if that made my 'easy' point well enough, obviously pints mess it up a bit (and you have to know a pint is 568ml - I've remembered ever since I went to study at Imperial, where the student bars include the well-named 'Metric' & '568') but 10ml is nice and round so it's fairly easy to get a rough idea or with rounder numbers - e.g. a whole 75cl bottle of wine at 12% is 3/4 * 12 = 9 units.

> 6 drinks a day, every day,

One glass of wine, 125 ml, at 8%, is one UK unit. Half a .75l bottle of wine at 14% would be 5.25 units.

It's not unusual in the UK for people to drink half a bottle of wine each evening, and then more on the weekends.

We drink a lot. It causes a lot of harm.

Did they control for all of the comorbidities associated with non-drinkers?

A lot of people who don't drink are not abstaining for some health augmentation or lifestyle choice, but because drinking any amount of alcohol would send them into a very early grave.

It seems like these hidden confounding factors are more plausible of an explanation that the "0 drinks per day" group has higher morbidity, compared to "alcohol is healthy in moderate amounts" explanation for the bottom of the J curve.

Which group of society would go to a very early grave by drinking ANY amount of alcohol?

Very rough estimates say there are ~6 million Americans with Chron's or Ulcerative Colitis, none of whom have ANY business drinking ANY amount of alcohol.

That's just one group. There are other groups, I'm sure, but that's the one I'm painfully familiar with.

And what happens if they drink a beer or have a glass of wine?

I guess it’s cognitive dissonance at play. If you try hard to find a way to interpret the results to justify your drinking habits, you will find it.

From figure 3, it looks like half a drink a day improves your brain volume size versus abstaining entirely.

For reference, a single shot (1.5 fl oz of 40% alcohol) has 14g of alcohol.

Yeah. And it's a daily average, so you can have 3 standard drinks every Saturday for example (aiming for the TMREL of 6g). I'm not sure about the maximum per day; there was a heart disease study that defined heavy drinking as 60g in a day,but I haven't seen any studies specifically on drinking patterns at this low intake.

Ergo, 7 shots once a week and I'm good to go!

A pint of beer is 20g.

I used to drink socially when I was young, and I usually had more than that.

Of what abv%? 4.1% light or 8% dipa.

To contain 20g of ethanol (25ml of ethanol) a pint (473ml) would need to be 5.3% ABV, squarely in mid-Pilsner ABV range.

Reply on news.ycombinator.com