Hacker News Reader

Comment by FabHK | Hacker News Reader

A) Is the following then the appropriate conclusion? Drink moderately, and find a spouse that drinks approximately as much as you.

B) I wonder whether the following set of hypotheses makes sense, and how one could test it:

- Social life is good for health

- Alcohol is bad for health, and the more, the worse (ie, a monotonous relationship)

- At lower levels of alcohol consumption, it brings improvements in social life whose health benefits outweigh the health costs of consumption

- At higher levels, this reverses.

Thus explaining the U-shaped effect (quadratic, as the study terms it).

View on HN

Not at all. The most likely explanation to me is the causality is reversed. Poor health or limited social life causes both abstinence and mortality.

Nothing in the abstract suggests they did any analysis to determine causality which makes their statement at the end "showing that survival varies as a function of one’s own and one’s partner’s drinking" misleading, as it could easily be interpreted to imply causality. Of course they phrased it that way on purpose because implying causality makes their results sound way more important than they are.

> Poor health or limited social life causes both abstinence and mortality.

Your theory doesn't explain the observed positive correlation to lifespan when partners shared the same drinking habit.

A compatible partner is part of "good social life".

It might - if both partners feel good enough to go out and have a drink together, they are probably healthy. Or at least healthier than those who don't feel good enough to do so.

> Of course they phrased it that way on purpose because implying causality makes their results sound way more important than they are.

How do you know that? I am at a loss to understand how someone on HN can read one sentence and begin drawing conclusions about what other ulterior motives they had while writing the sentence?

Maybe that sentence means exactly what it says and nothing more which is Z is a function of X and Y? Why perform psychoanalysis on that sentence? And if you have to perform psychoanalysis of that sentence, at least substantiate your conclusion by telling us what process you followed to rule out about a 100 other possible ulterior motives and conclude that "makes their results sound way more important than they are" is most definitely the ulterior motive they had?

That’s how I understood it, too. From what I’ve come to learn, social health is most correlated with longevity. Despite alcohol’s negative physical effects, it appears to have net positive effects overall in terms of longevity when consumed at moderate/low doses.

This finding keeps coming up in social science research: being connected to people seems to increase longevity. Which begs the follow up question: what is the underlying mechanism(s) that causes this effect?

Does social comparison promote self maintenance? Does the sharing of ideas lead to better cognitive hygiene? Does sharing power tools lead to less household accidents? Do we need to sniff other people's pheramones?

On a sceptical note, do extraverts, the sort of people who consent to be studied, continually arrive at conclusions that justify extraversion, missing the huge cohort of introverts who are quite content with minimal social contact?

> This finding keeps coming up in social science research: being connected to people seems to increase longevity.

Has a causal relationship in that direction actually been shown in research, or just correlation?

Because it seems very likely (well, certain for at least some people based on anecdotal experience) it goes in the other direction as well, i.e. someone who has serious health issues which cause them pain or low energy levels or whatever, would likely have less of a social life than if they were completely healthy even if they were the most social person imaginable before their health issues.

If for example I were diagnosed with cancer tomorrow and died of it in ten years time, I suspect it would significantly reduce the energy I have for social activity in the next ten years, but a study done just before I die finding correlation between my not having a good social life and my impending death wouldn't change the cause of death from cancer to lack of social contact.

(FYI I'm not arguing that social activity doesn't have any positive impact on physical health / length of life, in fact I'm sure that it can - I just don't know to what extent research has or hasn't shown how much it's one way or the other.)

Having seen elderly people decline, I honestly wouldn’t be surprised if it’s simply that that have a help network in place when they need it. I actually have a widowed neighbor approaching 90 with limited mobility but she’s been able to stay in her home comfortably enough because we have a rotation on our block that brings her meals (eg. We feed her on Wednesday, etc). She doesn’t really need the economic support but having someone cook and prepare and clean for her at each meal is a reason many people her age would move into an assisted living facility. Despite a high degree of social contact, duration of life once entering a facility like that is pretty low. She’s across the street from me so I even go take out her trash weekly to the curb and make sure her lawn guys are keeping things up to her standards (simple stuff like “hey, can you fix that sprinkler head”)

Alcohol consumption could be something as simple as being a net positive for health in moderation after accounting for the impact of various stresses the average person encounters in life. These things are often encountered as a couple for married folks.

I think there’s been research regarding the supposed “J-curve” for alcohol consumption — essentially, the original research that led to the idea of moderate consumption being better than none failed to differentiate between people who have never had alcohol, and those who stopped drinking due to health conditions. I wouldn’t be surprised if this “couples drinking together” also has a similar bias.

>At lower levels of alcohol consumption, it brings improvements in social life whose health benefits outweigh the health costs of consumption

Patently false, as we know today that the level of alcohol consumption that's best for you is ZERO.

French people seem evidence against that.

In what way? There has been plenty of studies on this and the general consensus is that while there may be certain health benefits to alcohol under certain conditions, in aggregate alcohol in any amount does not have a net positive effect on overall health.

French people are adopting the Western diet in considerable numbers with all of the fast food chains getting in there. Obesity is on the rise. I have not been there recently but I expect to see things have changed across all demographics apart from the elderly.

France cannot be used for comparison studies when it comes to diet and alcohol in the way it used to be.

I'm obviously talking about the ones that eat the traditional diet, not American fake food

The counter argument for the J curve/French Paradox study is that their data for people who don’t drink is biased — it doesn’t distinguish between people who never drink because they don’t want to, and those who stopped due to underlying health conditions (thus making it appear as if mortality among those who don’t drink is higher).

That's the current WHO position apparently, but it's only motivated by lack of evidence of a threshold below which alcohol is safe. I find that approach actually harmful, as it creates an impression that there's little difference depending on the amount consumed, when in reality it's the amount that makes ALL the difference.

I'm thinking about substance use in general that it might even have some benefits but with each repetition the problem use threshold is getting closer. And we have no way of knowing where the threshold is for each of us. It might be ten reps or ten thousand, but even considering the genetics, trauma, etc we cannot know whether the next rep is going to be over the threshold. In addition the threshold in most cases becomes visible after a long time it's been crossed.

The same thing applies to things like chocolate. Yet no one says that since we can't pinpoint a threshold of chocolate consumption harmfulness, we need to assume it's ZERO.

I think it's likely that the WHO is considering people like me in their recommendation.

I have friends who can drink a beer, maybe two, and be quite happy and satisfied. This is a phenomenon I have never been able to understand, because even a small amount of alcohol switches my reward system to REQUIRE more.

If I have zero alcohol, I'm fine. But if I have one beer, I WILL have 7 more. For some portion of the human population (I've seen estimates at 15-20%), the concept of moderation is impossible - and for us, zero is the correct amount.

If the WHO said "drink moderately if you don't experience dopaminergic response to ethanol, but drink zero if you do", it would be more nuanced and accurate, but public health organizations need to target their recommendations to the LCD, as simply as possible.

It's the same as dietary cholesterol and salt. No health organization is going to say "run a double-blind study on yourself to find out if you are part of the population for whom dietary cholesterol impacts serum cholesterol, or sodium intake affects blood pressure, then consult this chart to see what our recommendation is for you" Instead they'll just say GROG SAY SALT AND EGGS BAD, NO EAT.

I have a hunch if you tried a GLP-1 agonist, this reward center would evaporate.

https://www.nature.com/articles/s41598-023-48267-2

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8820218/

Yes, I've seen this, but I am a bit of a fitness enthusiast, and nobody's going to prescribe semaglutide or tirzepatide until I gain 200lbs and wreck my health, which I am unwilling to do.

I'm not sure if you're tried visiting a doctor lately with your own research, but they look down your nose at you if you try to advocate for your particular edge case, with pubmed articles in hand - because they are better than you, as a filthy plebian patient.

These are the people who murder a quarter million Americans annually due to preventable medical errors. Why anybody with an IQ over 90 would ever bother doing business with them is beyond me.

At this point, I'm just waiting for the more reputable darkweb market suppliers to catch up.

But yes, you're right, I'm looking forward to trying this approach, but I refuse to grant moral sanction or financial resources to the Gatekeeper class who demand we beg Daddy's permission to make decisions about our own bodies. They are all evil, full stop.

All fair points.

I'm similar to you there; the old addage "one is too many, two is too few" (or whatever the current version of that is) always ruled my drinking; I didn't need alcohol, but when I started I'd keep going - nice to have a name for it.

Unrelated to any of this, I have a number of mental health issues, and recently I found that alcohol no longer gives me any of the euphoria effects that is used to do (obviously I still had the negative effects); once I noticed that, my desire for alchol went to zero and I've not touched it since - no idea if I am still have the same response to alcohol, but I'm not going to test it, for obvious reasons :)

That would be "the level of alcohol consumption that's best for your _health_"

First off, what's "best for me" is subjective and as CS Lewis describes in "The Humanitarian Theory of Punishment", the pathologizing of my own conditions that I don't myself regard as pathology is a form of despotism.

There's just one thing after another: your rude, brusque dismissal "Patently false" and then appealing to false consensus "as we know today" (sorry you don't know this, there's some isolated recent evidence of alcohol being harmful to health, but hardly anyone actually abstains completely in such a puritanical manner, and LIVING is harmful to health at the end of the day).

I'm not sure if it's because of the alchohol, or the pre-selection of type of people that don't drink, by choice or health

Reply on news.ycombinator.com