Hacker News Reader

Comment by bin_bash | Hacker News Reader

Isn’t the problem with these studies that people that drink no alcohol are often not doing so because of bad health or because they were alcoholics?

e.g.: https://time.com/5376552/how-much-alcohol-to-drink-study/

View on HN

This study is only looking at people with heart disease -- not healthy individuals.

Still, there's confounders. The people who had heart disease and drank more probably have a more active social life than those with heart disease who don't drink.

Of course. There are lots of potential reasons.

The most obvious to me is that alcohol is a blood thinner, much like aspirin, which is known to be beneficial to people with heart disease.

Note that the maximum benefit was found to be only a single drink a day - not a large dose.

People who drink in moderation likely also eat bad food in moderation, or smoke in moderation, or argue with their partner in moderation, or work in moderation, or stress in moderation or all sorts of other things over their lifetime.

The blood thinning does sound like an interesting point but it's still really hard to see this as evidence without accounting for factors unknown.

A lot of things cause heart disease and those things in more moderation probably stave of off future problems.

Read the study. Heavy drinkers with heart disease also had lower risks than non-drinkers with heart disease.

It’s just that moderate drinkers were the group with the largest reduction in risk.

> Note that the maximum benefit was found to be only a single drink a day - not a large dose.

Maybe it's just me, but a drink a day seems like a worryingly large dose.

People with heart disease can’t also have been alcoholics or quit drinking due to other health concerns?

The study separated out "never drinkers" from "former drinkers" to address this bias.

https://bmcmedicine.biomedcentral.com/articles/10.1186/s1291...

Possibly, or a myriad of other factors as well, which is why designing proper control/test groups is extremely difficult.

Vast majority of Muslims don't drink due to religion, shouldn't that provide big enough sample size to definitely decide if light drinking is good or bad.

Vast majority of Muslims also smoke a lot.

There's a truly stupendous amount of generalization going on in this thread, is there any evidence for this?

Check the stats for "largest Muslim country by population" (12%+ of world Muslim pop)followed by "% smoking men in Indonesia" (63%+) to see where that particular generalisation comes from. Live next door and it's pretty sad, even before covid came along and added to the risks

So your argument is that 63% of 12% constitutes a "vast majority"? That seems pretty tenuous.

It might be better to avoid making sweeping generalizations about a religion with on the order of 2 billion adherents as if it were some kind of monolith.

Edit: also the prevalence of women in Indonesia smoking is between 2% and 6% depending on what population you're looking at.

https://tobaccocontrol.bmj.com/content/8/3/327

I read "vast majority" as hyperbole. The smoking rate is far higher in Muslim countries than Western though—for men.

Even then it's not really true. I'll grant there's a significant outlier of Indonesia, but after that it's places like China, Russia, and eastern Europe that lead the way.

I'm also not sure why you'd care whether it's men or women doing the smoking, given that the data shows that the risks are similar. If you correctly include the whole adult population, there are no "Muslim" countries other than Indonesia in the top 10. This is just a pejorative and it really needs to be quashed.

The top 10 is: Kiribati, Montenegro, Greece, Timor, Nauru, Indonesia, Russia, Bosnia and Herzegovina, Serbia, Chile

https://ourworldindata.org/smoking#prevalence-of-smoking-acr...

Smoking rates in OIC countries?

It's absolutely not the case that a "vast majority" of Muslims live in OIC countries. If that's what was meant, they should have said that.

It seems it should be easy to control for it.

Reply on news.ycombinator.com