> allthough one would think they'd get 0% alcohol related cancers

I assume "alcohol related" in this context means that alcohol consumption increases the risk for those types of cancers, but you might still get those types of cancers even if you have never consumed any alcohol. And "less than once drink per week" is assumed to be almost the same as never consuming any alcohol at all, so 17% is the risk for women who never consume any alcohol.

Then the phrase "alcohol related" is massively biased wording.

That's like calling death from bleeding out "gun related death", because people who get shot often die of bleeding out.

Ah yes that must be it. 17% of women get these cancers, which are cancers that you have a higher risk of when drinking, but in this case they are not caused by drinking. I though < 1 per day is still drinking 6 special Belgian beers of 8-10 % alcohol on a Saturday so I thought, that's still quite something. At least I'd be heavily hung over the next day. I expected the baseline to just be "non-drinking".

Very poorly written I'd say.

In your quote, the number is "<1 per week", not "<1 per day"

Oh... right, pff, this invalidates all my concerns indeed. Apologies.

Hahaha there's the rub.

This data is self reported of course, and "one drink daily" are people who actually drink a lot more.

If you actually drink less than daily, very little to worry about AFAICT.

Just the standard amount of worry then. For the standard person with the standard lifestyle.

I imagine it's because they count any cancer that could be caused by heavy-drinking as an "alcohol-related cancer", regardless of whether it was actually caused by alcohol or not.

How did they establish causality, or is it just a correlation potentially resulting from hidden variables? E.g., I do not know many smokers that do not touch alcohol, but plenty of non smoking abstinates. I also know plenty of sedentary obese heavy drinkers, where a higher percentage of the teetotal lot are more health obsessed in other ways.

They can't really. This data is almost always self-reported. They can attempt to eliminate common confounds but the data is very noisy. People have a lot of bias and inaccuracies when reporting the primary data. Even questions like "how many cigarettes per day/week/month/year do you smoke?" is subject to enormous inaccuracies. And the more questions re confounds which are asked, the more the noise, and the lower the number of questionnaires returned. And no matter how many confounds are included in the questionnaires, there are likely a thousand more which also matter. It's a fairly major issue in health research, and why we so often see headlines with conflicting findings every few years. See research into the health of eggs, for example. One should be highly skeptical of correlative health evidence at this point (and this is what most of the clickbait studies are). Instead, try to focus on the causative research where they identify specific genes or chemical mechanisms which cause outcomes. This is much more difficult, of course, and sparse.

I wish there was more causative research where nutrition was artificially changed in a highly managed population. For example in a prison or military barracks.

I suspect there's a lot we could learn about the health of eggs, for example, if we could just pay some prisoners to eat varying quantities of them for 20 years, then look at the development of their health.

While I applaud your desire to have better controlled data, there may be some selection bias interfering with the results.

You should message Vault-Tec about that opportunity you seek, they might already have a suitable Vault for that.