4 ms·

Seems like we plain do not have a good set of studies on alcohol (or coffee, or...). You can examine that alcohol has been consumed by many cultures for centur

Seems like we plain do not have a good set of studies on alcohol (or coffee, or...). You can examine that alcohol has been consumed by many cultures for centuries and evaluate whether the associated outcomes are acceptable to you.

AstralStorm 9y ago
Epidemiology does not work that way. Too many confounders and very low predictive power.
- projektir 9y ago
  Well, I'm not talking about epidemiology, just personal choice. People had to use incomplete information for personal choices for centuries, I'm afraid. I'm not sure what to do about that, the alternative seems to be chasing after the conclusion of the latest study and a lot of said latest studies are very poor, often in very insidious ways that are hard to identify. If something was literally poisonous and was consumed by lots of people, we'd be seeing drastic effects. Like drastically slashed life expectancy. If something was literally hyper-beneficial and extended your life by 20 years, we'd be seeing some cohort expressing that, as well. If there are lots of confounders, it just means it's not the only problem, which I think is already frequently true for many of these things (i.e., smoking and weight gain associated with lots of root issues). The effects of a lot of things taken in vacuum often seem less significant than their very confounders.
boomboomsubban 9y ago
>Seems like we plain do not have a good set of studies on alcohol What's wrong with the one used in this report? It samples 600000 people over 19 countries. We have plenty of other good studies on the issue, and most of the disagreeing studies are tied to alcohol companies. The life expectancy effects of alcohol are fairly well laid out in this study. Examine those before making your choice, noting that people have drunk alcohol for centuries does not help you make an informed decisions.

AstralStorm 9y ago

Epidemiology does not work that way. Too many confounders and very low predictive power.

projektir 9y ago
Well, I'm not talking about epidemiology, just personal choice. People had to use incomplete information for personal choices for centuries, I'm afraid. I'm not sure what to do about that, the alternative seems to be chasing after the conclusion of the latest study and a lot of said latest studies are very poor, often in very insidious ways that are hard to identify. If something was literally poisonous and was consumed by lots of people, we'd be seeing drastic effects. Like drastically slashed life expectancy. If something was literally hyper-beneficial and extended your life by 20 years, we'd be seeing some cohort expressing that, as well. If there are lots of confounders, it just means it's not the only problem, which I think is already frequently true for many of these things (i.e., smoking and weight gain associated with lots of root issues). The effects of a lot of things taken in vacuum often seem less significant than their very confounders.

boomboomsubban 9y ago

>Seems like we plain do not have a good set of studies on alcohol What's wrong with the one used in this report? It samples 600000 people over 19 countries. We have plenty of other good studies on the issue, and most of the disagreeing studies are tied to alcohol companies. The life expectancy effects of alcohol are fairly well laid out in this study. Examine those before making your choice, noting that people have drunk alcohol for centuries does not help you make an informed decisions.