One thing I have noticed is that studies that end up with conclusions that sound like BS end up being BS studies. The effect size of a 50% increase of dementia

One thing I have noticed is that studies that end up with conclusions that sound like BS end up being BS studies. The effect size of a 50% increase of dementia risk sounds way to high for coffee. Seems like there is a good chance of some correlated variable they are missing or a faulty assumption.