I remember how there was that study about how "alcohol is good for you" because the "mortality vs. alcohol consumption" was a slight bathtub curve at the low en

I remember how there was that study about how "alcohol is good for you" because the "mortality vs. alcohol consumption" was a slight bathtub curve at the low end. They completely forgot that many people who don't drink do so because of other confounding health complications that may increase mortality. Talk about not controlling for variables, yeesh :|