That study seems ... wrong. Estimating actual alcohol drunk from estimates of sales? Separate model for tourist consumption? Non linear spline model for distrib

That study seems ... wrong. Estimating actual alcohol drunk from estimates of sales? Separate model for tourist consumption? Non linear spline model for distribution of drinking at 0-12.5 a day rates? No accounting for differences between cultures which binge drink (Northern europe) and countries with steady, consistent alcohol consumption with meals (southern europe). Yeah, seems legit. Not. And then figure 5 basically tells us the common wisdom ... a drink a day is harmless or good for you. I think it's pretty well established that drinking a whole lot is pretty bad for your health, and not drinking so much is at least not real harmful. Anything that gives totalitarian proscriptions of booze, particularly when accompanied by figure 5; well, that reminds me of back in the day when they insisted smoking a joint would poke giant wholes in my pre-frontal cortex and make dying of a heroin overdose inevitable.