They close with this: However, a more critical review[0] of the epidemiological data that took into account genetic variants which impact the effect of alcohol intake on longevity shows that the described J-shaped relationship is not as robust as once thought. In conclusion the review simply states: The optimal level of alcohol intake for health and longevity is zero [0] https://academic.oup.com/ageing/article/49/3/3…
This is the conclusion of the most recent lancet study on alcohol and health. However, the lancet study comingles societal and personal health affects, which I find annoying (drunk driving, increased violence, etc)
From everything that I've read, alcohol has a j-curve benefit to cardio health, and a j curve detriment to cancer rates. When you look at mortality rates vs alcohol consumption, the J part of the curve doesn't return to 'tee-totaler' non drinker baseline until you're at ~ 4 drinks a day, which is a lot.