My rule of thumb is: if you have a bunch of conflicting studies on the risks of a certain activity like this, the risk is probably pretty low if it exists at al

My rule of thumb is: if you have a bunch of conflicting studies on the risks of a certain activity like this, the risk is probably pretty low if it exists at all. If it was a large risk it would be easy to see, and in the case of moderate alcohol drinking it isn’t. On the other hand, all cause mortality might not be the only end point we care about here. If I live just as long, but my health is poorer during the last decade, that’s a loss also. I don’t know how you would quantify that.