One Comment

  1. Salty Randall says:

    Metanalyses suffer from compounding error. 100 correlational studies, analyzed at a group, do not come any closer to causation than a single study. That said, there is no evidence that drinking the stated amounts “causes” cancer. Is it possible? Of course. But the myriad of variables that infest cross cultural and cross national populations make any such conclusion to be foolish.

Leave a Reply

Your email address will not be published. Required fields are marked *