This article drastically overstates the evidence, which comes not from experiments but from epidemiology--that is, purely observational studies where you ask pe

This article drastically overstates the evidence, which comes not from experiments but from epidemiology--that is, purely observational studies where you ask people what they eat and follow them to see what happens. This type of study has a horrible track record, and there's a solid theoretical basis for expecting them to produce false results. That there are "more than 400" such studies is irrelevant, because the problem with epidemiological studies isn't a problem with sample size, and the same confounder can distort all 400 studies in the same way.