"Large scale epidemiological studies may have problems but they are enough to tip the burden of proof towards those who would disagree with them." No! Other th

"Large scale epidemiological studies may have problems but they are enough to tip the burden of proof towards those who would disagree with them." No! Other things are often highly correlated with diet, and has massive effects on cardiovascular health (exercise in particular). Only randomized controlled dietary trials can be trusted for predictive power.