Eh, I'm not sure I totally agree with what you are saying: a) As you mentioned, smoking is a huge confounder in your example b) A lot of congenital heart dise

Eh, I'm not sure I totally agree with what you are saying: a) As you mentioned, smoking is a huge confounder in your example b) A lot of congenital heart disease is not preventable via dietary intervention. For people who do require dietary intervention it's going to depend on the disease. For a small portion of people, for instance, eating low cholesterol is important, but the vast majority it makes no difference. For these people eating low cholesterol is the biggest difference, not a generally balanced diet. c) There is a ton of science showing massive improvements in heart disease outcomes from losing weight. I'm not saying eating a balanced diet doesn't matter, but for most people actually being a healthy weight is the most important thing, and for many others the dietary intervention is pretty specific, ie: not low cholesterol not "eat a balanced diet."