The Seven Countries Study was designed to confirm, not discover

Ancel Keys's Seven Countries Study — the empirical backbone of the fat hypothesis — selected its countries without objective criteria. Keys studied Italy, Greece, Yugoslavia, Finland, Netherlands, Japan, and the US. He left out France and West Germany. Both had high saturated fat intake and low rates of heart disease. Including them would have undermined the hypothesis. It is hard to avoid the conclusion that he chose countries he already suspected would support his position.

The study showed a correlation between saturated fat intake and heart disease mortality, exactly as Keys predicted. It became canonical, the basis for a cascade of subsequent papers, and eventually the scientific foundation for the 1980 US Dietary Guidelines. What it never established was causation. Epidemiology finds patterns in populations; it cannot rule out confounders. Keys never conducted a controlled trial.

When Alessandro Menotti, the study's lead Italian researcher, went back to the data years later, the food that correlated most closely with heart disease deaths wasn't saturated fat. It was sugar.

Source claim: Keys selected only countries likely to confirm his fat hypothesis, excluded inconvenient counterexamples, and a later reanalysis found sugar — not saturated fat — was the stronger correlate with heart disease deaths.