Dietary fat is the primary cause of heart disease, and a healthy diet should be low in all fats.

Ancel Keys was a formidable scientist and a formidable personality. His Seven Countries Study, launched in 1958, tracking 12,770 men across Finland, Greece, Italy, Japan, the Netherlands, the United States, and Yugoslavia, became the most influential nutrition study of the twentieth century and the foundation of a dietary consensus that shaped what Americans ate for four decades.

Keys's conclusion was clear: saturated fat raises blood cholesterol, elevated cholesterol causes heart disease, therefore nations eating less saturated fat have lower heart disease rates. Mediterranean countries like Greece and Italy, where olive oil and fish dominated diets and heart disease rates were low, seemed to confirm this. Japan, where dietary fat was low and heart disease rates were very low, seemed to confirm it further.

The McGovern Senate Select Committee on Nutrition and Human Needs translated Keys's research into the first federal dietary goals in 1977: reduce fat consumption to 30 percent of calories, reduce saturated fat to 10 percent. The theory had intuitive appeal. Fat contains nine calories per gram compared to four for carbohydrates and protein. Fat is fat; eat fat, accumulate fat. The simplicity of the message was part of its power. It could fit on a food package. It could animate a federal guideline. The USDA Food Pyramid that followed in 1992 placed fats at the narrowest tip — "use sparingly" — while grains occupied the broad base, intended as the foundation of the healthy American diet at six to eleven servings daily.

The food industry responded with breathtaking speed. If fat was the enemy, fat needed to be replaced. The replacement was sugar and refined carbohydrates. Low-fat cookies, low-fat yogurt, fat-free salad dressings, and reduced-fat peanut butter crowded grocery shelves. Snackwell's cookies — fat-free, plenty of sugar — became a cultural phenomenon. Food manufacturers removed fat, added sugar to maintain palatability, and stamped "low fat" health halos on the packaging. What no one had adequately studied was whether replacing fat with carbohydrates improved health outcomes.

It did not. Obesity rates that had been relatively stable through the 1970s began climbing sharply through the 1980s and 1990s. By 2000, roughly one-third of American adults were obese, up from around 15 percent in 1978. The critique of the fat hypothesis was building in research circles through the same period. Walter Willett at Harvard's School of Public Health published mounting evidence that the fat-heart disease link was far weaker than Keys had suggested, that type of fat mattered enormously, and that refined carbohydrates were their own cardiovascular risk factor.

The clinical trials confirmed the revision. The Women's Health Initiative Dietary Modification Trial — the largest randomized controlled trial of a low-fat diet ever conducted, involving nearly 49,000 women over eight years — found no significant reduction in cardiovascular disease in the low-fat intervention group compared to controls. Results published in JAMA in 2006 were received with genuine surprise. Meanwhile, the PREDIMED trial, published in the New England Journal of Medicine in 2013, found that a Mediterranean diet supplemented with extra-virgin olive oil or nuts — explicitly high in fat — reduced major cardiovascular events by approximately 30 percent compared to a low-fat control diet.

The 2015 Dietary Guidelines for Americans quietly removed the total fat restriction that had governed federal nutrition policy since 1977. The guidelines noted that fat type, not fat quantity, was the relevant variable. The food supply had spent nearly four decades optimizing for the wrong number. The experiment in population-scale dietary engineering, conducted without the clinical evidence to justify it, had ended — but its effects on the American diet, and on American waistlines, are still being worked through.