Willett’s curiosity about disease, and how it might be prevented, dates to a particularly traumatic period in his own life. When he was 13, his father — a reproduction physiologist who worked to improve dairy production from cows like those his family raised in Michigan — died of brain cancer. As the oldest of the four children, Willett had to grow up fast. He began planting vegetables on a neighbor’s farm, later selling them to cover his college tuition.
During medical school at the University of Michigan, Willett became curious about the Quaker religion and attended a Friends Meeting. There, he was introduced to a nursing school student named Gail Pettiford. Although they were of different religions and different races — he was brought up in Congregationalist and Methodist churches, she was African-American — their personalities meshed. At the time, so did their diets, since both ate thoroughly Midwestern fare of red meat, potatoes, and homemade desserts. Willett’s mother kept a small bottle of olive oil somewhere in the back of her cabinet, but it grew rancid from disuse. Gail’s father would express his pleasure with a hearty meal by saying, “Tastes like more.”
Walter and Gail moved to Boston, where he did his residency in internal medicine and began his studies at the Harvard School of Public Health. After marrying, they moved to Tanzania, where he taught medical school for three years.
Back in Boston, Willett joined the Nurses’ Health Study in 1977, its second year. It had actually begun as a study of the long-term implications of oral contraceptives. But Willett, who had shifted his focus from internal medicine to nutrition and epidemiology, succeeded in expanding the nurses’ study to match his own interests.
He began collaborating with colleagues of Greek, Italian, and Spanish heritage, who exposed him to the flavors of Mediterranean cuisine. In time, he noticed that the people in his cohort studies who were eating the oils, whole grains, and fresh fruits and vegetables that were staples of the Mediterranean diet were faring better.
By the early 1990s, he’d seen enough. As many Americans were racing to replace fat with lots of pasta and SnackWell’s cookies, Willett encouraged his family to go the other way, eating lots of “healthy fat” oils, nuts, and avocados, and very little red meat or sugar.
Gail Willett was reluctant at first, especially given her love of ice cream and other sweets. Their two sons, Amani, who was around 14 at the time, and Kamali, 10, had even less enthusiasm. In day care, Amani had been known to trade his apple for a Twinkie. But Willett eventually won out. In his typical fashion, he didn’t push this change by fiat, but rather by a persistent refrain of “What do you think if we tried this?”
AT THE OFFICE, Willett and his team continued to mine their enormous data sets, publishing hundreds of papers with all sorts of insights about healthy living. Over time, his ideas took hold.
Willett has been particularly influential in a war on trans fats that he began waging in earnest in the early 1990s. We now consume about three-quarters less of them than we did just a decade ago. Willett also effectively campaigned to discredit the US Department of Agriculture’s familiar food pyramid, exposing how agribusiness lobbying, not scientific data, drove its design. He produced an alternative food pyramid, then promptly sent his version off to the USDA, as if to show the agency how it was done.
“It’s hard to think of a nutritionist in the country who has more influence,” says Gary Taubes, a best-selling science author who has written extensively about the ills of refined carbohydrates. He says the kinds of studies Willett does have been instrumental in helping to identify risk factors for disease: how high blood pressure is associated with heart disease, how the low breast cancer rates among women in Japan rise dramatically within two generations of their moving to the United States, and how obesity is associated with a host of chronic diseases.
But as Willett’s influence has grown, so has criticism of the very nature of his research. That’s because large cohort, or “observational,” studies can show associations and correlations, but never causation. The nurses’ study can suggest that people who took folic acid had a lower risk of colon cancer, but that doesn’t prove folic acid — and not something else — actually reduced the risk.
Taubes, who recently cofounded the nonprofit Nutrition Science Initiative for research, stresses that observational studies are equipped only to generate hypotheses. To prove them, scientists have to turn to a randomized controlled trial. In these, participants are randomly assigned into either an intervention group, which gets the treatment or special diet, or a control group, which gets a placebo or a standard diet. Continued...