Over the last quarter-century, Burkitt’s fiber hypothesis has become yet another example of Francis Bacon’s dictum of “wishful science”—there has been a steady accumulation of evidence refuting the notion that a fiber-deficient diet causes colon cancer, polyps, or diverticulitis, let alone any other disease of civilization. The pattern is precisely what would be expected of a hypothesis that simply isn’t true: the larger and more rigorous the trials set up to test it, the more consistently negative the evidence. Between 1994 and 2000, two observational studies—of forty-seven thousand male health professionals and the eighty-nine thousand women of the Nurses Health Study, both run out of the Harvard School of Public Health—and a half-dozen randomized control trials concluded that fiber consumption is unrelated to the risk of colon cancer, as is, apparently, the consumption of fruits and vegetables. The results of the forty-nine-thousand-women Dietary Modification Trial of the Women’s Health Initiative, published in 2006, confirmed that increasing the fiber in the diet (by eating more whole grains, fruits, and vegetables) had no beneficial effect on colon cancer, nor did it prevent heart disease or breast cancer or induce weight loss.

“Burkitt’s hypothesis got accepted pretty well worldwide, quite quickly, but it has gradually been disproved,” said Richard Doll, who had endorsed the hypothesis enthusiastically in the mid-1970s. “It still holds up in relation to constipation, but as far as a major factor in the common diseases of the developed world, no, fiber is not the answer. That’s pretty clear.”

As we have seen with other hypotheses, the belief that dietary fiber is an intrinsic part of any healthy diet has been kept alive by factors that have little to do with science: in particular, by Geoffrey Rose’s philosophy of preventive medicine—that if a medical hypothesis has a chance of being true and thus saving lives, it should be treated as if it is—and by the need to give the public some positive advice about how they might prevent or reduce the risk of cancer. This was immediately evident in a New England Journal of Medicine editorial that accompanied back-to-back April 2000 reports on two major trials—one on fourteen hundred subjects of the Phoenix [Arizona] Colon Cancer Prevention Physicians’ Network, and one $30 million trial from the National Cancer Institute—both of which confirmed that fiber had no effect on colon cancer. The editorial was written by Tim Byers, a professor of preventive medicine at the University of Colorado, who said that the two trials had been short-term and focused only on the early stages of cancer. For this reason, they should not be interpreted as “evidence that a high-fiber cereal supplement or a low-fat high-fiber diet is not effective in protecting against the later stages of development of colorectal cancer.” Byers was wrong, in that the results certainly were evidence that a high-fiber diet would not protect against the later stages of colorectal cancer; they simply weren’t sufficient evidence for us to accept the conclusion wholeheartedly as true.