The Science of Low Fat Diets

The idea that what we eat affects our health is extremely popular, both in the public imagination and among scientists. Controlled studies of the connection between diet and disease, however, have tended to disappoint in the past-a study of dietary fiber showed no effect on colon cancer, and studies of various vitamins have shown no protection against cancer. Low-fat is the latest theory to fall, but it is a failure the medical community finds particularly hard to accept. Though no research has ever confirmed a link between overall fat intake and incidence of cancer, the belief that high fat consumption increases the risk, especially of breast and colon cancers, is widespread. In large part, the hypothesis comes from comparing rates of cancer among different populations; the Japanese, for example, consume far less fat than Americans and have much lower rates of these cancers. As Jacques Roussouw, a researcher at the National Heart, Lung and Blood Institute and one of the heads of the study, explained to the Washington Post, "We set out to test a promising but unproven hypothesis that has proven to be less promising than we anticipated."

Researchers found absolutely no difference in rates of colon cancer between the two groups, and although low-fat participants had slightly lower average low density lipoprotein (LDL-"bad" cholesterol) at the end of the study, they suffered heart disease and strokes at the same rates as non-dieters. Breast cancer was the one disease where researchers saw some difference between the groups, with the low-fat group having a 9% lower incidence of invasive breast cancer. Even this difference was so small, however, that it was not statistically significant-in other words, it could easily be chalked up to random chance.

Though it was intended to end the low-fat debate with conclusive, scientifically validated data, if anything, the WHI study stirred up the pot. Many physicians and researchers who have committed themselves to the low-fat camp refuse to accept the study's conclusions, arguing instead that the failure to find a link between low dietary fat and disease prevention was due to flaws in the study. First and foremost, critics point to the fat stats, claiming that the low-fat group did not cut fat intake enough for it to affect their health. Had the women actually reached and maintained the 20% fat-calorie target, these experts argue, the results of the study would have looked very different. With fat consumption creeping up to 30% by the end of the study, the difference between the study group and the control group was, in some minds, relatively insignificant. The study's authors disagree, claiming that even the 8.1% difference in fat intake was statistically important and should have generated different outcomes for the two groups, if low-fat was really beneficial.

Critics also take issue with the approach the study took to fat, namely, treating all fat as created equal. The instructions were to reduce total fat consumption, and the study didn't distinguish between the so-called "good" fat in foods like nuts, fish and olive oil, and saturated and trans fats, which are prevalent in meat and processed foods. There is strong evidence that saturated and trans fats are linked to heart disease, and they are therefore suspected of being generally unhealthy; natural, unsaturated fats, on the other hand, have been shown to raise high density lipoprotein (HDL-"good" cholesterol) levels. A number of public health experts, therefore, argue that the study should have differentiated between types of fat, encouraging women to eat moderate amounts of "good" fats while sharply reducing saturated fat. This diet is known as the "Mediterranean" diet, because it resembles the eating habits of people in Mediterranean regions like Italy, who consume large amounts of olive oil but little saturated fat. Peter Libby of Harvard Medical School is a strong proponent of the Mediterranean diet, and told the New York Times that until a study disproved its benefits, he would stand by his "cherished hypothesis."

Dissatisfied experts also took issue with the choice of participants and the time span of the study. Post-menopausal women might not have been the best population for studying the benefits of a low-fat diet, because adopting such a diet earlier in life might be more effective. Finally, some experts maintained that while 8 years sounds like a long time, in the scheme of chronic diseases, it isn't. To fully see the benefits of radical diet changes, researchers would have to follow patients, these experts claimed, for a considerably longer period-perhaps for the rest of their lives.

Other scientists argue that the study's conclusions are solid and should not be ignored. David Freedman, a statistician at the University of California at Berkeley, who has written books on how to run sound clinical trials, told the New York Times that "the studies were well designed." They set out, he stated, to "confirm popular hypotheses about the protective effect of diet," but hard data told a different story. While the study does not speak to things it did not test, like the Mediterranean diet, with regard to the particular question it sought to answer-whether cutting overall fat prevented cancer or heart disease-Freedman believes it should be considered conclusive.

Sources

Kolata, Gina. "Low-Fat Diet Does Not Cut Health Risks, Study Finds." New York Times, February 8, 2006, page A1.

Prentice, Ross et al. "Low-Fat Dietary Pattern and Risk of Invasive Breast Cancer." JAMA, February 8, 2006, page 629.

Stein, Rob. "Low-Fat Diet's Benefits Rejected." Washington Post, February 8, 2006, page A1.

Stein, Rob. "Low-Fat Diet Prompts Modest Weight Loss." Washington Post, January 4, 2006, page A3.

"Reducing Total Fat Intake May Have Small Effect on Risk of Breast Cancer, No Effect on Risk of Colorectal Cancer, Heart Disease or Stroke." Press release on the study from the National Institutes of Health. (February 7, 2006) www.nih.gov/ news/ pr/ feb2006/ nhlbi-07.htm.