Chew on This

By Nancy Volkers

Chew on This

Diet, supplement results raise key issues in understanding research

By Nancy Volkers

Researchers recently reported some eyebrow-raising results from the Women’s Health Initiative (WHI)—one of the largest and most comprehensive clinical studies ever undertaken: Maintaining a low-fat diet doesn’t reduce the risk of breast cancer or colorectal cancer, and taking calcium and vitamin D supplements together doesn’t reduce the risk of colon cancer (see below).
However, that doesn’t mean a healthy diet is irrelevant, says Harvard University nutritionist Walter Willett. “It’s very important not to conclude that diet does not make a difference. Especially for heart disease, replacing trans fat and saturated fat with liquid vegetable oils is very beneficial.”

And although the WHI studies on vitamin D and calcium showed no benefit, “There’s actually a lot of evidence that taking vitamin D and calcium will reduce colon cancer risk, and other evidence that vitamin D will lower fracture risk,” Willett says.

So how do we interpret these new results? Experts in the field point out three key issues to keep in mind.


Research studies are constructed in different ways; the particular setup can limit how some results should be interpreted.

Most previous studies of diet and cancer risk were observational studies. For example, researchers studied the diets of a large group of people, and then compared the risk of breast cancer in women who ate a very low-fat diet to those who consumed a high-fat diet.

However, “Observational studies can only show associations,” says oncologist Barnett S. Kramer, director of the Office of Medical Applications of Research at the National Institutes of Health (NIH). “You need randomized controlled studies to study cause and effect.”

The WHI is the first randomized trial to evaluate the effects of a low-fat diet in postmenopausal women. In a randomized trial, women are assigned to an intervention group (for example, following a low-fat diet or taking calcium supplements) or to a control group.

“It is generally believed that a randomized trial provides the strongest evidence,” since the process of randomly assigning people to one group or the other helps balance out factors such as heart disease risk or family history of a certain cancer more evenly between the groups, says medical oncologist and WHI investigator Rowan Chlebowski of the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center in Torrance, Calif.

But even the best-designed studies can have shortcomings. Six years into the low-fat diet trial, the “low-fat” diet group was consuming an average of 29 percent of its calories from fat, lower than the comparison group (about 37 percent fat) but not as low as researchers had been aiming for (20 percent).

In the supplements trial, participants took 400 international units (IU) of vitamin D, the current recommended daily intake. However, Willett says, other trials have shown that people may need 700 to 1,000 IU to effect a reduction in risk.

Another limitation to the colon cancer results was that researchers did not tell women in the placebo group not to take vitamin D or calcium on their own if they wanted to. “That’s a problem,” Willett says. “By the end of the study, more than 69 percent of the women [overall] were taking calcium on their own, so you could hardly expect to find an effect.”


The women in the WHI were all between ages 50 and 79 when the study began. They had never been diagnosed with breast cancer or colon cancer, and had no evidence of other cancer types for the past 10 years. To be included in the study, a woman had to get at least 32 percent of her calories from fat. Women with life expectancies of less than three years and those with conditions that might affect their adherence to the trial’s guidelines (such as alcoholism or dementia) could not take part.

These restrictions mean that for all men, and for many women, these results may not be relevant. Furthermore, says Kramer, “We may not be able to extrapolate the WHI results—and I would not—to cancer survivors. Diet may have differing effects for them.”

Indeed, Chlebowski and others presented preliminary findings last year that suggested that breast cancer survivors who ate a low-fat diet reduced their risk of breast cancer recurrence over five years by 25 percent.

Longer follow-up of the study’s subjects may uncover additional clues about any potential diet–breast cancer link. “Cancer takes a long time to happen, so we think as we follow women for longer periods, the data will be more definitive,” says Cynthia Thomson, a nutritionist and WHI researcher at the Arizona Cancer Center in Tucson.


The complexity of medical research doesn’t lend itself to easy answers. “There is a trite statement, but it’s true: The best studies are the ones that not only answer questions but also raise additional questions,” Kramer says.

The print and broadcast media often have limited time and space to report findings from clinical trials. As a result, a subtle, yet important message from a study’s results may get placed far down in a news story, or may be left out altogether. For instance, biostatistician and WHI investigator Ross Prentice of the Fred Hutchinson Cancer Research Center in Seattle notes that one WHI report did suggest that a low-fat diet is associated with an estimated 9 percent reduction in breast cancer risk. In scientific parlance, however, this reduction wasn’t “statistically significant.” What that means is there’s still a reasonable possibility that the diet-cancer risk reduction link could be due purely to chance.

“It’s only natural that the media interpret [‘not statistically significant’] as ‘not important,’ though it just means that the result didn’t meet a statistical standard,” Prentice says. The WHI findings do not rule out a potential benefit of low-fat diets or supplements, according to Kramer.

Kramer attributes part of the problem in communicating subtleties in scientific results to human nature. “We just want to think of every piece of evidence as certain, and it isn’t,” he says. “Medical literature is like an ongoing discussion, and it’s rare that you’ll reach a punctuation mark.”



The Women’s Health Initiative (WHI) was established by the National Institutes of Health in 1991 to focus on heart disease, osteoporosis, breast cancer and colon cancer. It was planned as a 15-year endeavor that included three main areas: randomized clinical trials, observational studies and community-based strategies for developing healthy behaviors.

The study involved 161,808 women ages 50 to 79. The randomized, controlled clinical trial—the part of the study discussed in the accompanying article—included 68,132 women. The women could enroll in a hormone therapy trial, a vitamin D/calcium trial or a low-fat diet trial (some enrolled in all three trials). Women were followed for seven to eight years.

In February, Women’s Health Initiative researchers reported the following:

  • Following a low-fat diet does not reduce the risk of breast cancer
  • Maintaining a low-fat diet does not reduce the risk of heart disease or stroke
  • Taking calcium and vitamin D supplements together does not reduce the risk of colon cancer
  • Using calcium and vitamin D supplements together does not reduce the risk of fractures