Why It's Hard to Study the Role of Diet in Cancer

So . . . what should you eat to prevent cancer? Getting a definitive answer is tough because research on diet is especially challenging. Here are some of the reasons why it is hard to isolate the effect of diet on cancer:

It is hard to measure diet. Diet is not measured by a machine, but by what people report they eat. Because it is hard for people to remember exactly what and how much they eat, these self-reports may not be accurate.

It is hard to change diet. It's hard to get people to make and sustain big changes in their diet. In both studies, women in the low-fat diet groups attended multiple sessions with dietitians to help them keep to the diet. But even at the end of the first year, when compliance was highest, less than one-third of women had met their dietary goal (15 percent of calories from fat for one study; 20 percent of calories from fat for the other).

It is hard to isolate which dietary change matters. Researchers suspected that beta carotene, a chemical found in carrots, might reduce lung cancer risk. But in a randomized trial, beta carotene supplements did not work. It is still possible, however, that eating carrots would work. The researchers may have guessed wrong about the protective ingredient. Or there might be some complex interaction between beta carotene and other carrot ingredients. Or the beta carotene in supplements may be metabolized differently than the beta carotene in carrots.

In contrast, the low-fat diet researchers tested a broad dietary change. But now the problem is knowing whether the reduction in breast cancer risk observed resulted from the drop in dietary fat or some other change. For example, in the study of women with breast cancer, the low-fat diet group also lost weight (on average about six pounds). Was it their low-fat diet or was it their weight loss that lowered their breast cancer risk? No one knows.

-- Steven Woloshin, Lisa Schwartz, H. Gilbert Welch