Eating a lot of fiber-rich vegetables, fruits, and whole grains does not appear to reduce a person's chances of getting colorectal cancer, researchers found in the largest study yet to test the popular and longstanding idea about preventing the third most common cancer.
The research team, led by the Harvard School of Public Health, combined and reanalyzed data from 13 previous studies involving 725,628 adults and found that a person who ate 30 or more grams of fiber each day, the equivalent of more than seven servings of oatmeal, had about the same risk of getting colorectal cancer as a person who ate less than half that amount. The results were adjusted to take into account other risk factors, such as red meat consumption and age.
''This is the most exhaustive study, throwing everything into one pot," said Dr. Robert Mayer, director of the center for gastrointestinal cancers at Dana-Farber Cancer Institute.
Doubts about the relationship between fiber and colorectal cancer have been growing among oncologists and gastroenterologists, as smaller studies of fiber's benefits in recent years yielded widely inconsistent results. But several physicians said that many patients still strongly believe they can prevent colorectal cancer by eating a high-fiber diet. After being diagnosed with colorectal cancer, the doctors said, many patients ask whether they should have eaten more produce and whole-grain cereal.
''It became an urban myth," said Dr. David Ryan, medical director of the gastrointestinal cancer center at Massachusetts General Hospital. ''It takes a lot of time to deconstruct those."
But Dr. John Baron, an internist at Dartmouth-Hitchcock Medical Center in New Hampshire who wrote an editorial accompanying today's study, said that more research is needed about fiber's long-term impact on colorectal cancer, particularly since there are different types of fiber, which may have varying medicinal qualities. And the study did find that a diet of more than 25 grams of fiber a day was associated with a slightly reduced risk of cancer of the rectum alone.
Other physicians said they don't want to send a message that people shouldn't eat fruits and vegetables and whole-grain breads and cereals. Fiber is believed to have other benefits, including lowering cholesterol and helping prevent heart disease and preventing constipation among older people.
Dr. Ronald Bleday, chief of colorectal surgery at Brigham and Women's Hospital, said he recommends about 20 grams of fiber a day, even though he does not believe it prevents colorectal cancer. That's the amount of fiber in one bowl of oatmeal, two apples, one cup of broccoli, and two slices of whole wheat bread.
The new study, published in today's Journal of the American Medical Association, points to the general difficulty of proving that diet can affect one's chances of getting cancer. Studies have shown a strong relationship between a diet low in red meat and a reduced risk of colon cancer, but in many areas cancer researchers are not sure whether diet matters. One reason is that it's hard to study; cancer can take years to develop, and people don't tend to follow a specific diet for five to 10 years. Researchers also are realizing that genetics could end up having a much stronger influence on the occurrence of cancer than diet.
''Cancer is not a simple disease," said Yikyung Park, the lead researcher and a nutritional epidemiologist who now works at the National Cancer Institute. ''There are so many factors that affect risk. Diet is just one of them."
The disappointing studies into fiber and colon cancer also illustrate how strongly medical theories, even tenuous ones, can take hold with Americans.
The idea that fiber might protect people against colon cancer dates back to at least the late 1960s, when British surgeon Denis Burkitt noticed during his work as a medical missionary that people in southern Africa had lower rates of colorectal cancer, precancerous growths called adenomas, and diverticular disease than the English. He also noticed that they ate far more high-fiber vegetables and fruits. When researchers looked at other countries, they often found the same phenomenon: People in nations with the lowest colon cancer rates consumed the most fiber, said Bleday, of Brigham and Women's.
Studies that asked people with colon cancer and healthy people to recall their diets years earlier similarly concluded that a high-fiber diet seemed protective, although researchers now believe these studies contain significant bias because cancer patients tend to remember their diets as being less healthy than they were.
Also, the idea seems biologically feasible; fiber moves feces through a person's system more quickly, limiting the amount of contact that carcinogens have with the bowel, among other possible protective mechanisms.
But what Burkitt didn't appreciate, Mayer of Dana-Farber said, is that Africans ate far less red meat than the British, which may have accounted for their better intestinal health. As researchers conducted more sophisticated studies of fiber and colorectal cancer, the results were all over the map. Some showed that fiber is beneficial, others found no effect, and a few even showed fiber increases the risk of cancer, said Baron of Dartmouth-Hitchcock Medical Center.
In today's study, researchers picked 13 smaller studies that met certain quality criteria and that followed patients for between six and 20 years. Researchers sometimes combine and reanalyze studies, because the larger numbers of participants can reveal stronger results. There were 8,081 cases of colorectal cancer among the participants, who were divided into groups based on the amount of fiber they said they consumed. The researchers did not determine the prevalence of cancer for each group, just the relative risk of cancer across the groups.
Park said the studies she used have a weakness: Participants were surveyed about their diets just once at the beginning of the studies, so the studies did not take into account changes in eating habits or information about their diets as children or young adults.
Baron said the study, because of its large number of participants, may have uncovered a problem for people with extremely low-fiber diets, which most studies wouldn't catch because so few people fall into this category. In the study, participants who ate less than 10 grams of fiber a day had an 18 percent higher risk of colorectal cancer than people who ate more fiber, suggesting that ''colorectal cancer might be a sort of fiber-deficiency disease," he wrote.
The relationship between fiber and good health has been reinforced by food manufacturers. Makers of granola bars, cereal, oatmeal, and bread all advertise that their products are good sources of fiber, though most don't specify a connection to cancer.
Even the American Cancer Society says on its website that to lower the risk of developing colorectal cancer, ''it is important to eat plenty of fruits, vegetables, and whole grain foods and to limit intake of high-fat foods."
Yesterday, the organization said it probably won't change its guidelines, even though the organization's senior epidemiologist, Marjorie McCullough, participated in today's study. McCullough said in an e-mail that fruits, vegetables, and whole grains are too beneficial in other ways, including reducing obesity, a risk factor for cancer.