In The New York Times today, Gina Kolata raises the question: can a healthy diet prevent cancer? After surveying the results of recent case control and cohort studies, she concludes that the benefits of a healthy diet are "hypothetical and elusive" when it comes to preventing cancer.
There's a lot of great research in Ms. Kolata's article, but unfortunately the studies she cites examine only modest dietary changes, over relatively short periods, in adults. A broader look at the research reveals a very different picture, and a convincing case that diet is an important tool in preventing cancer, especially in children and young people.
Understanding Different Kinds of Studies
There are different ways of studying the relationship between nutrition and cancer.
Epidemiological studies look at populations with varying characteristics for comparison. These have shown overwhelmingly that there is a connection between diet and cancer.
Case control studies compare two groups: one with the disease in question, and one without. (Past food intake is determined by questionnaires.)
Cohort studies follow two groups over time, looking for differences that appear years later.
As the Times acknowledges, hundreds of epidemiological studies show in any number of convincing ways that there likely is a connection between diet and cancer. In The China Study, for instance (which the Times earlier called "the Grand Prix of all epidemiological studies" and "the most comprehensive large study ever undertaken of the relationship between diet and the risk of developing disease") researchers compared the varying diets in various Chinese towns, and found that as the amount of animal products increased in the diet, even in relatively small increments, so did the emergence of the types of cancers that are common in the West. The researchers noted that most cancers increased in direct proportion to the quantity of animal products eaten and decreased relative to the amount of fruits, vegetables, and beans consumed. Areas of China with exceptionally low intakes of animal products were virtually free of the cancers and heart disease that develop in most people living in Western countries.
The China Study is one of many epidemiological studies showing a diet-cancer connection. Among those aged 50-75, cancers of the digestive tract, breast, and prostate are 20 times higher in the United States than in Laos, Cambodia, and Thailand. The huge geographic variability in the incidence of these cancers suggests dietary factors as the main cause. When people from a low-risk country migrate to the United States, their cancer rates increase considerably, and their offspring get cancer at the same rate as other Americans. This demonstrates that the lower incidence of these cancers in Asia is not due to a lower genetic susceptibility in Asians, but rather to the lack of exposure to Western lifestyles.
Fat, particularly animal fat, has been implicated as a cause of cancer, while the consumption of fruits and vegetables had been shown to protect against cancer. For instance, Boyd et al reported on this in a 1993 study published in the British Journal of Cancer, as did Steinmetz et al in a 1996 article in the Journal of the American Dietetic Association, and La Vecchia et al in a 1998 article in the European Journal of Cancer Prevention. Studies on laboratory animals also implicate omega-6 oils and saturated fat intake as cancer promoters. (Examples include Hursting et al in a 1990 Preventive Medicine article, Zhao et al in Nutrition and Cancer in 1991, Fay et al in Cancer Research in 1997.)
The Times article draws heavily on the many recent cohort and case-control studies that have been something of a fly in the ointment. They have created confusion and doubt, because with a few exceptions, they have not confirmed the findings of the epidemiological studies.
Case control studies have shown an association with animal fat consumption and cancer, but this was not considered convincing evidence, as patients with cancer have a tendency to exaggerate their prior fat intake on diet recall questionnaires.
The cohort studies are more respected because they follow separate groups over a long time period. The cohort studies have not shown a clear-cut relationship between dietary fat (even saturated fat) and cancers of the breast, prostate, and colon or have only shown a moderate relationship. The Nurses Health Study showed that American women who reduced their fat intake did not see a decreased incidence of breast cancer.
Why do the epidemiologic and cohort studies show different results? Do these conflicting results mean that saturated fat is not a significant risk factor for cancer? Is a high-fiber diet that includes large amounts of natural, unrefined plant foods such as fresh fruit, raw nuts and seeds, vegetables, and beans not protective? Is this huge amount of data collected in the China Project and other convincing epidemiological studies wrong?
Cohort Studies: Measuring Too Little, Too Late
Because all the epidemiological studies can't be wrong, there are two possibilities. The first is that these cohort studies followed adults who are past the age when diet plays a significant role. The middle-aged adults who attempted to eat more carefully to prevent cancer were already past the age when diet has its most powerful effect. In China, for example, the dietary pattern observed was present during gestation, infancy, childhood, and beyond.
The second possibility is that the lower ranges of saturated fat intake tested were not sufficiently low to be protective. The dietary variation from one group to another may not have been enough to show a significant difference. (For instance, people eating lots of pasta and chicken, but not lots of leafy green vegetables, beans, nuts, etc. would be considered to be eating a low fat diet, but they would not be eating the diet that the evidence suggests would be optimal for cancer prevention. As I explain elsewhere, they key to a healthy diet is nutrient density.)
The bottom line is that these studies on adults in Western countries are not very accurate. They follow adults who made only modest dietary changes later in life, and who were likely past the age when dietary influence can have a profound effect on cancer occurrence.
Childhood Diet is Key
The piece of the puzzle that The New York Times missed is the research relating to diets during childhood. There is a mountain of evidence that suggests diet among children, especially very young children, can have a strong effect on the likelihood of later cancer. A lot of the reason why I wrote the book Disease-Proof Your Child was to put that research together all in one place.
When you examine what children eat and its effect on cancer, epidemiologic, case control, and cohort studies all fall much more in line with the idea that a healthy diet can reduce the likelihood of cancer.
This makes some sense: the growing body, with its dividing cells, is at greater risk when exposed to all types of negative and toxic influences. With loosely spooled DNA, children are literally more exposed.
Researchers have noted many examples of events in youth that cause cancer decades later. For example, the largest groupings of cancer spikes among those who survived the Hiroshima and Nagasaki bombings occurred 40 to 45 years after exposure (according to Pierce et al in a Radiation Research article published in 1996.)
Also in Japan, a forty-one-year National Nutrition Survey detected a massive drop in dietary fiber after World War II. An average intake of 27.4 grams per day in 1947 was reduced to 15.8 grams by 1963. Fat intake increased form 18 grams in 1950 to 56.6 grams in 1987. Twenty-three to 24 years after the heightened consumption of animal products began, there was a correlating increase in colon cancer. Those with the highest consumption of plant fiber in childhood had the lowest incidence of colon cancer.
In a 1998 study published in the European Journal of Cancer Prevention, Caygill et al found that incidence of breast cancer correlated strongly with body weight---several decades before the cancer occurred. Those who were overweight as young women were more likely to get breast cancer. Some researchers conclude that dieting later in life may be too late.
Dr. Jerald Silverman of the Comprehensive Cancer Center at Ohio State University used a grant from the American Institute for Cancer Research to study mice who are very susceptible to breast cancer. He fed some a high-fat diet. Other groups he switched to low-fat diets at various stages of their lives. The study showed the same things we see in human studies: those mice fed the high-fat diet had more cancer, and more of the cancer spread to the lung. The earlier the change to the healthier lower-fat diet, the better the mice fared.
Breast cancer is not the only cancer that has been shown to behave in this way. Colon cancer has a weak association with obesity in adults. But high body weight in adolescence correlates much more strongly with eventual colon cancer, according to Must et al in their follow up to the Harvard Growth Study, which was published in the New England Journal of Medicine in 1992.
Engeland et al tracked 1.1 million Norwegian women, and found that women who were heavier and taller as youngsters were 56 percent more likely to develop ovarian cancer. Their results were published in the Journal of the National Cancer Institute in 2003. Similarly, Harvard Medical School researchers have found that women who reported being overweight by age eighteen were twice as likely to be diagnosed with ovarian cancer later in life.
Numerous studies report links between high caloric intake in childhood and cancer. Frankel et al reported in the British Medical Journal in 1998 that they studied 3,834 subjects for more than half a century, and found a positive association between calories consumed during early life and later mortality from every cancer other than those related to smoking. With every additional MJ/day (238 calories) there was a 20% increased risk of mortality from the most common cancers. Fewer calories consumed during childhood provided protection against all three common cancers. The researchers noted that their findings were consistent with similar animal studies and human studies showing correlation between height (which can be at least partly caused by high caloric intake, especially from high-growth foods like dairy) and cancer, and concluded that their study "confirms the importance of optimal nutrition in childhood."
Just as there are studies showing that diet during childhood can increase the risk of cancer, there are other studies demonstrating that certain foods can reduce that risk. For instance, Sullivan published research in Family Practice News showing that teenagers who eat more high-fiber, high-antioxidant foods such as fruits, vegetables, and nuts have a lower occurrence of benign breast disease, the precursor marker of breast cancer.
In countless studies, early puberty has been repeatedly associated with breast cancer. But what causes early puberty? Lots of research implicates childhood diet.
In 1998, the UK Department of Health's Working Group on Diet and Cancer of the Committee on Medical Aspects of Food and Nutrition Policy made public their cohort study showing that higher consumption of produce and protein-rich plant foods such as beans and nuts is associated with a later menarche, and the higher consumption of protein-rich animal foods--meat and dairy--is associated with an earlier menarche and increased occurrence of adult breast cancer.
Similarly, a 1999 study published by Berkey et al in the American Journal of Epidemiology followed children from birth and found that the girls who consumed more animal products and fewer vegetables between the ages of one and eight were prone to early maturation and puberty. The strongest predictor of early puberty was a diet rich in animal protein before the age of five. Many studies have shown convincingly that estrogen levels in children can be managed through diet.
Cho et al followed 100,000 women between the ages of 26 and 46, and found that the younger the woman was, the greater effect diet could have on later breast cancer incidence. Those results were published in the Journal of the National Cancer Institute in 2003.
These are just some of the ways that a healthy diet, especially for young people, can reduce cancer risk. As I describe in Disease-Proof Your Child, an excellent diet can also reduce exposure to pesticides and hormones--which have been shown in studies to have links to cancer. The book also explains how a diet rich in the healthiest foods can have a dramatic effect in reducing asthma, ear infections, allergies, cardiovascular disease, and autoimmune disorders.