Michigan Today . . . Fall 1996

We all know the commercial slogan that has entered the popular culture: When such-and-such an expert speaks, people listen. But if some manufacturers had their way, you wouldn't get a chance to listen when Dr. Walter C. Willett speaks about health and the American diet.

Perhaps it's because he doesn't mince his words despite the calm and earnest manner in which he serves them up: "Some major food manufacturers attempt to control the content of research conferences. More than one has told academic organizers that they'll withdraw funding if I'm a participant."

Willett assumes the opposition is in response to his advice to the public to reduce or avoid consumption of trans-fatty acids (present in hydrogenated cooking oils used to cook fast food and in baked goods, dairy substitutes, snack foods, margarine and shortening), red meat and olestra (a no-calorie, polyester fat substitute made of fatty acids and sugar).

Manufacturers of processed foods---especially of fattening snacks---are eager to use olestra as a "nonfat ingredient" because it slides through the body unabsorbed. The problem is, Willett said in an interview in his Cambridge, Massachusetts, home, olestra flushes certain key nutrients out with it, including the carotenoids, a group of yellow-orange-red pigments that seem to protect us against cancer and heart disease. "One ounce of potato chips a day---one of those tiny bags with 15 or so chips---would contain enough olestra to drop absorption of the protective carotenoids by 60%," Willett argues. The means of carotenoid protection are unproved so far, he concedes, "but it's a big area of ongoing research."

Unilever, a Dutch corporation, also developed an olestra-like synthetic fat, Willett says, but announced that research it sponsored had indicated that the product should not be introduced into foods.

"Even if olestra can help people lose excess weight, which no research has showed," Willett adds, "isn't it strange to encourage people to gorge on junk food rather than just not eat it? Consuming too many calories is what makes a person fat, whether the source is a carbohydrate or fat. Fat intake is down in this country, but obesity is skyrocketing."

But after analyzing 150 studies of olestra, the US Food and Drug Administration approved the artificial chemical early this year, and it is expected to be marketed soon. Critics succeeded, however, in getting the FDA to require foods containing olestra to carry a label warning consumers that olestra can cause diarrhea and cramping, and that vitamins had been added to the foods to replace nutrients olestra leaches from the body.

Willett issues his pronouncements on nutrition atop a mountain of evidence. He heads the largest, longest-running and most respected study on the relationship between nutrition and human diseases. Science Watch, a publication that tracks trends in basic research, reported last June that in the entire field of medicine, Willett and his colleagues "account for three of the Top Ten papers," as ranked by average citations per year.

A 1970 graduate of the U-M Medical School, Willett, who is 50 years old, earned a PhD (1980) from Harvard's School of Public Health, where he is chairman of the Department of Nutrition, professor of epidemiology and professor of medicine at Harvard's medical school. His interest in food dates to his gardening days with his mother in Okemos, Michigan. He focused on the safe processing and storage of foods while majoring in food science at nearby Michigan State University, where his father was a researcher in the reproductive system of cattle. Studying the nutrition of residents of a Pottawatomi reservation in the Upper Peninsula when he was a U-M medical student sparked Willett's interest in nutritional diseases. "The project was headed by Dr. John Robson of the School of Public Health," he recalls, "and it influenced me to go into public health here at Harvard."

Dr. Frank Speizer had launched a long-term health study of 120,000 nurses in 1976 to look at the effect of oral contraceptive use. Willett joined the Harvard team in 1977 and began devising a study of dietary intake in the same population. ("Health care workers are highly motivated to fill out the surveys, which are conducted every two years, and they tend to be honest in their answers," he says.)

In 1980, Willett launched a massive ongoing survey of the diet and health of the nurses, who were aged 34 to 59 at the time. He has since added 52,000 male health care professionals in 1986 and 116,000 more nurses aged 25 to 42 in 1989. The size of the study is key, Willett says, "because if you want reliable data, you can't look at major disease events without looking at large numbers of people."

Data from the biannual questionnaires, plus samples and analyses of blood, urine, toenail clippings and drinking-water from subsets of the nearly 300,000 individuals being studied now fit in a shoe-box-size container of magnetic discs. "We have sort of a `virtual study' now," Willett says of his 80-person research team. "Without the progress in computerized information, we would have been awash in data by now."

Several factors affect the relationship between food and our health---evolution, the environment, genes, diet and individual life style. "We can tell that genetics is not the main factor in disease---diet, environment and life-style are," Willett says. "We've compared ethnic groups in their homelands with those who migrated to other countries and found that the migrants tend to develop the same health profile as other residents of their adopted countries."

Evolutionary factors are interwoven with our individual and cultural life-styles. "The circumstances of human life today are vastly different from the conditions under which we evolved," Willett says. "In paleolithic times, humans fashioned a diet sufficient to survive through reproductive age, not to survive cancer at age 65 and sustain health into their 80s. The ability desired back then was to be physically strong enough to hunt animals and ward off predators. The ability to store fat during times of famine was also a great biological advantage. We are designed to withstand a starvation diet, yet we now find food abundant and industry trying to get us to eat even more."

Willett upset the US Department of Agriculture's apple cart by overturning its 1992 food pyramid, a guide to daily food choices. USDA's pyramid suggested several servings daily of meat and dairy foods. Willett advocates the "Mediterranean diet" of the trim, hardy peoples of southern Italy and Greece. The Mediterranean diet includes plenty of tomatoes and other fruits and vegetables, whole grains and olive oil, with most dairy products consumed moderately in fermented form like yogurt or cheese. "The American diet is slowly moving in this direction," Willet says. "As time goes on, we'll see what impact these changes in diet have on national health.

Scientists have not yet shown many of the mechanisms involved in food's beneficial or harmful effects on our health, but Willett thinks that conclusions drawn from his study can help the public make wise dietary choices and guide policy makers on the regulation of the food industry. In response to questions, he offered the following observations and advice:

Heart disease:
The conventional wisdom is that saturated fat and "bad" cholesterol are the main risk factors, but we find that they may play only a modest role. Other risk factors are more important. By far, weight gain seems to me to be the biggest factor---and I don't mean just to the level of obesity. Even relatively pudgy or paunchy levels of weight gain in relatively inactive people appear to increase risk. Reducing saturated fat intake might reduce incidence of heart disease by 10 to 15%, but leanness would reduce the rate by 40% or more.

Fat:
Quantity of fat per se seems to make little difference in health. It's the type of fat one eats that is important. Substituting olive oil, a monounsaturated fat, for animal and hydrogenated fats would have a beneficial effect. Olive oil also contains antioxidant vitamin E, oxidation being the process of decay. Antioxidants appear to protect critical cell components from damage, although no one has yet proven the biochemical process involved.

Low-fat diets:
There is not good evidence that low-fat diets are beneficial in and of themselves. In fact, for people who are not already lean, substituting carbohydrates for fats seems to result in bad things---a change in metabolism, reduction in good cholesterol and rise in the triglycerides.

Colon cancer:
Reducing body fat seems to be one of the best ways to reduce the risk. Our cohort of male respondents measured their waists and hips, and we found that waist and hip expansion was a strong predictor for colon cancer. Calorie-intake, therefore, is key. Those who take in lots of calories need regular physical activity to control body fat and reduce the risk of cancer. There is a five-fold increase of risk if you put on body fat and don't exercise. Those who eat more chicken and fish and less red meat have lower rates. Something seems harmful in red meat beyond the fat. It could be iron, it could be the substances produced when meat is heated or it could be a combination of effects. Smoking and heavy alcohol consumption are risk factors, as is low folic acid intake. Folic acid [Vitamin B complex] supplements seem to cancel the alcohol risk.

Eggs:
Not much direct evidence of harm. They slightly increase lood-cholesterol levels, however, so it's good to minimize consumption, though there's no need to avoid them. They're a small risk compared with weight gain.

Alcohol:
It very substantially reduces heart disease risk if you have one or two drinks a day. But it also causes traffic accidents, induces over-drinking and presents a modestly greater risk of breast cancer and other cancers. Balancing risks and benefits is a complicated issue. There is no health reason for younger people who do not face risk of heart disease to drink alcohol, but if middle-aged persons do face such a risk and are not disposed to alcoholism, perhaps one or two drinks a day would be beneficial for some.

Sugar:
They are empty calories, which means that by acquiring 10 to 15% of our calories from sugar, Americans are displacing many nutrients present in beneficial foods. There seems to be evidence of a link between sugar consumption and diabetes. Sugar is not a poison, but limiting it is a good thing to do. Honey is no better.

Breast cancer:
Nearly 5,000 of the 120,000 nurses in our original study have developed breast cancer, or a bit more than 4%. That's very close to the rate among the general population---the lifetime risk is about 12%. Breast cancer accounts for about 20% of the deaths so far. Cardiovascular diseases account for another 20%, and lung cancer and other cancers are the other major causes of death in our first cohort of nurses.

Fractures:
High dairy intake as an adult does not seem to protect against fractures.

Vision:
High intake of fruits and vegetables containing carotenoids protects against cataracts. Smoking is a great risk factor---that wasn't appreciated when I was in medical school.

Prostate problems:
High consumption of animal fat and dairy products are risk factors, while there is no evidence implicating vegetable fats. The lycopeine in tomato products---especially cooked tomatoes---seems to protect against prostate cancer.

Sexual differences:
Health differences between men and women have been overplayed. Except for the differences in reproductive organs and sex hormones, important risk factors are generally similar. As a matter of fact I showed in my doctoral dissertation that smoking compromised women's health as much as it did men's. Some researchers were claiming at the time that smoking was less harmful for women.

Adult onset asthma:
There's a huge increase in cases, many thousands, but no one has found out why this is happening. Some researchers suggest that vitamin E and fish oil supplements may be protective. We've found no support for the fish oil hypothesis and are still looking at vitamin E, which may be protective.

Coffee and tea:
Our studies show coffee is not an important risk factor for heart disease, but one area of concern is for fractures. Coffee causes calcium loss in those who drink more than two or three cups a day. There is a 60 to 70% reduction in suicide among female coffee drinkers and an appreciable reduction among men, too. Perhaps that's because it's an antidepressant.

Liquid diets:
These can be a good source of vitamins, but don't have other important ingredients of natural food, plus there is no evidence that they provide any long-run benefits in weight loss. A few low-calorie diets have been fatal. We're going to look at them further.

Kidney stones:
We found no link between vitamin C and kidney stones.

Research:
The National Institutes for Health must continue to play the big role in supporting research on nutrition and public health, and fortunately the NIH has sustained funding levels in such research, as they should, because everyone has an interest in it. Biomedical research is one of the few areas where the US has leadership, and almost all of the funding comes rom NIH.

This Issue's Index   |   This Issue's Front Page   |   CURRENT Michigan Today