Authors: Mary Enig
We don’t know why, but this part of the Framingham study was never published.
Tecumseh: No Connection
A study similar to the Framingham study was carried out in the small town of Tecumseh, Michigan, by researchers from the University of Michigan and published in the
American Journal of Clinical Nutrition
, 1976. They asked over 2,000 people about the food they ate during a 24-hour period, as well as about the ingredients in the food.
The study subjects were then divided into three cholesterol-level groups: high, middle, and low.
The results: there was no difference between the amounts that the three groups ate of any particular food item. What’s more, the low-cholesterol group ate just as much saturated fat as did the high-cholesterol group.
British Bank Tellers: No Connection
In a report published in the
British Medical Journal
, 1963, researchers asked 99 middle-aged male bank employees in London to weigh and record everything they ate over a two-week period. Once more, the results showed no connection between what they ate and their cholesterol levels.
Just in case they had made an error, the researchers asked 76 of the tellers to do the whole procedure again for a week, at a different time of year: same results.
And finally, to be absolutely certain, the researchers reanalyzed the data from the men whose records were particularly detailed and accurate. Same result as before: no connection.
Israeli Civil Servants: No Connection
In a study published in the
Israel Journal of Medical Sciences
, 1969, researchers in Jerusalem studied the diets and blood cholesterol levels of 10,000 Israeli civil servants. These people came from many different backgrounds—Israel itself, Eastern, Southern, and Central Europe, Asia, and Africa—so their eating habits varied considerably. The amount of animal fat they ate ranged from 10 grams a day to 200, and their cholesterol levels ranged from high to low.
The researchers examined whether all the data from this extremely varied group of people indicated some kind of relationship between animal fat and blood cholesterol level, but they found nothing.
Just to make sure, the researchers did a smaller study of 62 people using a method similar to that used in the British bank tellers study: they conducted a survey over several days and at different times of year, a method that gives the most accurate results. Nevertheless, they still found absolutely no correlation between the amount of animal fat consumed and blood cholesterol levels.
High-Fat Diet—Good Cholesterol
In a study published in the
Journal of the American College of Nutrition,
2004, researchers at the State University of New York at Buffalo found that after 11 healthy adults ate a diet in which only 19 percent of calories came from fat for three weeks, their HDL or “good” cholesterol fell. Then they spent three weeks on a diet in which 50 percent of calories came from fat, and their HDL levels increased. Most important, the high-fat diet did not lead to an increase in LDL, the “bad” cholesterol.
Myth #4: Cholesterol Causes Plaque Buildup in Arteries
Another assertion generally taken as axiomatic is that high levels of cholesterol in the blood cause atherosclerosis, the buildup of the fatty plaques that obstruct arteries. Yet much research evidence indicates that high blood cholesterol has no relationship with degree of atherosclerosis.
Coconut and Cholesterol
Experts who promote the lipid hypothesis point to studies in which feeding coconut oil raised cholesterol levels in animals. However, the coconut oil used in these experiments was
fully hydrogenated,
that is, fully saturated with hydrogen, which turns all the fatty acids into saturated fats. (Partial hydrogenation creates mostly trans fatty acids, not all saturated fats.) The researchers were investigating the effects of essential fatty acid deficiency on the test animals, and full hydrogenation to eliminate all the essential fatty acids in the coconut oil. They used coconut oil because it is the only fat that can be fully hydrogenated and still be soft enough to eat.
As expected, the cholesterol levels in the animals went up, confirming other studies showing that essential fatty acid deficiency raises cholesterol levels. Yet, many commentators have pointed to this study as proof that coconut oil raises cholesterol levels, without mentioning that the researchers were not using natural coconut oil, but fully hydrogenated coconut oil that
created
the deficiency.
Other studies comparing various fats and oils found that natural coconut oil promoted the assimilation and storage of essential fatty acids, therefore
preventing
essential fatty acid deficiency. So it’s the lack of essential fatty acids that created the health risk, not the coconut itself. And in these studies, an increase in cholesterol levels was simply a marker for another condition—not the
cause
of health problems.
Canadian Veterans
A team of researchers in London, Canada, conducted a long-term study of 800 war veterans confined to a hospital, published in
Circulation
, 1963. For years, the researchers analyzed the veterans’ cholesterol levels and studied the arteries of those who died to determine the level of atherosclerosis for each veteran. They found that, although the levels of cholesterol varied considerably from one individual to another, each man’s cholesterol remained more or less the same during the entire period of the study. That is, if a veteran had low cholesterol at the beginning of the study, it was still low when he died. Yet the men who had low cholesterol had just as much atherosclerosis as those with high cholesterol.
What About the Japanese?
The Japanese have low blood cholesterol, and their risk of heart attack is much lower than that of any other nation. According to the lipid hypothesis, atherosclerosis should be rare in Japan.
Two studies sought to confirm this. In the first, published in the
American
Journal of Cardiology
, 1985, researchers from Kyushu University looked at the aorta (the main artery of the body, which carries blood out of the heart) in 659 Americans and 260 Japanese, for signs of atherosclerosis. They found very little difference between the two groups, at all age levels.
The second study, carried out at the Geriatric Hospital in Tokyo and presented at the International Symposium of Atherosclerosis, 1977, examined the arteries of the brain in 1,408 Japanese and over 5,000 Americans. In every age group, the Japanese had
more
atherosclerosis in these arteries than the Americans did.
As for the coronary arteries—the ones involved in heart attacks—the Japanese have less atherosclerosis in these arteries than do Americans, which may be why they are less prone to heart attacks.
But if high cholesterol is what causes atherosclerosis, this should occur in all types of arteries, since the amount of cholesterol in the blood is the same in all parts of the body. It seems much more likely that something other than cholesterol causes atherosclerosis. Blood pressure, for example, does vary in different arteries and at different times, for example when you are under stress. But few scientists seem willing to look at these complicated details.
Cholesterol Lowering: No Effect on Plaque Buildup
A recent study, published in the
American Journal of Cardiology
, 2003, found that patients who successfully lowered their cholesterol levels did not reduce plaque buildup in their arteries. One group took a strong dose of a cholesterol-lowering drug while the other took a lower dose of the same drug. Researchers then measured the amount of blockage in their arteries. After more than one year, both groups showed a 9.2 percent
increase
in plaque buildup, suggesting that plaque buildup is not related to cholesterol level.
If This Is All True…Why Does the Lipid Hypothesis Still Rule?
In science, the burden of proof rests on the scientist. If even one study contradicts a hypothesis, then we must at the very least be open to reevaluating the hypothesis, even if many studies support it. We’ve just described
many
studies that contradict the lipid hypothesis, yet it remains the ruling theory of heart disease in modern medicine.
Why? Consider the fact that the widespread credence given to the lipid hypothesis certainly has helped a number of industries. For example, the hypothesis provides the pharmaceutical industry with a rationale for selling cholesterol-lowering drugs, which today is a multibillion-dollar business. The hypothesis has also been a boon for the food industry, since it justifies the use of inexpensive vegetable oils in processed food products, rather than more expensive coconut oil and animal fats.
The following pages tell the story of how, due in part to the exertions of these industries, the lipid hypothesis became the basis of government policy. Going back 60 years, we’ll trace the evolution of this idea and the huge impact it has had on what Americans are eating today. And we’ll tell the tale partly from a personal perspective, for an important participant in the struggle was Dr. Mary Enig.
The Lipid Hypothesis Gets Rolling
David Kritchevsky’s 1954 studies, published in the
American Journal of Physiology,
July through September 1954, attracted the attention of scientists concerned about the emerging epidemic of heart disease because they appeared to support the lipid hypothesis. Kritchevsky’s studies demonstrated that
Kritchevsky’s work
appeared
to show that by reducing the levels of animal products in their diet, Americans could avoid heart disease—as long as scientists ignored the many studies (including those we’ve described above) that contradicted Kritchevsky’s findings. In fact, Kritchevsky’s studies had no relevance to the lipid hypothesis for two reasons. First, the kind of plaque the rabbits developed was completely different from the kind of plaque humans develop. Second, the rabbits were fed purified cholesterol that was oxidized (damaged by processing), not the kind contained in ordinary food.
The lipid hypothesis was introduced to the American people in 1956, when the American Heart Association (AHA) aired a fund-raiser on all three major television networks. The panelists presented the hypothesis as the cause of heart disease and recommended the Prudent Diet, in which corn oil, margarine, chicken, and cold cereal replaced butter, lard, beef, and eggs. It was on this television show that cardiologist Dr. Dudley White, who served as one of the panelists, made his comments objecting to the lipid hypothesis and the Prudent Diet.
“See here,” he remarked, “I began my practice as a cardiologist in 1921 and I never saw an MI patient until 1928. Back in the MI-free days before 1920, the fats were butter and lard and I think that we would all benefit from the kind of diet that we had at a time when no one had ever heard the words ‘corn oil.’”
Despite these nationally televised comments, and in spite of the numerous contradictory studies already published in the scientific literature, the lipid hypothesis had already gained enough momentum to keep it rolling. (One such contradictory study was the Anti-Coronary Club project; see sidebar, .) Both the food processing industry and the pharmaceutical industry could see the benefits of making patients out of healthy people, simply by making them afraid of cholesterol.
The following year, the food industry initiated advertising campaigns touting the health benefits of products low in fat or made with vegetable oils. A typical ad read: “Wheaties may help you live longer.” Wesson recommended its cooking oil to consumers “for your heart’s sake,” while an ad in the
Journal of the American Medical Association
described Wesson Oil as a “cholesterol depressant.” Mazola advertisements assured the public that “science finds corn oil important to your health.” Medical journal ads recommended Fleishmann’s unsalted margarine for patients with high blood pressure. Prominent physicians endorsed the use of vegetable oils as substitutes for saturated fat. Such claims continued well into the 1980s, after which they were quietly dropped when studies indicated that polyunsaturated oils contributed to cancer.
The American Medical Association at first opposed the commercialization of the lipid hypothesis, warning that “the anti-fat, anti-cholesterol fad is not just foolish and futile…it also carries some risk.” The AHA, however, was committed to supporting the lipid hypothesis. In 1961, the AHA published its first dietary guidelines aimed at the public. The authors, Irving Page, Ancel Keys, Jeremiah Stamler, and Frederick Stare, called for the substitution of polyunsaturates for saturated fat, even though Keys, Stare, and Page had all previously noted in published papers that the increase in CHD was paralleled by increasing consumption of vegetable oils. In fact, in a 1956 paper, Keys had suggested that the increasing use of hydrogenated vegetable oils might be the underlying cause of the CHD epidemic.
These ongoing concerns about hydrogenated vegetable oils prompted the inclusion in a 1968 statement by the AHA on diet and heart disease of a carefully worded disavowal, making clear that the types of fats thought to lower cholesterol were not the same ones present in hydrogenated fats. The AHA statement implied that proponents of the lipid hypothesis could not claim that hydrogenated fats were heart healthy.
However, that disclaimer was never released. Among the likely reasons were the shortening industry’s strong objections to the disclaimer and a letter from Fred Mattson, a researcher at Procter & Gamble, to Campbell Moses, medical director of the AHA, urging against its distribution, even though 150,000 copies had already been printed. The letter that Mattson wrote to Dr. Moses is part of the Congressional Record, for it was submitted as testimony during hearings held in 1977 by the McGovern Committee on Nutrition and Human Needs. The final recommendations to the public that the AHA did publish omitted the warning that partially hydrogenated (that is, trans) fats did not protect against heart disease.