The Cholesterol Myths by Uffe Ravnskov, M.D., Ph.D.

3. The diet has little to do with your blood cholesterol level

En français

An excerpt from my previous book The Cholesterol Myths (available for free on the web or as an e-book either from Smashwords or from DropBox). A link to the next section is available at the end

A reduction of animal fat and an increase of vegetable fat in the diet is said to lower the blood cholesterol. This is correct, but the effect of such dietary changes is very small. Ramsay and Jackson (37) reviewed 16 trials using diet as intervention. They concluded that the so-called step-I diet, which is similar to the dietary advices that are given nationwise by the health authorities in many countries, lower the serum cholesterol by 0 to 4% only. There are more effective diets, but they are unpalatable to most People.

Studies of African tribes have shown that intakes of enormous amounts of animal fat not necessarily raises blood cholesterol; on the contrary it may be very low. Samburu people, for instance, eat about a pound of meat and drink almost two gallons of raw milk each day during most of the year. Milk from the African Zebu cattle is much fatter than cow's milk, which means that the Samburus consume more than twice the amount of animal fat than the average American, and yet their cholesterol is much lower, about 170 mg/dl (38).

According to the view of the Masai people in Kenya, vegetables and fibers are food for cows. They themselves drink half a gallon of Zebu milk each day and their parties are sheer orgies of meat. On such occasions several pounds of meat per person is not unusual. In spite of that the cholesterol of the Masai tribesmen is among the lowest ever measured in the world, about fifty percent of the value of the average American (39).

Shepherds in Somalia eat almost nothing but milk from their camels. About a gallon and a half a day is normal, which amounts to almost one pound of butter fat, because camel's milk is much fatter than cow's milk. But although more than sixty percent of their energy consumption comes from animal fat, their mean cholesterol is only about 150 mg/dl, far lower than in most Western people (40).

Proponents of the diet-heart idea say that these African tribesmen are accustomed to their diet and that their organisms have inherited a cleverness to metabolize cholesterol. However, a study of Masai people who had lived for a long time in the Nairobi metropolis showed this to be wrong (41). If the low cholesterol of the Masai tribesmen was inherited it should have been even lower in Nairobi, because here their diet with all certainty included less animal fat than the diet of the Masai tribesmen. But the mean cholesterol level in twenty six males in Nairobi was twenty-five percent higher than that of their cattle-breeding colleagues in the countryside.

And there is more evidence. Although it is possible to change blood cholesterol a little in laboratory experiments and clinical trials by dieting, it is impossible to find any relationship between the make up of the diet and the blood cholesterol of individuals who are not participating in a medical experiment. In other words, individuals who live as usual and eat their food without listening to doctors or dieticians show no connection between what they eat and the level of their blood cholesterol.

If the diet-heart idea were correct individuals who eat great amounts of animal fat would have higher cholesterol than those who eat small amounts; and individuals who eat small amounts of vegetable fat should have higher cholesterol than those who eat great amounts. If not, there is no reason to meddle with people's diet.

In the early 1950's the Framingham study included dietary analyses. Almost one thousand individuals were questioned in detail about their eating habits. No connection was found between the composition of the food and the cholesterol level of the blood. Wrote Drs. William Kannel and Tavia Gordon, authors of the report: ”These findings suggest a cautionary note with respect to hypotheses relating diet to serum cholesterol levels. There is a considerable range of serum cholesterol levels within the Framingham Study Group. Something explains this inter-individual variation, but it is not diet.” For unknown reasons, their results were never published. The manuscript is still lying in a basement in Washington.

In a small American town called Tecumseh, Michigan a similar study was performed by a team of researchers from the University of Michigan headed by Dr. Allen Nichols (42). Experienced dieticians asked in great detail more than two thousand individuals what they had eaten during a twenty-four hour period. The dieticians also asked about the ingredients of the food, analysed the recipies of home-cooked dishes, and exerted great care to find out what kind of fat was used in the kitchen. Calculations were then performed using an elaborate list of the composition of almost 3000 American food items. Finally the participants were divided into three groups, a high, a middle, and a low level group, according to their blood cholesterol.

No difference was found between the amounts of any food item in the three groups; of special interest was that those with a low blood cholesterol ate just as much saturated fat as did those with a high cholesterol.

These studies concerned adults, but no association has been found in children either. At the famous Mayo Clinic in Rochester, Minnesota, for instance, Dr. William Weidman and his team analyzed the diet of about one hundred school children (43). Great differences were found between the amount of various food items eaten by these children, and also great differences between their blood cholesterol values, but there wasn´t the slightest connection between the two. The children who ate lots of animal fat had just as much or just as little cholesterol in their blood as the children who ate very little animal fat . A similar investigation of 185 children was performed in New Orleans with the same result (44).

Is it really wise to meddle with people's dietary habits if their food has no influence on their cholesterol? And how do those who believe that fat food is dangerous explain all these negative results?

The most common objection says that information about dietary habits is inaccurate, and it is. But even if it is uncertain what people say they ate yesterday, a crude relationship should appear if a sufficiently large number of individuals were questioned meticulously. If not, the influence of the diet, if any, is so minute that it cannot possibly have any importance.

Diet-heart supporters also argue that most people in Western communities already eat great amounts of fat and cholesterol. This argument declares that we have already crossed a threshold of too much animal fat in the diet so that more fat does not make any impact on our blood cholesterol.

The argument is in conflict with the studies I have mentioned above. For instance, astonished by their negative results Dr. Nichols and his team from Michigan (42) tried to find explanations. But they did not find that all individuals ate much fat. Wrote the authors: ”The distribution of daily intake of total fat, saturated fat, and cholesterol by the individuals in this study was quite broad”.

Consider now that it is the goal of the National Cholesterol Education Program to lower the intake of animal fat of all Americans to about ten per cent of their caloric intake. Almost fifteen per cent of the Tecumseh participants (42) already ate that little animal fat, and yet it was impossible to see a difference between the cholesterol of those who ate that little and of those who ate much more. Does it make sense to recommend this drastic reduction of animal fat intake if the cholesterol of those who already eat that little is just as high as the cholesterol of the others?

In the study from the Mayo Clinic (43) there was also a wide range of fat intake. The lowest intake of animal fat was 15 grams per day (less than 10 per cent of the caloric intake); the highest was 60 grams per day. In the Bogalusa study, the range was still broader. The lowest intake of all fats (no information was given about the range of intake of animal fat) was 17 grams per day, the highest 325 grams per day.

In Jerusalem a team of researchers, led by Dr. Harold Kahn studied the diet and blood cholesterol of ten thousand male Israeli civil servants. The dietary habits varied considerably between people coming from Israel, Eastern Europe, Central Europe, Southern Europe, Asia and Africa. The intake of animal fat varied from ten grams up to two hundred grams daily, and there were also considerable differences between their cholesterol values (47).

If the intake of animal fat were of major importance for the cholesterol level in the blood it should be possible to find some kind of relationship from a study of so many individuals with such great variations in blood cholesterol and dietary habits. But there was no relation in this Israeli study either. Extremely low cholesterol values were seen both in those who ate little and in those who ate the most animal fat, and high cholesterol values were seen at all levels of animal fat intake.

The scientists from Israel also studied the value of various ways of dietary questioning. Many studies have recorded the diet of a 24 hour period only. Even if this information were accurate it may not be representative of the diet for the rest of the year, far less for a whole life time. The Israeli scientists found that the best information came from a questioning over several days in different seasons of the year, the method used in the study of the bank staff members. Using this expensive and time-consuming method in a smaller study of sixty-two individuals they could not find a correlation either; the correlation coefficient between animal fat intake and blood cholesterol was zero point zero (48).

Some of my scientific papers about this and similar issues:

The questionable role of saturated and polyunsaturated fatty acids in cardiovascular disease. J Clin Epidemiol 1998;51:443-460.   Read also a dissent to the paper: Golomb BA. Dietary fats and heart disease-dogma challenged? and my answer; same journal and same issue. This paper won the Skrabanek award 1999
A hypothesis out-of-date: The diet-heart idea. Published in Journal of Clinical Epidemiology (2002 Nov;55:1057-63. Same issue: Dissent by W.S.Weintraub, and Reply by U. Ravnskov An evaluation of our discussion is available 
Is atherosclerosis caused by high cholesterol?
published in Quarterly Journal of Medicine (2002; 95:397–403)
High cholesterol may protect against infections and atherosclerosis published in Quarterly Journal of Medicine (2003;96:927-34). 
Should medical science ignore the past? BMJ 2008;337:a1681

My popular-scientific books, where you can read much more:

A shortened, simplified
 and updated version of 
my first book 
The Cholesterol Myths
Here you can read about 
how scientists and editors 
of scientific journals have 
deceived a whole world,
unintentionally or on purpose.
Also available as a 
Kindle version



About the author

to section 4

© Uffe Ravnskov