Mixed Age Groups
False correlations can be shown from the design of a study. For instance put together young and old individuals in the same study. Compare atherosclerosis in vessels with blood cholesterol levels.
The statement by members of the National Cholesterol Education Campaign was that the “Severity and frequency of raised plaques in the aorta and coronary arteries are strongly correlated with blood cholesterol levels”.
The above paragraph is a good example of a false conclusion drawn from a faulty correlation.
Young individuals with their lower cholesterol numbers and lesser atherosclerosis are mixed with older individuals with their higher blood cholesterol numbers and greater amounts of sclerosis. The results will seem to indicate that high blood cholesterol produces higher atherosclerosis.
The young individuals will be represented by data in the lower left part of the graft, and the older individuals will be represented by data in the upper right part of the graft.
The question that needs to be asked is does a 60 year old with high cholesterol have more sclerosis than a 60 year old with low cholesterol levels.
If the answer to this question is that higher cholesterol levels are always accompanied with a greater amount of atherosclerosis, then the diet-heart proposition would be true.
But as noted above in study after study, there was no correlation between blood cholesterol levels and degree of atherosclerosis. A man with a given blood cholesterol level may have low or high atherosclerosis.
Studies done by Landes and Sperry in New York, Paterson in Canada, and Mathur in India have already been referred to. A link below will allow you to review their data, and how saturated fat and cholesterol came to be viewed as the cause of heart disease.
Similar studies were done in Poland, Guatemala, and in the US. There was no correlation between the level of cholesterol in the blood and the amount of atherosclerosis in the vessels of autopsied individuals. These data have been ignored by the saturated fat-cholesterol in the diet heart disease proponents.
Normal and Genetic Metabolism Errors
If individuals with familial hypercholesterolemia, which have high cholesterol levels, usually over 350, and have a large amount of atherosclerosis, are included in studies with normal individuals with normal cholesterol numbers and lesser atherosclerosis, the results will be skewed.
In such a graph there may be a slight correlation between high blood cholesterol and CHD (coronary heart disease) death. But this correlation will disappear if those presumed to be afflicted with this relatively rare disease are removed from the data.
In a study of 50 middle aged men who died with CHD, there was a slight correlation between blood cholesterol and degree of sclerosis of the inner surface of coronary vessels.
• Blood cholesterol ranged from about 160 to almost 400 mg/dl.
• Percent sclerosis varied from about 6% to 85%.
• There was a slight coefficient of correlation of 0.29 between cholesterol level and atherosclerosis.
If those with levels of blood cholesterol 350 and higher, presumed to be those afflicted with familial hypercholesterolemia, were removed from the data, the slight coefficient of correlation of 0.29 disappeared.
Unfortunately a lot of studies referred to by the diet-heart proponents were of the mixed data type according to Uffe Ravnskov, M.D., Ph.D., a practicing physician and researcher, and author of The Cholesterol Myths. Only a close study of the original data revealed such discrepancies or poor research design.
From my vantage point, it seems as if research was done, and data presented, to shield actual results from the uninitiated consumer.
This practice has also been done with research on MSG (monosodium glutamate) and with chemical sweeteeners.
False correlations are dealt with in the Seven Countries Study.
Click here to review Landes and Sperry and others who found no correlation with cholesterol levels and atherosclerosis. A new window will open to help you get back to where you were.