In 1985 an NIH Consensus Conference was able to “establish beyond any reasonable doubt the close relationship between elevated blood cholesterol levels (as measured in serum or plasma) and coronary heart disease” (JAMA 1985, 253:2080-2086).
I have been making an analogy between scientific behavior and the activities of the legal system and following that idea, the wording of the conference conclusion suggests a criminal indictment. Since the time of the NIH conference, however, data on the role of cholesterol fractions, the so-called “good (HDL)” and “bad (LDL)” cholesterols and, most recently, the apparent differences in the atherogenicity of different LDL sub-fractions would seem to have provided some reasonable doubt. What has actually happened is that the nutrition establishment, the lipophobes as Michael Pollan calls them, has extended the indictment to include dietary fat, especially saturated fat at least as accessories on the grounds that, as the Illinois Criminal Code put it “before or during the commission of an offense, and with the intent to promote or facilitate such commission, … solicits, aids, abets, agrees or attempts to aid… in the planning or commission of the offense. . . ..”
A major strategy in the indictment of saturated fat has been guilt by association. The American Heart Association (AHA), which had long recommended margarine (the major source of trans-fats), has gone all out in condemning saturated fatty acids by linking them with trans-fats. The AHA website has a truly deranged cartoon film of the evil brothers: “They’re a charming pair, Sat and Trans. But that doesn’t mean they make good friends. Read on to learn how they clog arteries and break hearts — and how to limit your time with them by avoiding the foods they’re in.”. While the risk of trans-fats is probably exaggerated — they are a small part of the diet — they have no benefit and nobody wants to defend them; dietary saturated fat, however, is a normal part of the diet, is made in your body and is less important in providing saturated fatty acids in the blood, than dietary carbohydrate. Guilt by association is a tricky business in courts of law — just having a roommate who sells marijuana can get you into a good deal of trouble — but it takes more than somebody saying that you and the perpetrator make a charming pair.
The failure of the diet-cholesterol-heart hypothesis in clinical trials as been documented by numerous scientific articles and especially in popular books that document the original scientific sources. It is unknown what the reaction of the public is to these books. However, amazingly, there is only one book I know of that takes the side of the lipophobes and that is Daniel Steinberg’s Cholesterol Wars. The Skeptics vs. the Preponderance of Evidence. A serious book with careful if slightly biased documentation and an uncommon willingness to answer the critics, it is worth reading. I will try to discuss it in detail in this and future posts. First, the title indicates a step down from criminal prosecution. “Preponderance of the evidence” is the standard for conviction in a civil court and is obviously a far weaker criterion. One has to wonder why it is that the skeptics have the preponderance of the popular publications — if the scientific evidence is there and health agencies are so determined that the public know about this, why are there so few — maybe only this one — rebutting the critics.
The Skeptics vs. the Preponderance of Evidence
In any case, what is Steinberg’s case? The indictment on page 1 is somewhat different than one would have thought.
“….the [lipid] hypothesis relates to blood lipids not dietary lipids as the putative directly causative factor. Although diet, especially dietary lipid is an important determinant of blood lipid levels, many other factors play important roles. Moreover, there is a great deal of variability in response of individuals to dietary manipulations. Thus, it is essential to distinguish between the indirect “diet-heart” connection and the direct “blood lipid — hard” connection failure to make this distinction has been a frequent source of confusion. (his italics)”
What? Are we really supposed to believe that diet is an incidental part of the lipid hypothesis? Are we supposed to believe that our cholesterol is just a question of the variability of our response to diet. Has the message really been that diet is not critical and that heart-disease is just the luck of the draw (until we start taking statins)? This is certainly the source of confusion in my mind. Of course by page 5, we are confronted with this:
“In 1966, Paul Leren published his classic five-year study of 412 patients who had had a prior myocardial infarction. He showed that substitution of polyunsaturated fat and saturated fat-rich butter-cream-venison diet favored by the Norwegians reduced their blood cholesterol by about 17 per cent and kept it down. The number of secondary current events in the treated group was reduced by about one-third and the result was significant at the p < 0.03 level.”
In a future post, I will describe Paul Leren’s classic five-year study which, by 1970, had a follow-up to eleven years and the results will turn out not to be as compelling as described by Steinberg. For the moment, it is worth considering that, given the strong message, from the AHA, from the American Diabetes Association, from the NIH Guidelines for Americans, the criterion really should be beyond a reasonable doubt. There shouldn’t be even a single failure like the Framingham Study or the Women’s Health Initiative. In fact, the preponderance of the evidence when you add them all up, isn’t there.
Richard David Feinman 