The world is changing fast and to keep up you need local knowledge with global context.
Starting in the 1940s, pioneering American physiologist Ancel Keys helped establish the epidemiological link between cholesterol and cardiovascular disease. Keys was wrong. Very wrong. But instead of using the opportunity to learn, as one is more inclined to learn from what doesn’t work than what does, Keys opted to bury studies contradicting his sacred findings and continued to force his dietary theory down the throats of anyone who would listen. Literally. In Part 11, the final instalment, of Professor Tim Noakes’ series on insulin resistance, Noakes concludes the series with a rap sheet detailing Ancel Keys’ criminal misconduct. Two manifestations of the human condition stand out. The first is Keys’ single-minded insistence to continue on whatever path most likely to substantiate his chosen hypotheses. The second is the susceptibility of human beings to believe theories posited by charismatic, self-assured individuals, at the expense of solid scientific evidence to the contrary. Sound familiar? – Nadya Swart
It’s the insulin resistance, stupid: Part 11
By Prof. Timothy Noakes
The Final Act
Scientists usually achieve fame for being right. It matters not how they achieve their correctness, only that in the end their ideas prove to be correct.
Ancel Keys’ problem was that he was wrong. But his poor science was not the reason why he was wrong. He was wrong because he failed to bring the same scientific rigor to John Gofman’s diet-heart and lipid hypotheses that he had shown when conducting his starvation studies (2). It was as if he had lost interest in the conduct of hard science. Rather, it appears as if his “science” had become a method for marketing himself, his ego, and his next big idea.
Ultimately, with the Seven Countries Study (SCS) especially (3), he sought to entrap a multinational legion of compliant buddies in what became “a scientific adventure in cardiovascular disease epidemiology” (4). In time, his colleagues became his global acolytes. With their backing, it became a relatively simple matter to control the American Heart Association (AHA) and National Heart, Lung, and Blood Institute (NHLBI) — and more important, to control where these organizations would invest their research monies.
The proof of his relative disinterest in the hard science of cardiovascular epidemiology is the manner in which he acquired and misused the initial information he presented in 1953 (Figures 4 and 5 in reference 5), as well as the fact that he simply ignored any epidemiological evidence that conflicted with his hypotheses.
The strongest proof is the manner in which he ignored the quite reasonable and carefully (and respectfully) presented arguments of Jacob Yerushalmy and Herman Hilleboe (5, 6). Though he responded (7), he tellingly ignored their key argument that association does not prove causation. He never addressed that key criticism; logically, because he could never win that argument (Figures 4, 5, 7, and 9 in reference 5). He knew the criticism to be true because in 1952 (8), he had criticized Haqvin Malmros (9) for making the same “crude” error. Let’s recall (5) what Keys wrote in 1952:
The recent analysis of mortality data in Scandinavia (9) is more acceptable but still suffers from the inherent weakness of attempting to discover causation from a parallelism between crude estimates of national averages for two variables. Besides the dubious point of attributing war-time changes in vital statistics to actual changes in atherosclerosis, the analysis glosses over the fact that the presumed variations in cholesterol intake were parallel to similar variations in total dietary fat intake and, at least to some extent, to changes in calorie intake and physical activity. (8, p. 115-116, my emphasis)
Keys is simply but forthrightly making the point that associational studies cannot ever exclude all possible confounders. It is also interesting that here Keys mentions physical activity. But in the SCS, physical activity was not considered as a potential confounder in the final conclusions (10; 11, p. 13).
Instead of embracing the challenges presented by other members of the scientific community better to develop his own hypotheses, Keys simply ignored them. Later he became combative and dismissive (12, 13) of the counterarguments of John Yudkin (14,15), Raymond Reiser (16), and George Mann (17, 18). He used classic colonial paternalism to dismiss Mann’s evidence that despite a diet of almost exclusively milk and meat, the Maasai tribespeople of Tanzania have low blood cholesterol concentrations (17) and large coronary arteries with marked intimal thickening (18), all without complicated atherosclerotic lesions: “The peculiarities of those primitive nomads have no relevance to the diet-cholesterol-CHD (coronary heart disease) relationships in other populations,” he claimed (19, p. 26; 20, p. 153).
Keys could well afford to disregard the criticism he received from his peers because he had the complete confidence and support of the AHA and NHLBI, which were both equally dismissive of any challenges to their hegemony (22-24). The diet-heart mafia had usurped complete control, as I described earlier (25).
And right at the beginning, Keys had dismissed Gofman and colleagues’ description of the lipoproteins in blood (26) as “giant molecules” that added nothing to the understanding of coronary heart disease beyond that already provided by the blood cholesterol concentration (27, 28).
Yet, when Keys came to develop his twin hypotheses, neatly stolen from Gofman, he incorporated Gofman’s novel findings (5).
The point is, such practices do not make for good science. Instead, in this example, it has produced great harm.
Ultimately, Keys and his acolytes must be held accountable for the wretchedness their over-promotion of his wrong ideas has wrought on the health of so many people.
As Mann wrote, “Saturated fat and cholesterol in the diet are not the cause of CHD. That myth is the greatest scientific deception of this century, perhaps of any century. One day the true cause of CHD will be found and the disorder will be prevented. When that happens Keys and his cronies will be exposed” (1, p. 212).
The Rap Sheet
Ancel Keys, you are charged with the following. How do you answer to these charges?
The prosecution speaks:
Dr. Keys, it is our charge that the twin hypotheses you developed in 1953 were not your original ideas. Rather you “borrowed” both from Dr. John Gofman and his colleagues. The moment you came across their work (26), you immediately realized that, already in 1950, Gofman and his team had developed an eminently testable hypothesis to explain why the incidence of CHD had apparently reached epidemic proportions in the U.S. and elsewhere.
You also knew that, at that time, there was no “hotter” research topic, nor one with more readily accessible research funding than that involving the causes and prevention of CHD. You understood that when President Harry S. Truman signed the National Heart Act into law in 1948, he effectively “opened the funding floodgates for researchers studying the diagnosis, treatment and prevention of heart disease” (10, p. 220). For you, this was a most opportune development since you knew that to advance your academic career after the ending of World War II, you were in urgent need of a new and more promising research direction.
Searching for a change in your research direction, you correctly concluded this was the one you should be following.
But you also knew that as a physiologist with no training in medicine or cardiology or epidemiology, your chances of becoming part of that elite heart disease research establishment were extremely remote. So you needed to do something unusual and urgent to grab attention, particularly that of the select group of U.S. scientists controlling the distribution of those research funds.
Realizing others like Gofman were better qualified than you and that they already held all the aces, you needed to find a way to bypass the usual scientific process — a process by which, as you knew, academic recognition and acceptance occur only after decades of concentrated research and numerous publications. But you had neither the money nor the luxury of time to follow that rigorous research path, since in 1950, Gofman and his team had already laid down the marker. They were clearly the prime contenders to lead this research field in the U.S. and perhaps globally. You, on the other hand, in 1950 had zero qualifications to be considered a serious player in any aspect of heart disease research.
As an aside, your inability even to measure dietary intakes accurately in your research subjects would be exposed as a significant, indeed a perhaps fatal flaw in the majority of your research studies, most especially the SCS (29, 30). In that study, it appears you had very little real interest in carefully documenting what the participants in different countries were actually eating. The clear proof is that you studied fewer than 500 subjects, representing fewer than 4% in a test population of more than 12,700. But wasn’t the goal of that study to establish the dietary contribution to CHD?
Surely it does not take a lawyer to tell you that successful science requires that you meticulously collect and record the fundamental data on which your scientific conclusions will be based. If you do not, then your results will be meaningless.
By contrast, the recently reported Prospective Urban Rural Epidemiology (PURE) Study was able to accurately measure the diets of 135,335 subjects (100%) in 18 countries (30, 31).
I can only conclude that your primary goal was not the conduct of hard science.
This leads to the prosecution’s claim, developed subsequently, that you were not driven by any passion to discover what the subjects in your studies were actually eating. Instead you “knew” what they had to be eating according to whether or not they came from populations with high or low rates of CHD.
So, it seems you showed what T. C. Chamberlin (32) warned can happen so easily: “The moment one has offered an original explanation for a phenomenon which seems satisfactory, that moment affection for his intellectual child springs into existence.” Then begins a process in which there is “pressing of the theory to fit the facts and pressing of the facts to make them fit the theory” (32, p. 569).
But let’s return to the threat Gofman posed to you and your ambitions in 1950 and 1951.
It could not have escaped your attention that Gofman and his colleagues had laboratory skills you could never hope to match. Your cold genius was your understanding that to gain ascendancy in this field, you had to change the direction of the discourse. But what could you possibly do with your significant absence of research qualifications or experience in this discipline?
The first glimmer of a possible solution came to you when you read the work of Malmros (9). It was your first lightbulb moment, for you realized epidemiological associational research (for which in 1951 you needed no specific qualifications or detailed training) would be the way for you to go. You would take the research out of the laboratory where Gofman ruled to the much weaker and still evolving “science” of associational nutritional epidemiology. On the way, you would usurp Gofman’s twin hypotheses and subtly undermine the value of what he was attempting (27, 28).
Then, fortuitously, you found your answer in the 1953 publication of the World Health Organization (WHO) Annual Epidemics and Vital Statistics, which listed both mortality rates from heart disease and dietary fat intakes in 22 countries (Figure 5 in reference 5). You knew immediately that this was your golden chance.You grabbed it with both hands, and the rest, as they say, is history. This history was one that you alone would initiate and, with the help of your influential acolytes in the AHA, the National Institutes of Health (NIH), and NHLBI, would drive for the rest of your life (30).
You decided to present that information, conveniently gathered as part of your “armchair research” (all the while avoiding the rigors of the usual scientific process to which Gofman and his colleagues were committed), as if it provided definitive proof of your hypotheses. You backed yourself, believing that with your particular brand of “charm” and persuasive skills, you would be able, in time, to convince the key opinion leaders to believe you and only you. You hoped they would accept, without question or skepticism, that your conclusions, based on such simple associational data, definitively proved causation. In the words of one internationally respected cardiologist (33), “The 1957 Seven Countries Study by Ancel Keys … established unequivocally the pathophysiological role of dietary saturated fats acting through serum cholesterol concentrations in the causation of atherosclerotic vascular disease” (33, p. 1233).
History has proved you were correct in your presumption. Congratulations. It shows you have an exceptional understanding of human nature.
However, instead of presenting all the information available to you, you selectively and willfully misrepresented only the most favorable data (Figure 4 in reference 5) from all the evidence (Figures 5 and 7 in reference 5) available to you at that time. Your exclusion of critical information, as your lifelong opponent and somewhat bitter enemy Dr. George Mann has charged, is a form of cheating (11, p. 6). As our rap sheet will reveal, cheats do not just cheat once. They tend to cheat serially.
Over the following five decades, you used your special appreciation of human nature to advance your hypotheses, to the exclusion of all other competing ideas.
You became, as investigative journalist Nina Teicholz has written with feeling, the “Greatest Man in the History of Nutrition” (29, p. 46) — by far.
We now move to the more specific charges:
1. When Yerushalmy and Hilleboe communicated the flaws in your 1953 paper, both at a scientific meeting and later in a scientific publication (6), they showed many factors other than dietary fat might explain the relationship you were claiming (Figure 7 in reference 5) and that, in any case, associational studies cannot prove causation. You simply ignored their argument (7), even though you (8) had used the same one to point out the flaws you found in Malmros’ thinking in 1950 (9). While you would sometimes thereafter pay dutiful lip service to the truth that associational studies cannot prove causation — for example, in the introduction to your best-selling book (21, p. vi) — you always implied your associational studies were different (20).
But you knew, as Yerushalmy and Hilleboe had warned, that if you repeated your story with sufficient frequency and enough conviction, and you distributed it through as many different communication outlets as possible, eventually the key players would believe you. When you appeared as the key person in two prominent cover stories in the U.S. in 1956 and 1961, you knew you and your ideas had conquered the world.
In the end, you believed your own hubris. On Dec. 12, 1961, you told the reporter from The New York Times you were convinced saturated fat in the diet caused coronary heart disease (34, p. 199).
2. That is the clear message you took to the world with the publication of your popular book, first released in 1958 (21). The book included the scaremongering table included in a previous column as “Table 1: Heart attack incidence at different blood cholesterol concentrations as reported by Ancel Keys” (25). That table predicted a 400% increase in heart attack rates in persons with blood cholesterol concentrations greater than 260 mg/dL (6.7 mmol/L).
Yet these data came from studies such as the Framingham Heart Study (FHS), which was in its infancy and would in time prove that prediction of a 400% increase to be completely misleading (Figures 1, 3, and 4 in reference 35). You again failed to warn your readers, now numbering in the millions around the world, that those data were from simple associational studies that could not exclude the possibility that other factors were just as likely to explain the association.
For example, persons with blood cholesterol concentrations >260 mg/dL may be at increased risk of heart attack for many reasons other than their high blood cholesterol concentrations. Perhaps they are male, they smoke, have high blood pressure, are older, and do little physical activity, for example. Those “risk factors” are the ones research, including that from the FHS, has now clearly established truly increase risk of CHD (36).
The problem was that your book launched global lipophobia, a fear of fat (34), without informing the reading public that your scaremongering was based solely on unproven hypotheses, not on proven scientific facts.
There is a certain arrogance, Dr. Keys, in the manner with which you presented such dire warnings based on the “inadequacy of insufficient evidence” (19, pp. 22-41).
Since you are not a medical doctor, you perhaps failed to understand one of the four cornerstones of ethical medical practice. It is this: First, do no harm.
3. We have mentioned earlier that cheats never cheat just once. Your conduct in failing to report the complete (adverse) findings of the Minnesota Coronary Experiment (MCE) is perhaps the most egregious indictment of your scientific malfeasance (30).
We refer to the failure of your team, of which you were the principal investigator, to publish the results of the MCE timeously when they became available in 1975 and were first presented, incompletely, at a medical congress (37, 38). The incomplete data were then published for the first time in 1989 (39), but that publication failed to include material information that finally became available only in 2016 (40), and that information only came to light because it was “discovered” by Dr. Ivan Frantz’s son (41), who immediately realized it might have some historical value. He was correct.
The 2016 publication (40) established that the substitution of the omega-6 polyunsaturated fat linoleic acid for dietary saturated fat did indeed reduce blood cholesterol concentrations as your research predicted it would (42).
But not only did this reduction in blood cholesterol concentrations not reduce CHD mortality (as your hypothesis predicted it must); it increased heart attack risk by 22% for every 30 mg/100 mL (0.78 mmol/L) reduction in blood cholesterol concentrations. In addition, this dietary substitution worsened survival in those over 65 years of age (Figure 10 in reference 30).
In other words, your meticulously designed and conducted study (and here the prosecution must compliment you for undertaking one of the very best dietary intervention trials in history) proved lowering blood cholesterol concentrations by doing exactly what you, in your book (21, pp. 7-8) and your Time magazine interview (43), had said the global public should be doing actually increased heart attack risk instead — the opposite of your claim.
So the outcome was that those millions around the world who read your book or the Time magazine article and who, as a result, followed your advice were unknowingly harming their health by increasing the probability that they would develop CHD and have a shortened life expectancy.
Given your global status as “Mr. Cholesterol” (43), your responsibility at that time was to announce to the world that your ideas were wrong and potentially harmful. You failed to do that. Let me remind you, Dr. Keys, that this is another example of negative evidence, an event that should have happened but did not, much like the failure of the dog to bark in the illuminating Sherlock Holmes story (44).
You allowed the true results of that study to go unpublished from 1975 until your death in 2004, a 29-year period during which the low-fat diet became institutionalized around the world as the only diet that could prevent heart disease. Significantly, your research instead had proven that it could not and showed your dietary recommendations were dangerous.
Ultimately, your deception was hidden for another 12 years before finally being exposed in 2016 (40).
Instead of revealing all to the world when your own carefully designed and executed study disproved your hypotheses, you allowed the results of the study to go unpublished and unannounced. You buried the study (30).
You did this for only one reason: to protect your academic legacy. Your ego would not allow you to be truthful.
And in acting in this way, you sold the health of the world down the river (45) into a form of dietary slavery that has devastated global health and had unimaginable consequences.
That, Dr. Keys, is your ultimate legacy.
4. The Seven Countries Study (SCS) was specifically designed to “prove” that the association between dietary fat intake and heart disease rates you had extracted from the WHO data (Figure 4 in reference 5) was indeed true when accurately measured in representative samples of individuals living in those seven countries. Yet there were a number of reasons why that study could never provide the definitive evidence for which you were searching. The prosecution’s argument is that, throughout your life, you failed to draw attention to those flaws.
First, the SCS was fatally flawed from the outset because the choice of countries was not random (46; Figure 9 in reference 5). Because of this, it cannot be assumed the results were not influenced by residual confounders in the non-randomly selected countries.
Even if, as K. D. Pett and colleagues (47) have argued, your selection of those seven countries was done for purely practical reasons and not to produce a predetermined outcome, the fact remains: The selection was not random. And this matters, for it essentially negates any interpretations of the findings, other than that they are almost certainly confounded.
Second, you made no allowance for differences in habitual levels of physical activity between the inhabitants of the different countries. Let’s recall Mann’s criticism (11): “The most sensible explanation for the national differences in CHD that they observed (in the SCS) was that exercise protects.” Mann also claims the table that might have shown this protective quality of exercise “mysteriously disappeared from the final publication (of the SCS results)” (11, p. 13; Table 1 in reference 30). To where did that table disappear, Dr. Keys, and why?
Third, the dietary analyses you conducted were hopelessly inadequate, as they likely were in all your previous studies (Figure 10 in reference 5). As we have asked repeatedly: How could you possibly believe an analysis of less than 4% of all the subjects in the SCS would produce information from which you could draw any coherent and generalizable conclusions?
In fact, you knew this was not possible; your subsequent actions confirm your guilt, for you actively chose to hide that failing — and others — by publishing the work in an obscure medical journal, subsequently released in book format (48, 49), which you knew would be read by very few. How do you answer to Teicholz’s conclusion:
Keys did not seem eager to report on his dietary data at all, and, indeed, I had trouble tracking down some of it. He published most of the data in a Dutch journal, Voeding, where he knew it would go unnoticed, not in one of the mainstream British or American publications where he published most of his other Seven Countries papers. (29, p. 41)
Indeed, even in those obscure publications, you were forced to acknowledge the uncertainties of the data produced by your woeful research methods:
The difficulties and limitations of estimating nutrient intakes from dietary survey data are critically examined below in the article “Dietary Survey Methods in Studies on Cardiovascular Epidemiology.” In view of these limitations, and those mentioned above, it would be improper to claim that the survey data reported here are highly accurate representations of the true long-time diets of men in these areas. But, as will be seen, every effort was made to obviate controllable sources of error. Though the data on individuals have serious limitations, it is believed that the general picture is substantially correct for each of the samples, and for the differences among the samples. (49, p. 7, my emphasis)
Actually no, Dr Keys. The goal of science is not to provide evidence that will buttress your personal beliefs. It is to provide objective data. Clearly you had not collected sufficient objective data to support your biases, so you simply hid the data and asked the world to accept that your conclusions were always correct.
The point is that if this is how you believe science should be conducted, then why did you bother to conduct the SCS in the first place? Why did you not simply tell us what we needed to believe? It would have saved a great deal of money and effort.
The prosecution believes this is just another rather blatant example of how the main focus of your research was always to provide “proofs” that could be used to market your hypotheses — and most certainly not to disprove them.
I am reminded of the old joke about the drunk man holding onto the lamppost at night. He uses the lamppost not for its illumination but for its support, exactly as you have used your scientific research.
Without believable dietary data, the SCS is nothing more than a marketing exercise.
In fact, as I will now argue, the SCS produced substantial evidence that disproves much of your advocacy. But, in another telling example of negative evidence, you never did acknowledge this.
5. Despite all these fatal flaws, the SCS produced a number of findings (50), and these conflicted with your hypotheses (51). You chose not to communicate these negative findings truthfully to the scientific community or the general public, the majority of whom continue to believe the SCS was a triumphal confirmation of your twin hypotheses (47). These findings include the following:
- Individual serum cholesterol concentrations were unrelated to the composition of the diet (50, p. 162). This negates your diet-heart hypothesis, which requires that the blood cholesterol concentration is determined by the fat content of the diet. Exactly similar findings were reported in the Framingham Heart Study (FHS), but they were conveniently hidden (35), a subterfuge about which you must have known. Importantly, unlike the SCS, the FHS rigorously analyzed the dietary data in a large number of participants.
In addition, the FHS was directed by many of your key research colleagues, including Drs. Jeremiah Stamler and W. B. Kannel. While you were not directly involved in that study, you would have been aware of the findings as they were analyzed. In time, you would have become aware of how they also conflicted with, indeed disproved, your hypotheses. But you failed to express those concerns in public.
The key point is that when diet appeared to have a relationship that fitted your hypotheses, you eagerly published that evidence, but when the evidence proved the opposite, you simply buried it.
Recall, also, that another study in which diet was carefully studied, the Tecumseh Study (52, 53), also found no evidence that diet determined the blood cholesterol concentrations. That study determined, “Serum cholesterol and triglyceride values were not positively correlated with selection of dietary constituents” (53, p. 1948), and therefore concluded, “These findings suggest that serum cholesterol and triglyceride levels in Americans are more dependent on degree of adiposity than on frequency of consumption of fat, sugar, starch, or alcohol” (p. 1948).
Could it possibly be that differences in levels of habitual physical activity or insulin resistance might be factors that you have overlooked in developing your hypotheses?
But more to the point, your special status as “Mr. Cholesterol” (43) gave you an awesome responsibility. The most important part of that responsibility was to keep the world informed of where the heart disease research was leading.
When the evidence became clear that dietary factors do not explain differences in blood cholesterol concentrations, you should have made that information available to the world immediately. You failed to do that.
Your ego again got in the way. You lacked the ego strength to admit that you could possibly have been wrong. This is yet another example of the negative evidence of the dog that did not bark when it might have.
- Elevated serum cholesterol concentrations were consistently related to higher rates of CHD (p. 160), including 10-year CHD death rates in the different populations that were studied (p. 161). This is entirely consistent with your lipid hypothesis. Well done.
However, death from all causes (all-cause mortality) were only slightly correlated with average serum cholesterol concentrations in the different populations/countries (p. 161). This means higher blood cholesterol concentrations must have provided protection against some (non-cardiac) causes of death.
Again, this has not been properly communicated to either medical professionals or the general public.
A simple statement by you would have sufficed. For example, what’s wrong with this: “Hi, I’m Dr. Ancel Keys. I’m the one who brought you the diet-heart and lipid hypotheses (that I “borrowed” from Dr. Gofman). In those early days I believed it was necessary to use a low-fat, low-cholesterol diet to lower blood cholesterol concentrations as much as possible. It turns out I was both right and wrong. While our work clearly suggests this may be true as far as heart disease risk goes, this seems not to be the case for other conditions. Our new evidence shows persons with low blood cholesterol concentrations don’t live any longer than those with high blood cholesterol concentrations, and this leads us to believe higher blood cholesterol concentrations may be associated with lowered risks for death from non-cardiac diseases.”
See? How difficult was that?
Imagine if you had had the courage to say that. You may have spared us all from the statin disaster, the idea that the lower the blood cholesterol concentration, the better for all.
- Indeed, the SCS found serum cholesterol concentrations were negatively (inversely) related to non-cardiovascular disease risk in the different populations/countries studied (p. 161). This means decreasing serum cholesterol concentrations in individuals “tended to be associated” with increasing rates of non-coronary deaths (p. 161).
Again, as I have already argued, this has not been properly communicated to either the medical profession or the general public. Yet another example of negative evidence.
- Longevity was unrelated to the amount of saturated fat or cholesterol the SCS subjects ate, or their blood cholesterol concentrations.
Once more, this has not been properly communicated to either the medical profession or the general public.
6. The prosecution argues the messaging the public should have received from the results of the SCS were the following (30):
- The blood cholesterol concentration bears no relationship to the amount of fat in the diet (at least according to the findings of the SCS).
- Having a high blood cholesterol concentration may increase your risk of developing CHD.
- Having a high blood cholesterol concentration may reduce your risk of dying from a non-cardiac cause.
- Having a low blood cholesterol concentration may increase your risk of dying from a non-cardiac cause.
- Having either a high or a low blood cholesterol concentration will not influence how long you live.
7. In short, the findings of the SCS showed that even if a higher blood cholesterol concentration is linked in some way, perhaps causally, to an increased risk of CHD, this effect is more than offset by the competing effects of either a high blood cholesterol concentration protecting against other non-cardiac conditions or a low blood cholesterol concentration promoting death from those same (non-cardiac) causes.
The key point is that longevity was unaffected by either eating a high-fat diet or having a higher blood cholesterol concentration.
In other words, the SCS showed that following your diet would not materially influence a person’s long-term health.
But this was not communicated to either the medical profession or the general public. Instead, members of both groups continue to believe eating your low-fat, low-cholesterol diet will improve health by reducing risk of CHD and that the SCS provided that evidence.
From your position as principal investigator of the SCS, you should have ensured the truth was presented to both medical professionals and the general public. You did not.
More negative evidence.
8. The SCS continues to be promoted as the study that proved “the primary dietary finding that saturated fat was correlated with heart disease” (47, p. 4), since “the correlation between saturated fat and coronary heart disease was due to a correlation between saturated fat intake and higher serum cholesterol levels, which subsequently affect coronary heart disease risk” (47, p. 19).
For instance, J. Andrade et al. write:
Keys and his colleagues were able to determine that in societies where fat was a major component of every meal (i.e., the U.S. and Finland), both the blood cholesterol levels and the heart-attack death rates were highest. Conversely, in cultures where diets were based on fresh fruit and vegetables, bread, pasta, and plenty of olive oil (i.e., the Mediterranean region) blood cholesterol was low and heart attacks were rare. The report published in 1970 had a decisive impact on CVD prevention, as it described one of the first studies to clearly show that dietary saturated fat leads to CVD, and that relationship is mediated by serum cholesterol. (54, p. 67)
This belief in the association of dietary saturated fat with CVD had profound consequences. As Andrade et al. also write:
Starting in the 1940s pioneering American physiologist Ancel Keys helped establish the epidemiological link between cholesterol and cardiovascular disease. Once Keys and others clarified the role of dietary saturated fat in disease development, early intervention studies examined the effects of diet and drug therapy. In the 1990s large trials of statins confirmed the benefits of lipid lowering therapy. (p. 66)
So you see, Dr. Keys, your hypotheses remain utterly embedded in the teaching of cardiology and the practice of modern medicine.
But how can these conclusions possibly have any validity if (i) the dietary data you collected were so scanty, and (ii) a key overall finding was that individual dietary intake was unrelated to blood cholesterol concentration? Your conclusion would have to be that the initial non-random selection of countries produced a false associational finding that could not be confirmed when individual patients were studied.
But this information was not conveyed to the general public or the medical profession. Therefore, we continue to have these completely false interpretations of what the SCS actually found.
You could have stopped these misinterpretations at the source, but you did not.
More negative evidence.
9. We come, next, to the prescription of the experimental, untested low-fat, low-cholesterol diet for President Dwight D. Eisenhower.
When Paul Dudley White, MD, chose to prescribe your theoretical and untested low-fat, low-cholesterol diet to one of the most important politicians in the world, he did so on the basis of how he had been influenced by your advocacy of the diet. This raises an important ethical question: Was this action appropriate, especially given the fact that the president had essentially no active risk factors for CHD and his blood cholesterol was below the range considered to indicate increased risk of CHD (55)? At the very least, the experimental and unproven nature of the diet should have been explained to the president, and his response to the diet should have been carefully monitored.
And, in fact, it was. Now, I cannot blame you for missing the diagnosis of insulin resistance/Type 2 diabetes mellitus in the president (54). This was the responsibility of Dr. White and the other members of the president’s medical team. But you were the scientist, and your job was to make sure the experiment did not go badly.
The evidence clearly suggested that the experiment went spectacularly badly (55, 56): The president was perpetually hungry; he gained weight; his blood cholesterol concentration rose progressively and inexorably; and his mood was not always ideal. But worse, his coronary artery disease progressed remorselessly until he died in intractable heart failure with two of his three major coronary arteries almost completely obstructed.
If ever there was direct evidence that your low-fat, low-cholesterol diet promoted rather than prevented coronary atherosclerosis, this was it.
But you apparently learned nothing from this experience. Instead, you again buried it.
More negative evidence.
10. It is important to record here that intimidation was the main technique you used to deter those who challenged your ideas (12, 13). This was most obvious in your actions against Prof. John Yudkin, whose career was terminated (57) in part because he did not believe you.
The irony is that Yudkin’s ideas may prove to be closer to the truth in the end. The SCS ultimately provided at least some evidence that sugar is a risk factor for CHD (30, 58).
11. Finally, you and your wife did not eat as you wanted the world to eat. In your 1961 Time magazine interview, you reported that you “do not eat ‘carving meat’ — steaks, chops, roasts — more than three times a week” (43, p. 6).
Indeed, the breakdown of the number of pages devoted to different recipes in your book (21) is quite interesting:
This is hardly the vegan/vegetarian diet some (47) of that persuasion would have us believe is the “prudent” diet you personally practiced and promoted.
That, Dr. Keys, is how the evidence has now stacked up against you.
You are charged with promoting two false hypotheses for which you did not ever present definitive evidence.
Rather, when the data from your own studies actively disproved those hypotheses, you did nothing to inform the scientific community and general public that you were wrong. Instead, you hid the data. Your mission and your character did not allow you to admit error.
Seldom in medical history has there been a more catastrophic result of such scientific dishonesty. What you failed to observe at the time was that prescribing a low-fat, low-cholesterol diet to persons with insulin resistance — persons such as President Eisenhower — causes them to become perpetually hungry, gain adipose tissue, and become progressively more insulin resistant, leading ultimately to the development of Type 2 diabetes mellitus (T2DM) with all its complications.
Your ignorance is not an excuse.
The lipophobia driven by your unproven claim that everyone should eat less fat in order to prevent CHD produced the first man-made pandemics in modern human history: the global epidemics of obesity and T2DM that began after 1977 when the dietary guidelines changed to incorporate your false teachings.
Had you exhibited the requisite character to admit error, you could have prevented these frightful outcomes.
Now, perhaps, it may be too late.
How do you plead to these charges?
- Mann GV. The Way It Seems. Los Angeles, CA: Vantage Press, 1988.
- Keys A, Brozek J, Henschel A, et al. The Biology of Human Starvation. Minneapolis, MN: University of Minnesota Press, 1950.
- Keys A. Seven Countries: A Multivariate Analysis of Death and Coronary Heart Disease. Cambridge, MA: Harvard University Press, 1980.
- Kromhout D, Menotti A, Blackburn H. The Seven Countries Study: A Scientific Adventure in Cardiovascular Disease Epidemiology. Marjan Nijssen-Kramer. Bilthoven, Netherlands: Studio RIVM, 1993.
- Noakes TD. It’s the insulin resistance, stupid: Part 6. CrossFit.com. 30 Oct. 2019. Available here.
- Yerushalmy J, Hilleboe HE. Fat in the diet and mortality from heart disease. A Methodological note. New Y State J Med. 57(1957): 2343-2354.
- Keys A. Epidemiologic aspects of coronary artery disease. J Chron Dis 6(1957): 552-559.
- Keys A. Human atherosclerosis and the diet. Circulation 5(1952): 115-118.
- Malmros H. The relation of nutrition to health. A statistical study of the effect of the war time on arteriosclerosis, cardiosclerosis, tuberculosis and diabetes. Acta Med Scand 246(1950): 137-153.
- Olszewski TM. The causal conundrum: The diet-heart debates and the management of uncertainty in American Medicine. J Hist Med Allied Sci 70(2014): 218-249.
- Mann GV. A short history of the diet/heart hypothesis. In: Coronary Heart Disease: The Dietary Sense and Nonsense. An Evaluation by Scientists. Mann GV, Ed. London, England: Janus Publishing Company, 1993; pp. 1-17.
- Keys A, Grande F, Anderson JT. Bias and misrepresentation revisited: Perspective on saturated fat. Am J Clin Nutr. 27(1974): 188-212.
- Keys A. Sucrose in the diet and coronary heart disease. Atherosclerosis 14(1971): 193-202.
- Yudkin J. Diet and coronary thrombosis. Hypothesis and fact. Lancet 1957;273:155-162.
- Yudkin J. Dietary fat and dietary sugar in relation to ischaemic heart disease and diabetes. Lancet 2(1964): 4-5.
- Reiser R. Saturated fat in the diet and serum cholesterol concentration: a critical examination of the literature. Am J Clin Nutr 26(1973): 524-555.
- Mann GV, Shaffer RD, Anderson RS, et al. Cardiovascular disease in the masai. J Atherosclerosis 4(1964): 289-312.
- Mann GV, Spoerry A, Gary M, et al. Atherosclerosis in the Masai. Am J Epidemiol. 95(1972): 26-37.
- Taubes G. Good Calories, Bad Calories: Fats, Carbs, and the Controversial Science of Diet and Health. New York: NY: Anchor Books, 2008.
- Keys A. Coronary heart disease – The global picture. Atherosclerosis 1975;22:149-192.
- Keys A, Keys M. Eat Well and Stay Well. Doubleday and Company, Inc. Garden City; NY: 1963.
- Anonymous. Dietary goals for the United States: Statement of the American Medical Association to the Select Committee on Nutrition and Human Needs, United States Senate. Rhode Isl Med J 60(1977): 576-181.
- Ahrens EH. Dietary fats and coronary heart disease: Unfinished business. Lancet 314(1979): 1345-1348.
- National Research Council. Toward Healthful Diets. Food and Nutrition Board; National Academy of Sciences, Washington, D.C. 1980. Available here.
- Noakes TD. It’s the insulin resistance, stupid: Part 9. CrossFit.com. 11 December 2019. Available here.
- Gofman JW, Jones HB, Lindgren FT, et al. Blood lipids and human atherosclerosis. JAMA 11(1950): 161-178; Jones HB, Gofman JW, Lindgren FT, et al. Lipoproteins in atherosclerosis. Am J Med 11(1951): 358-380; Gofman JW. Diet and lipotrophic agents in atherosclerosis. Bull NY Acad Med 28(1952): 279-293; Gofman JW, Glazier F, Tamplin A, et al. Lipoproteins, coronary heart disease and atherosclerosis. Physiol Rev 34(1954): 598-607; Gofman JW, Tamplin A, Strisower B. Relation of fat intake to atherosclerosis. J Am Diet Assoc 30(1954): 317-326; Lyon TP, Yankley A, Gofman GW, et al. Lipoproteins and diet in coronary heart disease. A five-year study. Cal Med 84(1956): 325-328.
- Keys A. “Giant molecules” and cholesterol in relation to atherosclerosis. Bull J Hopkins Hosp 1951;88:473-483.
- Keys A. Cholesterol, “Giant molecules” and atherosclerosis. JAMA 147(1951): 1514-1519.
- Teicholz N. The Big Fat Surprise: Why Butter, Meat and Cheese Belong in a Healthy Diet. New York, NY: Simon and Schuster, 2014.
- Noakes TD. It’s the insulin resistance, stupid: Part 7. CrossFit.com. 11 November 2019. Available here.
- Dehghan M, Mente A, Zhang X, et al. Association of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): A prospective cohort study. Lancet 390(2017): 2050-2062.
- Chamberlin TC. The method of multiple working hypotheses. Science 148(1965): 754-759.
- Mayosi BM, Forrester T. Commentary: “Serum-cholesterol, diet, and coronary heart-disease in Africans and Asians in Uganda” by AG Shaper and KW Jones. Int J Epidemiol 41.5(2012): 1233-1235.
- Levenstein H. Fear of Food. A History of Why We Worry About What We Eat. London, U.K.: University of Chicago Press, 2012.
- Noakes TD. It’s the insulin resistance, stupid: Part 8. CrossFit.com. 24 Nov. 2019. Available here.
- Mahmood SS, Levy D, Vasan RS, et al. The Framingham Heart Study and the epidemiology of cardiovascular disease: A historical perspective. Lancet 383(2014): 999-1008.
- Brewer ER, Ashman PL, Kuba K. The Minnesota Coronary Survey: Composition of their diets, adherence, and serum lipid response. Circulation 51.suppl 2(1975): 1-269.
- Dawson E, Gatewood L. The Minnesota Coronary Survey: Methodology and characteristics of the population. Circulation 52.suppl 2(1975): 271.
- Frantz ID, Dawson EA, Ashman PL, et al. Test of effect of lipid lowering by diet on cardiovascular risk. The Minnesota Coronary Survey. Arteriosclerosis 9(1989): 129-135.
- Ramsden CE, Zamora D, Majchrzak-Hong S, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73). BMJ 353(2016): i1246.
- Begley S. Records found in dusty basement undermine decades of dietary advice. Raw data from a 40-year-old study raises new questions about fats. Sci Amer. 19 April 2017; Whoriskey P. This study 40 years ago could have reshaped the American diet but it was never published. The Washington Post 12 April 2016.
- Anderson JT, Grande F, Keys A. Cholesterol-lowering diets. J Am Dietet Assoc 62(1973): 133-142.
- Anonymous. Medicine: The Fat of the Land. Time 13 Jan. 1961. Available here.
- Skotnicki M. “The Dog That Didn’t Bark:” What we can learn from Sir Arthur Conan Doyle about using the absence of expected facts. Briefly writing, July 25, 2012. Available here.
- Gandhi L. What does “Sold Down The River” really mean? The answer isn’t pretty. Code Switch. npr.org. 27 January 2014. Available here.
- Wood PDP. A possible selection effect in medical science. J Roy Stat Soc 30(1981): 131-135.
- Pett KD, Kahn J, Willett WC, et al. Ancel Keys and the Seven Countries Study: An evidence-based response to revisionist histories. White Paper commissioned by the True Health Initiative. 1 August 2017. Available here.
- Keys A, Aravanis C, Sdrin H. The diets of middle-aged men in two rural areas of Greece. In: Dietary studies and epidemiology of heart disease. The Stichting tot wetenschappelijke Voorlichting op Voedingsgebied. The Hague, Netherlands: 57-68.
- Keys A. Introduction. In: Den Hartog C, Buzina K, Fidanza F, Keys A, Poine P. The Stichting tot wetenschappelijke Voorlichting op Voedingsgebied. The Hague, Netherlands: 4-9.
- Keys A. The inception and pilot surveys. In: Kromhout D, Menotti A, Blackburn H. The Seven Countries Study. A scientific adventure in cardiovascular disease epidemiology. Marjan Nijssen-Kramer. Bilthoven, Netherlands: Studio RIVM, 1993. pp.159-174.
- Keys A. Seven Countries: A Multivariate Analysis of Death and Coronary Heart Disease. Cambridge, MA: Harvard University Press, 1980.
- Nichols AB, Ravenscroft C, Lamphiear DE, et al. Daily nutritional intake and serum lipid levels. The Tecumseh study. Am J Clin Nutr 29(1976): 1384-1392.
- Nichols AB, Ravenscroft C, Lamphiear DE, et al. Independence of serum lipid levels and dietary habits. JAMA 236(1976): 1948-1953.
- Andrade J, Mohamed A, Frohlich J, et al. Ancel Keys and the lipid hypothesis: From early breakthroughs to current management of dyslipidemia. BC Med J 51(2009): 66-72.
- Noakes TD. It’s the insulin resistance, stupid: Part 5. CrossFit.com. 22 Oct. 2019. Available here.
- Lasby CG. Eisenhower’s Heart Attack: How Ike Beat Heart Disease and Held on to the Presidency. Lawrence, KA: University Press of Kansas, 1997.
- Leslie I. The sugar conspiracy. The Guardian 7 April 2016. Available here.
Menotti A, Kromhout D, Blackburn H, et al. Food intake patterns and 25-years mortality for coronary heart disease: Cross-cultural correlations in the Seven Countries Study. Europ J Epidemiol 15(1999): 507-515.
Cyril Ramaphosa: The Audio Biography
Listen to the story of Cyril Ramaphosa's rise to presidential power, narrated by our very own Alec Hogg.