The Ancel Keys Cholesterol Con. Part 12. 1984-1993
1984 NIH Consensus Development Conference
Any reasonable intellect would have to conclude that the LRC CPPT findings released in 1984, did not support Keys’ Lipid Hypothesis. They could only ever be interpreted as evidence that even if the Lipid Hypothesis was partially true, the contribution of a raised blood cholesterol concentration to CHD was miniscule. Even more importantly, as had been found with the WHO Co-operative Trial in the Primary Prevention of Ischaemic Heart Disease using Clofibrate (1,2), trying to lower blood cholesterol concentrations with drugs caused measurable harm, whatever the benefits in reducing CHD risk may be.
Nevertheless the finding were enough to encourage the NHI to the next phase in driving the global cholesterol mania. After all the NHI had spent a great deal of money proving the dangers of “cholesterol” and it was not about to allow that error to be remembered as a waste of money.
Accordingly the NHI formed yet another committee, the Consensus Development Panel whose members were listed (3, p.8), to consider “whether decreasing cholesterol levels to a comparable degree by dietary means rather than by cholestyramine treatment would give a comparable decrease in events” (3, p.7). Which would of course not be difficult since, when analysed with the appropriate statistical methods, the cholestyramine study failed to produce a significant reduction in “events” (4). Rifkind who had overseen that LRC CPPT study proposed that the committee should be under the chairmanship of another of Keys’ most willing acolytes, Daniel Steinberg.
But Steinberg was hardly an independent investigator. His laboratory had been one of the 12 laboratories in the failed LRC CPPT trial (4). And he would played a central role in misrepresenting the study as if it had been a massive success.
The 14-person Consensus Panel was tasked to answer five questions (3, p.8):
- Is the relationship between blood cholesterol levels and coronary heart disease causal?
- Will reduction of blood cholesterol levels help prevent coronary heart disease?
- Under what conditions and at what level of blood cholesterol should dietary or drug treatment be started?
- Should an attempt be made to reduce the blood cholesterol concentrations of the general population?
- What research directions should be pursued on the relationship between blood cholesterol and coronary heart disease?
Another group of 20 experts were tasked to present reviews of all the evidence pertaining to the relationship between cholesterol and coronary heart disease; to identify what proportion of the US population was at risk and finally to product instruction to US physicians on how to treat this novel threat. What they decided would have a global impact.
The 20 speakers presented their deliberations before the Consensus Panel and an audience of more than 600 physicians and researchers (3, p.8-12; 5, p.56-64). The Consensus Panel was allocated just 60 hours to review all the evidence and to produce a consensus statement; clearly an impossible task.
Included amongst the key expert speakers were Robert I. Levy MD who had co-directed the LRC CPPT trial, who represented the NHI and who was therefore heavily committed to finding cholesterol guilty on even the flimsiest of evidence. Levy’s argument was that cholesterol is the specific culprit in the causation of CHD and that there is no safe blood cholesterol concentration since risk rises inexorably with increasing blood cholesterol concentrations. This statement is clearly false.
Future Nobel Laureates for their work showing that the absence of LDL receptors in the liver cause the condition of Familial Hypercholesterolemia (FH) (6), Joseph L. Goldstein and Michael S. Brown, presented their findings which were then considered the definitive proof that elevated blood cholesterol concentrations are the sole and direct cause of CHD. The problem for Goldstein and Brown was that whilst their Nobel Prize discovery of the molecular basis for FH would be used to justify the widespread use of statin drugs, their findings actually undermine the Diet-Heart hypothesis. Since the elevated blood cholesterol concentrations in FH are the result of a genetic disorder and are not due to high-fat diet. Indeed a low-fat diet is of little value in the management of this condition (7). But this point is never mentioned in polite cardiological society.
Other speakers included Keys’ acolytes Jeremiah Stamler who spoke about the evidence from epidemiological studies; Basil Rifkind who presented the (non-existent) evidence from the LRC CPPT trial; Scott Grudy, from the Goldstein/Brown laboratory, who reviewed the efficacy of dietary management in CHD; and the heavily conflicted Antonio Gotto who addressed the question: What are the optimal cholesterol levels towards which we should aim for the American public at large?
Since Gotto was in the pocket not just of the pharmaceutical industry (8) but also of Big Food (9), the nature of his message would have been predictable: Get the blood cholesterol as low as possible, preferably to zero. And replace saturated fat with corn oil or related sources of manufactured polyunsaturated fatty acids.
Another speaker who brought significant bias to the discussions was Keys’ colleague at the University of Minnesota School of Public Health, Henry Blackburn. He spoke about “the appropriateness of public health measures to reduce blood cholesterol levels”.
The expert speakers did include three “sceptics” – Dr Michael Oliver from Edinburgh University, Scotland, who spoke about screening for hypercholesterolemia; E.H. “Pete” Ahrens’s whose topic was “the lack of appropriateness at this time of public health measures to change American dietary habits”; and Robert Olson who reviewed the evidence of whether or not lowering the blood cholesterol concentration prevents (coronary) heart disease.
A key presentation on the first day was from Dr Richard Peto from Oxford University who, according to Steinberg, “gave an electrifying presentation that included what may have been his first application of meta-analysis, which he is generally credited with introducing into epidemiology” (3, p.9). His conclusion was that when all of the diet studies were “lumped” together, there was a statistically significant decrease in coronary heart disease risk when blood cholesterol concentrations were reduced.
But not everyone agreed with Peto’s optimistic summation. So University of Chicago biostatistician and frequent adviser to the NIH trials, Paul Meier argued that: “We simply have to acknowledge that the [trial] results have been disappointing. These studies show no whisper of benefit in reducing total morality, and I am not entirely comforted by Richard Peto’s ability to explain away the significant excess of non-coronary events” (5, p.60).
Oxford Professor Salim Yusif’s also dissented: “Many of the trials that had the most striking effect on the reduction in cardiac deaths did not report total mortality. If you include only the trials that reported total mortality the effect is actually none” (5, p.59). Yusif’s opinion was that the evidence was strong that lowering blood cholesterol indeed reduced the risk of CHD but “it was crucial to examine all the effects of treatment. When you did that the net result was zero” (5, p.59).
Steinberg ensured that Yusif’s microphone was cut off before he could develop his inconvenient argument. Importantly Yusif was a co-author of Peto’s manuscript and so knew more than anyone else in the audience about the evidentiary basis for Peto’s conclusions.
Long-term cholesterol “sceptic” Michael Oliver added his opinion that over all the trials, the statistical trend towards increased deaths from other causes was as strong as the trend towards reduced deaths from coronary heart disease (5, p.59).
Following Peto’s presentation “which had an important effect on the Panel’s deliberations”, the star turn was that of Basil Rifkind who presented the results from what the NHLBI had labelled “the keystone in the arch” (3, p.9), “the crown jewel of lipid research” (5, p.60) – specifically the failed LRC CPPT cholestyramine trial that had achieved a statistically-significant (but biologically-irrelevant) finding only after appropriate statistically massaging (4).
Rifkind enthused: “It is thought to be the first study in man to establish conclusively that lowering (blood) cholesterol reduces heart attacks and heart attack deaths” (5, p.60). But the study did neither. To explain why, if these effects were real, total mortality was not reduced by the intervention, Rifkind’s excuse was: “The conclusion of the investigators was this small discrepancy most likely reflected a chance occurrence” (5, p.61).
But Rifkind’s explanation did not fool everyone. From the audience Dr Paul Meier of the University of Chicago commented: “To call ‘conclusive’, as Dr Rifkind did, a study which no difference in total mortality, and by the usual statistical criteria, an entirely non-significant difference in coronary incidents, seems to me a substantial misuse of the term” (5, p.61). His response to Rifkind’s attempted explanation of the failure to show a reduced total mortality, Meier was equally direct: “Any statistician would turn in his badge if he couldn’t find an excuse like that for any outcome. It’s just too easy to do” (10, p.41).
The controversy did not end there. In discussing dietary recommendations, three different experts presented three different diets as the ultimate solution for CHD. The first problem these three speakers sought to resolve was: Who was being targeted by this dietary advice? Would a diet designed to prevent CHD in middle-aged men be ideal for everyone including children and pregnant women?
In the end the Panel decided, yes, all would be treated equally since their goal was to ensure that the blood cholesterol concentration should become the sole blood marker necessary to predict the future health of any human of any age, gender or ethnic origin.
By the final morning of the Conference it was clear that Steinberg did not seek consensus. He had long before made up his mind on what he wanted the outcome to be – in line with his subsequent writings on the topic, the defining characteristic of which are his dogmatic certainty that only his solipsistic opinions are the correct ones (11-13). His interest was to present “a campaign document and a call to action for the NHI’s cholesterol education program” (5, p.62) – a program that was even then being drafted by the his like-minded supporters in the NHLBI. So he was in a great hurry to get the Consensus proceedings behind him.
Indeed Edinburgh’s Michael Oliver felt the whole process was simply too rushed: “As a visitor, I want to suggest you be a little cautious, that you stay within the areas where you are confident… We approached the problem a little differently. We had ten bureaucrats and we sat fifteen days over a period of two years. We spent another fifteen days writing, reading, and preparing the reports” (5, p.62). Eliot Corday a member of the Federal Drug Agency and an NHI advisor, concurred: “After listening to the testimony today which promises so much, why don’t we hold off a bit. We’re not ready to come out with a consensus and pour present concepts in concrete” (5, p.62).
But, for reasons that will become shortly become obvious, the tight agenda had to be maintained. At all costs.
The discussion on the third day focused on the definition of what constitutes a dangerously high level of blood cholesterol and how it should be managed. Predictably many panellist argued that the “normal” cholesterol in the US was far too high and that a desirable goal for the entire US population should be a blood cholesterol concentration less than 200mg/dL (5.2mmol/L) (or even less – (3, p.10)). This outcome was predictable since the Panel had been loaded with “experts” whose careers required them to come to that conclusion, regardless of the evidence.
Finally the Panel came up with “a simple set of numbers that satisfied everyone. We proposed ‘desirable’ levels of <200 for persons younger than 20 years; <220 for those 30 to 39 years; and <240 for those older than 40 years, and we proposed the same guidelines for men and women” (3, p.10). This despite the fact that there was no evidence then or even now that elevated blood cholesterol concentrations carry identical risk in men and women. Clearly the absence of hard evidence for all this advice was not something that particularly concerned the Consensus Panel.
So at the end of the third day the Consensus Statement was finalised, released and subsequently published (14). It read: “Elevation of blood cholesterol levels is a major cause of coronary artery disease. It has been established beyond a reasonable doubt that lowering definitely elevated blood cholesterol levels (specifically blood levels of low-density lipoprotein [LDL] cholesterol) will reduce the risk of heart attacks caused by coronary heart disease. This has been demonstrated most conclusively in men with elevated blood cholesterol levels, but much evidence justifies the conclusion that similar protection will be afforded to women with elevated levels. (Untrue as there was no such evidence in 1984 – my addition). After careful review of genetic, experimental, epidemiologic, and clinical trial evidence, we recommend treatment of individuals with blood cholesterol levels above the 75th percentile (upper 25% of values). Furthermore, we are persuaded that the blood cholesterol levels of most Americans are undesirably high in large part because of our high dietary intake of calories, saturated fat, and cholesterol (Untrue. This ignored the evidence from the Framingham Heart study (6,15), the Seven Countries study (16,17) and numerous other studies (17) showing that persons with CHD do not have higher intakes of saturated fat and cholesterol that do those without CHD. It also ignored the evidence from the MRFIT trial that a low fat diet plus other successful attempts to lower CHD risk factors had zero impact on CHD outcomes (6) – my addition). In countries with diets lower in these constituents, blood cholesterol levels are lower and coronary heart disease is less common. (This is also a dubious conclusion since it then relied on the evidence from the Seven Countries study and that evidence was extremely dubious because the dietary analyses were so poorly conducted (16,17) – my addition). There is no doubt that appropriate changes in our diet will reduce blood cholesterol levels (Always be wary of the statement – there is no doubt. Because that proves that the authors know there is doubt. This statement is true for metabolic ward studies (18) but completely untrue for free living populations. Recall the words of William Castelli from the Framingham study: “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 (as measured here). Clearly if there is to be an attempt to manipulate blood cholesterol levels in the general population, it would be desirable to know what these powerful but unspecified forces are” (5, p37; 19, p. 27). Or in the words of other Framingham scientists: “…zero or near zero correlations were found between the various components of the diet and serum cholesterol levels” (20, p.77) – my additions). Epidemiological data and more than a dozen clinical trials allow us to predict with reasonable assurance that such a measure will afford significant protection against coronary heart disease” (14, p.2080-2081).
History has been particularly harsh on this final conclusion. The most modern analysis of 44 published studies found that: “The preponderance of evidence indicates that low-fat diets that reduce serum cholesterol (concentrations) do not reduce cardiovascular events or morality. Specifically, diets that replace saturated fat with polyunsaturated fat do not convincingly reduce events or mortality. These conclusions stand in contrast to current opinion (21)” (22, p.5).
As I argue subsequently this conclusion does not go far enough. There is now sufficient evidence that replacing dietary saturated fat with polyunsaturated fat causes harm. Therefore it is no longer ethical to advise patients to restrict the amount of saturated fat that they ingest. Even the hubristically sounding American Heart Association’s Presidential Advisory has had to conclude that: “A dietary strategy of reducing intake of total dietary fat, including saturated fat, and replacing the fats mainly with unspecified carbohydrates does not prevent CHD” (21,p.e17).
Incorrectly the same Presidential Advisory claims that: “Randomized clinical trials showed that polyunsaturated fat from vegetable oils replacing saturated fat from dairy and meat lowers CVD” (p. e17). This statement is not universally true as shown by the data from the Recovered Sydney Diet Heart Study, the Recovered Minnesota Coronary Experiment and the grandly titled Women’s Health Initiative Randomized Controlled Dietary Modification Trial (23).
But as always the core dogma is emphasized: “Saturated fat increases LDL cholesterol, a major cause of atherosclerosis and CHD, and replacing it with polyunsaturated or monounsaturated fat decreases LDL cholesterol” (p. e17). Hence the assumption that anything that lowers the blood cholesterol concentration must also prevent CHD.
Yet this is clearly untrue for the prescription of cholesterol-lowering statin drugs, the statins, which are “of doubtful benefit” according to some authors (24, p.959).
But 35 years ago, Rifkind was absolutely adamant in his certainty. Largely as a result of the LRC CPPT he concluded that the “cholesterol arguments are dying down. The focus of the debate has shifted. A year or two ago, people would have been arguing about whether cholesterol should be reduced in high-risk people. Now there is no dissenting voice on whether cholesterol should be reduced in high-risk people” (10, p.41).
The Consensus Conference Panel also produced dietary guidelines in line with those of the American Heart Association (AHA) Nutrition Committee (25) which is predictable given that the AHA and the NHI function as a twin organization, joined at the hip. Thus their dietary advice was: Exercise and reduce total calories to maintain normal body weight; decrease total calories from dietary fat to 30% (<10% saturated fat); and reduce daily cholesterol intake to <300mg. (3, p.10). This despite clear evidence in the literature from the master himself, Ancel Keys, that dietary cholesterol intake has essential zero impact on blood cholesterol concentrations (26) and the other studies discussed earlier which have failed to show that the low-fat prudent diet protects against CHD.
Finally to ensure that they still had work in the future, the Panel advised that under the guidance of the NHLBI, there should be established “a national program, involving all of the major medical and public health associations, to educate both physicians and the public on the importance of controlling cholesterol levels”.
Thus was born the National Cholesterol Education Program based as it was on the failed LRC CPPT trial results and driven by a group of scientists who, like Keys himself, had decided that cholesterol was guilty, whatever the evidence might show.
This Consensus Statement and the related Recommendations for Treatment of Hyperlipidemia in Adults: A Joint Statement of the Nutrition Committee and the Council on Arteriosclerosis (27) finally buried Gofman (28) and Fredrickson’s classifications of the hyperlipidemias (17,29) which established that high carbohydrate diets produced abnormal blood lipid profiles, in particular elevated blood triglyceride concentrations, that are at least as likely to predict risk of CHD as are elevated blood cholesterol concentrations.
Thus the Nutrition Committee and the Council on Arteriosclerosis (27) under the chairmanship of the industry-directed (8,9) Dr Antonio Gotto, MD proposed that the “approach to lowering triglycerides is identical to that described for mildly elevated cholesterol levels, namely, decreased intake of saturated fatty acids and cholesterol, and in obese patients, curtailment of total calories” (27, p.1079A).
Thus according to Gotto’s committee, the treatment of hypertriglyceridemia requires weight reduction as “the first order of business” (p.1079A); “a second goal in dietary therapy of elevated triglycerides is to decrease intakes of saturated fatty acids” (p.1079A) as, apparently, “these fatty acids intensity lipoprotein overproduction” (p.1079A) which is clearly false since it has been known since the work of Reaven in the early 1960s that increased blood insulin concentrations driven by high carbohydrate diets drive hepatic production of triglycerides (figure 1 in 30). But who’s to question this “tiny” error amongst friends?
So the treatment of hypertriglyceridemia requires the replacement of dietary saturated fatty acids. But with what? “Some years ago, polyunsaturated fatty acids were thought the ideal substitute for saturated acids when triglycerides are high. Polyunsaturates will lower both VLDL and LDL in hypertriglyceridemic patients. They certainly are preferable to saturated fats. The AHA, however, takes the position that as a general rule, carbohydrates are a better substitute than polyunsaturated fatty acids. The reasons are: 1) Polyunsaturates have a higher caloric density and can promote weight gain. Weight loss is easier on a low-fat diet (False – my addition). 2) Polyunsaturates can cause increased synthesis of cholesterol in some hypertriglyceridemic patients… 3) Polyunsaturates can raise the cholesterol content of bile (31) and might increase risk of cholesterol gallstones (32). 4) The long-term consequences of high intakes of polyunsaturates are unknown. They alter the composition of cell membranes (33), and in experimental animals, they can act as cocarcinogens (34). Therefore, despite the fact that an increased carbohydrate intake can stimulate synthesis of VLDL-triglycerides (a fact ignored up to this point – my addition), carbohydrates seem preferable to polyunsaturates” (p.1079A).
What is perhaps most remarkable about this circumlocution is the statement from the AHA that “the long-term consequences of high intakes of polyunsaturates are unknown” even though the AHA had, on behalf of their commercial “sponsors”, been promoting the increased use of polyunsaturated fats (PUFAs) since 1961 (35) as if the safety and efficacy of PUFAs was already established.
Yet all of a sudden PUFAs might be damaging cell membranes and causing gallstones and cancer! Pretty much as Fred Kummerow had warned already in 1978 (23). Predictably no Heart Association anywhere in the world any longer remembers this cautionary note of 1984. For the carefully orchestrated industry-directed advocacy of nutrition scientists like Willett (23,36), Hu and Mozaffarian (23) have effectively “health-washed” polyunsaturated fats as the only form of fat safe for human consumption
These pronouncements (25,27) were the final nail in the coffin of any future possibility that carbohydrate-sensitive hypertriglyceridemia would ever again be considered a possible cause of CHD. What the Sugar Research Foundation had begun with its successful payment to the Harvard School of Public Health (2) to bury any possible evidence that sugar and carbohydrates might play in the causation of CHD, these Consensus Statements had finally consecrated.
But these statements was not universally well received. Edward “Pete” Ahrens typically direct statement was: “I think the public is being hosed by the NIH and the American Heart Association. They desire to do something good. They’re hoping to God that this is the right thing to do. But they are not acting on the basis of scientific evidence, but on the basis of plausible but untested data” (5, p.63).
In an editorial published in the Lancet in May the following year (37), Ahrens extended his criticism. He argued that patients must be treated as individuals. But this was ignored by those who took a public health approach to the problem, mandating that the blood cholesterol concentration should be below a certain value for the entire US population. Again he warned that epidemiological studies on which the Consensus Panel relied so heavily provided only correlations. “But correlations, no matter how strong, are never proof” (37, p.1085). Instead his more cautious opinion was: “I would have been content with the consensus statement if it had confined itself to what we do know and what we do not. It promises benefits without giving the evidence to back up that promise. By failing to emphasize what we do not know, the statement sweeps these weaknesses in our evidence under the rug, as if they were trivial. I have disagreed with that position (38)” (37, p.1087).
In that 1979 article (38) Ahrens had pointed out that his own research had clearly shown that “plasma lipids can be predictably altered by changes in the dietary fat 25 years ago (18)” so that this finding ensured that “we are intellectually and emotionally involved in the outcome of any test of the lipid hypothesis” (p.1346). “However” he continued, “if the public’s diet is going to be decided by popularity polls and with diminishing regard for the scientific evidence, I fear that future generations will be left in ignorance of the real merits, as well as the possible faults, of any given dietary regimen aimed at prevention of CHD” (p.1346).
Ahrens then discussed his four reasons “for resisting the current advocacy of a low-fat national diet” soon to be labelled the “prudent diet”.
- There has been no previous test of the “prudent” diet.
- The “prudent” diet will have only a small effect on plasma lipid levels.
- Any one diet produces different results in different people.
- Crucial questions remain to be resolved. In this context “Hyperlipidaemia is only one of the risk factors predisposing to CHD, and the genesis of hyperlipidaemia in more than 95% of the population is simply not understood..( my added emphasis)” (p.1347).
His conclusion was that: “In view of these many considerations and uncertainties, I feel it is irresponsible to make the dietary recommendations that are being so widely proposed to the general public at this time” (p.1348).
In an editorial published alongside that of Pete Ahrens, Michael Oliver (39) criticized the concept of a “consensus development conference”. This he argued is a contrived situation “and unlikely to be achieved within the short time set aside for such a meeting” (p.1088). Unless of course, “The panel of jurists…was selected to include experts who would, predictably, say….that all levels of blood cholesterol in the United States were too high and should be lowered” (p.1088). Which was the natural outcome of this Consensus Conference.
Oliver offered a series of alternative procedures for formulating policies before warning: “But well-orchestrated so-called consensus conferences between doctors and the public with the implicit intention of exerting psychological and political pressure should not be permitted too loud a voice – however powerful the protagonists might seem to be – and be recognized as special pleading and evaluated as such” (p.1089).
Thomas Chalmers of the Mt Sinai Medicine School and the Harvard School of Public Health stated that the Panel had made “an unconscionable exaggeration of all the data” (5, p.63; 10, p.40). Whilst he was not averse to treating persons at high risk: “I am not against taking measures to lower cholesterol if you are in a high-risk group. But if you are at high risk you have something to gain. If you are in the low-risk category, the side effects become more significant” (10, p.41).
Furthermore the was concerned by the absence of evidence that cholesterol-lowering saves lives: “The American public might be more interested in whether they will live longer than in how they will die” (10, p.41). He also argued that, “there is absolutely no evidence that it’s safe for children to be on a cholesterol-lowering diet. I don’t think that anyone believes people start laying down plaques before puberty and then (it happens) only in males. So why subject children to the diet starting at age two? The [consensus panel’s] argument that you have to start the diet early to make it a habit is fatuous” (10, p.41).
Salim Yusuf expressed his concern that whereas the death rate from heart disease may be reduced in drug or diet interventions, death rates from other conditions, frequently cancer, are often increased, suggesting that cholesterol lowering “is not always safe for everyone” (10, p.41). Paul Meier of the University of Chicago, whilst supporting a move to advise adults to modify their diets to lower their blood cholesterol concentrations was adamant: “But none of this excuses misrepresenting the evidence. Our first obligation is to be honest and forthright” (10, p.41).
But honesty and forthrightness had never been the sustaining values of the campaign to demonize cholesterol.
In his response to these criticisms Steinberg (3, p.10-12) focused his own comments on the writings of Science editor, Gina Kolata (10), and author Thomas Moore (5). For daring to suggest that “There are two sides to every story”, Kolata was chided that as a mere reporter she would have “no way of knowing which point of view represents the majority opinion” (3, p.11). So what Kolata “probably did not know is that the small group that she talked with after the conference represented a rather small minority of the experts in the field” (p.11).
Moore was criticized for comments he made in an article first published in The Atlantic in September 1989 (40) and which became the basis for his subsequent book (5). In the article Moore had written that “…the dissenters have been overwhelmed by the extravaganza put on not just by the heart institute but by a growing coalition that resembles a medical version of the military-industrial complex. This coalition includes….the ‘authorities’…the heart institute itself…and the American Heart Association” (3, p.11).
Moore was also criticized for pointing out the conflicts of interest shared by the “Gang of Five” who were heavily involved in the Consensus Conference and its offshoot the National Cholesterol Education Program – Scott Grundy, Antonio Gotto, John LaRosa, Robert I. Levy and Daniel Steinberg. Moore inconveniently pointed out that all these influential scientists had significant involvement with the US drug company Merck which was about to launch the very first cholesterol-lowering drug into the US market (5, p.68-70).
Was it perhaps a little too convenient that the NHI National Consensus Development Conference (NCDC) and its follow-up the National Cholesterol Education Program (NCEP), were happening just prior to Merck’s launch of drugs that would be marketed for their cholesterol-lowering effects? Which was exactly what the NCDC and NCEP were arguing had to be done with the utmost urgency to save the world from the coming CHD epidemic (which just happened already to be in retreat).
Perhaps it was really quite convenient that the Consensus Panel had come out so strongly in favour of lowering blood cholesterol concentrations in all Americans as a matter of great urgency. At a time when drugs able to achieve this end were about to be released onto the US market. Co-incidence?
But the criticisms and warnings of these independent scientists were without effect
For the Consensus Statement was in truth merely a pretext for the launching of what would become the most lucrative medical intervention in the history of US medicine.
It would be driven by The National Cholesterol Education Program (NCEP).
1987. The launch of the NHI National Cholesterol Education Program
By 1985 it was clear that despite the simple message of the Consensus Conference, US physicians had yet to become enthusiastic supporters of the Diet-Heart and Lipid Hypotheses. They still lacked the dogmatic certainty of the AHA, the NHLBI and Keys’ acolytes, including the Gang of Five, that cholesterol is the most important cause of CHD. In their scepticism, physicians lagged somewhat behind the vast majority of the general public in the US who were substantially more certain about this relationship.
But now that Sternberg and his collection of group thinkers had decided that the LRC CCPT provided the definitive scientific evidence of the value of (drug-induced) blood cholesterol lowering on future CHD risk, the NIH decided it was time to act on the outcome of their fake Consensus Conference. Their target would not be the general public. Rather it would be the still-sceptical US physicians. They needed to be brought on board the cholesterol bandwagon.
So suitably “Armed now with the results of the CPPT and a consensus (sic) among the leaders in the field, the NIH decided to get into high gear” (3, p.12). They would use the National Cholesterol Education Program (NCEP) to drive change: “..the NCEP played a key role in this sea change in ‘cholesterol awareness’” (3, p.12)
Producing that sea change in “cholesterol awareness” would be the task of the 1985 NCEP under the Director of the National Heart Lung and Blood Institute (NHLBI) Claude Lefant (3, p.12-13). Once again a huge team of like-minded persons would be assembled. In the end representatives from 24 national health professional organizations including the American Medical Association, the American Public Health Association and the American Heart Association and 10 other federal agencies were corralled into action under the chairmanship of James I. Cleeman (41, 42).
The stated goal of the NCEP was simple: “To reduce the prevalence of elevated blood cholesterol in the United States and thereby contribute to reducing coronary heart disease morbidity and mortality” (5, p.27-28). Which would of course be perfect if there was any evidence that the two parts of this sentence were causally related. What if there were other more effective methods to reduce CHD morbidity and mortality? Because, quite clearly, reducing blood cholesterol concentrations in all US citizens in whom the levels were elevated was going to be very, very, very expensive. And not without risk to the health of the US population.
Importantly as Moore emphasizes, the NCEP which would affect the lives of millions by placing them on medically-supervised diet and drug regimens not just in the US but around the world, was not authorized by the White House or the US Congress. Instead it was launched by the NHI “on its own authority, consulting mainly with special panels of hand-picked physicians” (5, p.29). Physicians who it needs to be added had significant conflicts of interest – who stood to gain both financially and career-wise from a national adoption of this national program to demonize cholesterol and to make paying customers of healthy individuals whose sole error was to have blood cholesterol concentrations greater than 200mg/dL (5.2mmol/L).
A key focus of the NCEP was to publicize the conclusions of the National Consensus Conference findings based on the (failed) LRC CPPT as if that trial was the definitive evidence that a dietary change that lowers blood cholesterol concentrations will reduced subsequent CHD risk. That the trial used a drug, not a dietary intervention, to lower blood cholesterol concentrations was simply ignored. As was the fact that the trial was a failure. Although the results may have achieved statistical significance using a water-down statistical metric, the finding was of no biological significance.
The main intervention promoted by the NCEP was the detection and management of blood cholesterol concentrations in the entire US population according to the conclusions of the Steinberg’s National Consensus Conference. Thus the published guidelines (43) “detail which patients should go on to have lipoprotein analysis, and which should receive cholesterol-lowering treatment on the basis of their low density lipoprotein (LDL)-cholesterol levels and status with respect to other coronary heart disease risk factors. Dietary therapy is the primary cholesterol-lowering treatment. The report specifies the LDL-cholesterol levels at which dietary therapy should be started and the goals of therapy, and provides detailed guidance on the nature of the recommended dietary changes. If, after six months of intensive dietary therapy, LDL-cholesterol exceeds specified levels, drug treatment should be considered” (43, p.36).
According to Steinberg, these guidelines “quickly became the gold standard on who to treat and how to treat. Many other countries followed the U.S. lead and convened expert groups to develop their own guidelines. Except for relatively minor differences in cut-off points, these were remarkably similar to those of the NCEP” (3, p.13).
The propaganda campaign unleashed by the NCEP was hugely successful. It increased the percentage of the public who thought that “reducing blood cholesterol levels would have a large effect on heart disease” from 64% in 1983 to 72% in 1986. The result was that by 1986, 46% of the public had had their blood cholesterol concentrations checked; 60% believed that reducing dietary fat and cholesterol were equally important; and 23% reported that they had made those dietary changes (44).
The effects on physicians’ opinions were even more impressive: The percentage of US physicians who thought that “reducing blood cholesterol levels would have a large effect on heart disease” rose from a paltry 39% in 1983 to 64% in 1986. Also by 1986, physicians were beginning to think that lowering blood cholesterol concentrations has preventive value although not yet as much as lowering blood pressure and quitting smoking.
By 1986, physicians had also lowered the blood cholesterol concentrations at which they would initiate either dietary therapy (from 6.7mmol/L in 1983 to 6.2mmol/L in 1986) or drug treatment (from 8.79mmol/L in 1983 to 7.76mmol/L in 1986) (45). (This despite the absence of evidence that diet determines the blood cholesterol concentrations of free-living individuals in the US (15,16)).
In 1997 the NCEP extended its reach to include young adults (46). It began by recommending that all adults below age 20 should have their blood cholesterol concentrations measured “as endorsed by representatives of more than 40 medical and health organizations” (p.1646). The basis for this evidence was a single study of 1017 medical graduates of Johns Hopkins Medical School between 1948-1964 who were followed for up to 40 years (47) and in whom the risk of CHD events was five times greater in those with the highest blood cholesterol concentrations when they were students, compared to those with the lowest. The risk of heart attack was also 9 times greater in those with the highest blood cholesterol concentrations.
The authors also cited the Framingham study linking higher risk for CHD events with increasing blood cholesterol concentrations but failed to acknowledge that, as in the Framingham data (figure 1 in reference 6), the quartile of subjects with the highest blood cholesterol concentrations would have included a number of medical students with familial hypercholesterolemia (FH). The majority of persons with FH are at increased risk of cardiovascular events that are only weakly related to their elevated blood cholesterol concentrations but are more likely linked to inherited blood clotting abnormalities and greater degrees of insulin resistance (48).
Also conveniently ignored was that, at the time the paper was written in 1997, there was no evidence that lowering elevated blood cholesterol concentration, last measured in medical students at age ~22 years, by either diet or drugs, will beneficially influence long term outcomes. Nor was any mention made of the findings from the MRFIT study already published in 1982 (4,6), showing an absence of any long term benefit from multiple interventions aimed at lowering long term CHD risk, even when blood cholesterol concentrations were successful reduced.
And let us not forget that in a paper published in 1986 (49) one of the authors of these guidelines, Dr Scott Grundy MD, had already grossly exaggerated the real risk of even large increase in blood cholesterol concentrations (figures 4 and 5 in reference (6).
Despite these serious scientific limitations, the success of the NCEP has been remarkable. Today anyone who does not believe that cholesterol is the single most important cause of heart disease, reversible with the prescription and use of statin drugs that are “powerful cholesterol-lowering drugs with robust mortality benefit” (50, p.1) – the latter statement is patently false (22,23) – is a member of a tiny, despised minority that is guilty of causing “significant harm to society and individuals…(as a result of) the wanton spread of medical misinformation. It is high time that we this stop, and we lay at the feet of the purveyors of internet and social media content the responsibility to fix this” (50, p.2). Presumably this is a call for the censoring of anything on social media and the internet that conflicts with the message of the drug manufacturers and which might harm their exorbitant profits.
The cited editorial was signed by the editors of 28 leading international cardiology journals indicating the seriousness with which the industry must feel the threat of the material presented in this book. So serious that it wishes to terminate debate on the topic.
Indeed this was exactly what Steinberg’s Consensus Conference had aimed to achieve, now 33 years ago. And this recent editorial (50) is a measure of how successful it has been, at least within this club of academic cardiologists.
The end result was that “a program (the National Cholesterol Education Program) that may have begun as sincere but misdirected zeal for the public good became intertwined with greed. The world was learning how much money could be made by scaring people about their cholesterol…one fact was therefore certain. The American public and the physician community would receive a hard sell on the dangers of cholesterol, an effort sure to drive people by the millions into the National Cholesterol Education Program…” (5, p.67).
And would the American public be the ultimate beneficiaries of this program?
“In fact, however, the enormous costs of the National Cholesterol Education Program are more insidious…the costs of this program are no less real. They are just better concealed. But the money for this effort will be collected on the authority of doctors, primarily for the benefit of drug companies, in a scheme engineered by a small group of men and women who mistakenly believe they are doing something good” (5, p.81).
1989. The initial results of Keys’ Minnesota Coronary Experiment are finally published
The results of Key’s first RCT intervention trial the Minnesota Coronary Experiment (MCE) (51), the planning of which had begun in 1968 and the conduct of which happened between 1968-1974, were first presented at a scientific conference in 1976. But then for reasons that have never been properly explained, the results were lost, only to be published in a rather brief report in a scientific journal in 1989. The result discussed in an earlier column (16) apparently showed that there was no difference in outcomes between the group receiving the altered diet in which saturated fat was partially replaced with corn oil, and the group that continued to eat its usual diet (Figure 9 in reference 15).
Thus as far as the world was concerned at the time, the MCE had failed to find any benefit from replacing dietary saturated fat with corn oil. Nor, apparently, had it detected any harm from this substitution.
But the true findings of that study would be hidden for a further 27 years until re-analysed and republished as the Recovered MCE (52) to be discussed subsequently.
Nina Teicholz makes the point that the senior author of the study Dr Ivan Franz, chose to publish the results of the MCE in a journal – Atherosclerosis, Thrombosis and Vascular Biology – that would likely be read only by cardiologists. She continues: “When asked why he did not publish the results earlier, Franz replied that he did not think he had done anything wrong in the study. ‘We were just disappointed in the way it came out’, he said. In other words, the study was selectively ignored by its own director. It was another inconvenient data point that needed to be dismissed” (53, p.96). Taubes makes the additional point that: “Proponents of Keys’ hypothesis who consider the Helsinki Mental Health study reason enough to propose a cholesterol-lowering diet for the entire nation, never cited the Minnesota Coronary Survey as a reason to do otherwise” (19).
Importantly, the Helsinki Mental Health study, also known as the Finnish Mental Hospital Study (54), was conducted in two mental hospitals near Helsinki between 1959-1971 amongst middle-aged men. Patients in one of the hospitals received a cholesterol-lowering diet low in saturated fats and relatively high in polyunsaturated fats whilst patients in the other hospital received standard hospital foods. After six years the diets in the two hospitals were reversed and the trial was continued for another six years. Blood cholesterol concentrations fell in patients when eating the cholesterol-lowering diet. The study found that the incidence of coronary heart disease was reduced in both hospitals by about 50% in the six year periods when the cholesterol lowering diet was eaten.
However there were significant issues with the trial (19, p.76-78). In particular the study was not randomized. Although male deaths at one hospital did go down dramatically during the intervention; this finding was not repeated in the second hospital. Nor was heart disease lessened in the women. But the largest problem was that the population under study was changing all the time with new admissions and releases happening at both hospitals throughout the 12 years of the trial. So this was not a legitimate randomized controlled trial although that is the way it is usually presented by those who use this study as further definitive support for Keys’ Diet-Heart hypothesis (55).
The study was criticized (56) as too statistically weak to be used in support of the Diet-Heart hypothesis. The MCE authors responded by saying that although the study design was imperfect, a perfect trial would be “so elaborate and costly…(that it) may perhaps never be performed….(so) we do not see any reason to change or modify our conclusions” (57, p.1267). As a result “The imperfect trial meanwhile would have to stand (as)..The research community accepted this ‘good-enough’ reasoning, and the Finnish Mental Hospital Study earned a spot as one of the linchpins of evidence for the diet-heart hypothesis” (53, p.77-78).
We return in due course to review the rather more disturbing findings of the RMCE (52).
1989- 2005. The results of Dr Robert Knoop’s Dietary Alternatives study and of the Boeing Employees Fat Intervention Trial (BeFIT) are published
Two studies published in 1977 – one from the Cooperative Lipoprotein Phenotyping Study (58) and the other from the Framingham Study (59) – both reported an inverse association between blood HDL-cholesterol concentrations and risk for developing CHD. Thus: “The finding that HDL cholesterol concentration is inversely related to subsequent development of CHD supports but does not prove the possibility that HDL elevation may prevent the development of CHD” (58, p.771) and “At these older ages the major potent lipid risk factor was HDL cholesterol which had an inverse association with the incidence of coronary heart disease (p<0.001) in either men or women” (59, p.707). The Framingham data also found that the association between LDL cholesterol and CHD risk was “weaker” (59, p.707). Importantly Keys’ key acolyte, William Castelli MD, was a co-author of both papers, whilst other acolytes, William Kannel and Thomas Dauber MD, co-authored the second paper (59).
These findings suggesting that blood LDL cholesterol concentrations are “weakly” associated with CHD risk and that blood HDL cholesterol concentrations are inversely related (i.e. “protective” against CHD) posed a major threat to Keys’ Twin Hypotheses. Particularly as the studies were published at the exact time in 1977 that Senator George McGovern’s US Senate Select Committee on Nutrition and Human Needs was about to release its Dietary Goals for the United States (60), strongly based on Keys’ hypotheses. In the end, as we shall see, this evidence was successfully suppressed and even today HDL-cholesterol does not have equal status with LDL-cholesterol as a predictor of future CHD risk.
Gary Taubes (19, p.166) notes that within 2 years the same authors had moderated their position and were again stating that risk of CHD in persons younger than age 50 is “strikingly related to the serum total cholesterol level” although they acknowledged that “at any level of cholesterol the risk varies widely, depending on a number of factors” (61, p.86) (Figure 1). Presumably if the risk varies widely depending on other risk factors, there is likely a combination of factors that largely negate any risk due to an elevated blood cholesterol concentration.
Legend to figure 1. Eight-year probability (per 1000 persons) of developing CHD in the Framingham population as related to different levels of five “risk” factors – increasing blood cholesterol concentrations (mg/dL); presence/absence of glucose intolerance; level of systolic blood pressure – normal (105mmHg) or elevated (195mmHg); use or avoidance of cigarettes; electrocardiographic (ECG) evidence of left ventricular (LV) hypertrophy (enlargement). Reproduced from figure 2 in reference 61.
Figure 1 shows that the risk of CHD rises increasingly with the addition of one or more risk factors – an increasing blood cholesterol concentration; glucose intolerance; hypertension; cigarette smoking and electrocardiographic evidence of left ventricular hypertrophy.
The figure also provides the calculation of exactly by how much the addition of these risk factors increases the risk for an individual with the different combinations of risk factors.
For example, an elevated blood cholesterol concentration alone carries a risk of 3.9 deaths/1000/8 years for an annual % death rate of 0.05%, that is 1 death in 2000 every year in those aged 35. In contrast those with the highest risk because they have hypercholesterolemia, glucose intolerance, systolic hypertension, who also smoke and who have left ventricular hypertrophy (the tallest column on the right of figure 1) experience 50.2 deaths/1000/8 years for an annual death rate of 0.63%; that is 1 death in every 161 persons every year.
The point is that although risk increases impressively with the presence of increased risk factors, for the individual, the real increase in risk is relatively trivial. And an elevated blood cholesterol concentration makes up only one of 5 variables contributing to that risk.
Furthermore with age, the importance of an elevated blood cholesterol concentration as a risk factor lessened so that “the strength of the association wanes progressively with advancing age, so that serum total cholesterol is no longer a predictor of risk in men beyond age 65 (59)” (61, p.86).
If the Framingham data seemed to show that the blood total cholesterol concentration was not as powerful a risk factor as Keys and his acolytes had hoped, the idea that diet-induced changes in blood HDL-cholesterol concentrations might offer additional information about CHD risk was taken up by Robert H. Knopp MD of the University of Washington.
In 1989 Knopp and his colleagues (62) began a series of studies of the effects of diets with reduced fat content on blood lipids. Their initial interest was to investigate whether or not people could “maintain such dietary changes over time” (63, p.1412). In time they would begin to question whether low-fat diets have detrimental effects on blood triglyceride and HDL-cholesterol levels, of the kind first recognized by Reaven and colleagues (30,64,65).
Their original studies (66) known as The Dietary Alternatives Study (DAS) evaluated the effects of four diets (Diets 1 to 4) differing in their fat and carbohydrate contents on blood lipid values in persons with either hypercholesterolemia alone (HC) or with hypercholesterolemia combined with hypertriglyceridemia (CHL).
In order, Diets 1 to 4 contained the following fat, carbohydrate and percentages:
- Diet 1: 30% fat; 53% carbohydrate; protein 17%
- Diet 2: 26% fat; 57% carbohydrate; protein 17%
- Diet 3: 22% fat; 61% carbohydrate; protein 17%
- Diet 4: 18% fat; 65% carbohydrate; protein 17%
The novel finding of their initial study (66), predictable from the work of Albrink, Man and Kuo (67), was that after one year on the diet, blood triglyceride concentrations increased significantly (13% and 20%) in HC subjects eating Diets 1 and 4 whilst decreasing somewhat in CHL subjects regardless of their carbohydrate intakes.
In 1995 they published the findings from the 2-year follow up of these subjects (68). It showed that in the HC group, a “sustained high carbohydrate diet can result in a continuous increase” in blood triglyceride concentrations (68, p.993) whereas the effect was much less in CHL subjects. They concluded that a low-fat high-carbohydrate diet can be used in the management of both HC and CHL.
To extend the study the authors next focused on the effects of the diet on blood lipid measures other than solely blood triglyceride concentrations (69). By then they had become interested in other findings that they considered incompletely studied. They noted that the National Cholesterol Education Program advocated diets that restricted dietary fat intake to less than 30% of total calories with saturated fat contributing less than 10% (Step I diet) or less than 7% (Step 2 diet). But even more fat-restrictive diets have been advocated by, amongst others, Dean Ornish MD (70), the American Heart Association’s Nutrition Committee and Council on Atherosclerosis (27), and Scott M. Grundy MD and colleagues (71).
Whilst it might be intuitively obvious that progressively lower dietary intakes of total fat, saturated fat, and cholesterol must produce progressively lower blood cholesterol concentrations, they noted that the evidence in the literature at that time did not support that conclusion (71,72).
For example the 2 month cross-over design feeding study of Grundy et al. (71) failed to show that a 20% fat diet produced lower blood cholesterol concentrations than did a diet containing 30% fat. Nor were blood cholesterol concentrations different in subjects with peripheral vascular disease eating either the American Heart Association (AHA) diet for the treatment of hyperlipidemia (31% fat; 49% carbohydrate) or the Pritikin Maintenance Diet (14% fat; 64% carbohydrate) for 12 months (72).
Knopp et al. (69) concluded that there are few comparative data proving that a lower dietary fat intake produces lower blood cholesterol concentrations in free-living individuals. Recall that the Framingham study failed to find any dietary variable that predicted the blood cholesterol concentration in a free-living population of middle-aged men (6,15).
To address these questions, over a period of 18 months, Knopp and his colleagues (69) screened 8372 men working at the Boeing factory in Seattle, Washington and selected the 444 men with the highest blood cholesterol concentrations for further study. The 444 selected men with hypercholesterolemia were then sub-divided into two groups as before – those with (combined hyperlipidemia – CHL) and those without an associated hypertriglyceridemia (hypercholesterolemia alone – HC). Approximately equal numbers of subjects (50-70 per group) were then assigned to one of the four diets that differed in their fat and carbohydrate contents and which they had used in their earlier studies (66,68,69). As in the previous studies, the HC group was randomized to all 4 diets; the CHL group to Diets 1-3.
Notably the study was one, not just of reducing dietary fat intakes, but also of increasing dietary carbohydrate intakes.
The initial study reported that blood total cholesterol, low density lipoprotein and apoprotein B concentrations all fell significantly in the HC group whereas triglyceride levels increased progressively with dietary fat restriction (figure 2). Responses were more variable in the CHL group.
Legend to figure 2: The Dietary Alternatives Study evaluated the effects of 4 different diets with progressively lower total fat content (Diets 1 – 4) on blood total cholesterol, apoprotein B, LDL-cholesterol and triglyceride concentrations. Reproduced from figure 1 in reference 69.
Figure 2 shows that whereas the increasingly fat-restricted diets all lowered blood total cholesterol, LDL-cholesterol and Apoprotein B concentrations somewhat, this did not appear to be a graded effect; responses were not uniformly greater as the dietary fat content was reduced from 30% (Diet 1) to 18% of daily energy intake (Diet 4). Indeed in all diet groups the blood total cholesterol concentrations remained elevated in persons with hypercholesterolemia despite the low-fat diets.
Thus the range of average blood cholesterol concentrations in the 7 different HC and CHL groups after 12 months of the dietary intervention ranged from 5.8-6.5mmol/L, compared to a range of 6.2-6.9mmol/l at the start of the trial. Thus the best the lowest (18%) fat diet could achieve was to lower average blood cholesterol concentrations by ~0.4mmol/L or about 7%.
In contrast in the HC group, blood triglyceride concentrations were markedly elevated in those eating Diets 3 and 4 with the highest carbohydrate intakes. The authors were unequivocal in their conclusions: “These findings indicate that hypertriglyceridemia is induced by an aggressively fat-restricted, high-carbohydrate diet and that the elevation persists long term” (69, p.1514). Exactly as Gerald Reaven MD had warned (64,65).
Recall for the moment that in 1984 the hubristically termed Nutrition Committee and the Council on Atherosclerosis of the American Heart Association (27) was advocating low-fat high-carbohydrate diets for the management of hypertriglyceridemia.
But more troublingly and not reproduced in graphical form in any part of the report was the finding that blood HDL-cholesterol concentrations fell significantly in the HC group eating the highest carbohydrate diets, Diets 3 and 4.
Since at the time there was enough evidence to suggest that elevated blood HDL-cholesterol concentrations are associated with reduced CHD risk (58,59), this finding should have raised the alarm. Particularly since the fall in HDL-cholesterol on the high-carbohydrate diets was associated with increased blood triglyceride concentrations and therefore an adverse change in the HDL-cholesterol/Triglyceride ratio.
So whilst the authors warned about the undesirable effects of the high-carbohydrate diet on blood triglyceride concentrations, they chose not to highlight the adverse blood HDL-cholesterol responses to the low-fat, high-carbohydrate diet.
Their main conclusion was that restricting dietary fat intake to less than 26% offers “little further advantage” and may produce “undesirable effects in HC subjects” (69, p.1509).
Subsequently Knopp and his colleagues turned their attention to possible gender differences in this response to the low-fat high-carbohydrate diet and initiated the BeFIT study.
For this study (73) 9770 male and female Boeing employees were screened and again those with hypercholesterolemia (HC) or combined hyperlipidemia (CHL) were invited to participate in a one-year trial. Four hundred and nine subjects were randomized to an immediate and delayed intervention group; the delayed group received the dietary intervention only 6 months after the immediate group had begun their intervention. The dietary intervention was the Step II diet used in the previous studies and the division into HC and CHL was as before.
The study found that in both men and women, total cholesterol and LDL-cholesterol concentrations fell on the diet. But only in women were blood HDL-cholesterol concentrations significantly reduced on the Step II diet. Thus they identified a gender difference in the response of men and women with HC or CHL to a low-fat diet.
Since low HDL-cholesterol concentrations (and elevated blood triglyceride concentrations) are independent risk factors in women eating a standard US diet (59,74) and since other studies have also shown the same response to a low-fat diet in women with normal blood cholesterol concentrations (75), they noted: “The explanation and significance of the HDL decrease (in women) is unknown, and more work is needed to determine the factors responsible, whether the decrease is sustained, and whether it is deleterious…”. “Low HDL cholesterol is an independent CVD risk factor in women consuming Western diets (59,74,76). Whether alternative dietary regimens that might maintain higher HDL cholesterol levels would offer long-term advantages is unknown” (73).
Their next paper (77) in the beFIT series confirmed the gender difference in the response to low-fat diets. Whereas men and women showed equivalent (~5%) reductions in blood total cholesterol and LDL-cholesterol concentrations to the low-fat diet eaten for 6 months, women showed greater reductions in HDL-cholesterol (8%) and in HDL2-cholesterol (17%) concentrations. ApoA-1 concentrations decreased only in HC women. Blood triglyceride concentrations increased in all groups and there was no specific sex effect. These differences were not due to gender-based differences in dietary compliance and therefore appear to be truly gender-based.
The authors again cited numerous studies showing that low HDL cholesterol concentrations are associated with increased CHD risk and concluded the paper with the statement: “the NCEP Step II diet can provide long-term benefit to cardiovascular health by lowering LDL-C (provided one believes that LDL-C has any role in determining cardiovascular ill-health – my addition). However, alternative diets that lower LDL-C but do not reduce HDL-C should be evaluated and may prove preferable for HC women” (p.1585).
On the basis of these studies Knopp et al. concluded that in men “fat intakes below about 25% total energy and carbohydrate intakes above ~60% total energy yield no further LDL-C lowering in HC and CHL male subjects and can be counterproductive to triglyceride, HDL-C and apo B levels. This lack of benefit appears to be explained by enhanced exogenous synthesis of palmitic acid, which negates the benefit of further saturated fat restriction. The HDL-C decrease in women may have a similar cause and points to an underlying male-female difference. Alternative dietary approaches to limit saturated fat intake deserve intensive study” (78, p.191-192).
In a subsequent review article Knopp et al. (79,80) concluded: “Most consistent are greater changes in high-density lipoprotein (HDL), HDL2, and apolipoprotein A-I levels in women compared with men with high-carbohydrate or high-fat feeding. Dietary fat restriction in women appears to have a less beneficial lipoprotein effect than in men. Dietary fat restriction for heart disease prevention may be less ideal in women than in men” (p.472).
Thus: “The greater HDL reductions in women with fat restriction could have deleterious consequences for the concentration-dependent antioxidant, anti-inflammatory and anticytokine effects of HDL… In addition, HDL and triglycerides are stronger mediators of cardiovascular risk in women than in men (81,82). Alternative dietary interventions in women should be explored to minimize the tendency of low-fat, high-carbohydrate diet in women to raise triglyceride and lower HDL cholesterol more and lower LDL less than in men. Such a diet might be comprised of a higher amount of allowable fat but still limited in saturated and trans fatty acids and lower in carbohydrate…” (79, p.477).
In short the authors were implying that a low-fat diet may not be ideal for women – a message that has simply not made any impression on the lore of cardiology.
A further group of studies (83-85) published since the meta-analysis of Cobb and colleagues (75) confirm that the blood lipid response of men and women to low-fat diets are not identical. Thus: “The impact of these potentially atherogenic changes in response to reducing dietary total fat and saturated fat must be weighed against the clearly demonstrated benefits of reducing LDL cholesterol levels (86)…” (84, p.448).
Perhaps the authors could have chosen a more convincing, more independent body of evidence than the NCEP Consensus (86) to support their conclusion that lowering blood LDL cholesterol concentrations is beneficial to health.
But the message that women respond differently from men to a low-fat diet has been so successfully hidden that, until I was directed to these articles by Nina Teicholz’s book (53, p.159-166), I had never heard this matter raised in any academic circles.
1991. The results of the Helsinki Businessmen Study are published
The first results, representing a 15-year follow-up of this study conducted between 1974 to 1980, were reported in 1991. The goal of the study was to determine whether lifestyle modification, in particular stopping smoking, achieving an ideal weight through adoption of a “healthy” diet, moderation of alcohol use, and active drug treatment of those with hypertension or hyperlipidemia that did not respond to diet, would reduce the subsequent development of CHD in a group of Helsinki businessmen. The control group received the standard medical care. Persons in the intervention group also received 4-monthly visits from the investigators. At the end of 5 years, subjects were informed that they should continue doing as they had during the 5 years of the trial.
As presented previously (6), the 1991 findings (87) came as something of a shock to the investigators. Although the prevalence of coronary risk factors was reduced by 46% in the intervention group at the end of the 5 year trial, 5 years later the coronary risk profile and medication use were similar in both the intervention and control groups. In the subsequent study period the 67 deaths in the intervention group was significantly greater by 142% than the 46 deaths in the control group. Cardiac deaths were also significantly greater in the intervention group (34 vs 14) as were deaths due to violence (13 vs 1). As a result, cumulative numbers of total deaths and of heart disease deaths became greater in the intervention group by 45% already 5 years after the beginning of the trial (figure 7 in reference 6).
A subsequent report (88) tried to paint a somewhat different story than this dismal picture which “may even have led to undue pessimism about reducing coronary risk factors (89)” (88, p.449). Which is of course, yet more proof that the researchers were searching for a predetermined outcome. And when they failed to find that result, they went searching for a way around this “pessimistic” finding. But science is not meant to be optimistic or pessimistic. It’s meant to be unemotive and independent.
For that report the authors analysed CHD outcomes over 18 years of follow-up in the total study group, including all members of both the intervention and control groups. Whilst acknowledging that the new analysis continued to show that a higher mortality in the intervention than in the control group, yet “the present analysis of the background population of 3313 men gives an important new perspective on this outcome” (88, p.452). The re-analysis of the data produced the graph shown in figure 3.
Legend to figure 3. Eighteen year survival curves in 5 different groups in the Helsinki Businessmen study. Note that survival was worse in the intervention group than in the control group but was slightly improved in the group with the fewest number of coronary risk factors. Reproduced from figure 2 in reference 88.
Figure 3 shows that 18-year survival was best in the group with the lowest prevalence of risk factors (blue line); was not quite as good in the control group (yellow line); was worse in the intervention group (red line); still worse in those who had refused to participate in the trial (green line); and was the very worst in the group that was excluded from entering the trial because they had pre-existing disease, or were already using medications, or for other reasons (purple line).
The authors drew the conclusion that this study “shows that the traditional cardiovascular risk factors (89) are also important predictors of mortality among men of the highest social class” (88, p.453). And interestingly “The findings on 1 hour blood glucose suggests that factors related to glucose tolerance explain in part the excess mortality in the intervention group compared with the control group” (p.449).
In essence the logical trickery used by the authors is that in an attempt to invalidate data from a randomized controlled trial that produced results they did not like, they instead converted the study into an observational, associational study.
But associational studies cannot prove causation because of residual confounding factors present in the different groups. The confounders in this study are only too apparent. The Low Risk group were likely low risk because they chose to be that way. And that choice likely influenced their superior survival in ways that may have had nothing to do with the traditional cardiovascular risk factors. Or that is the way these findings should correctly be interpreted with the appropriate caution.
Thus persons in the low risk group were leaner which raises the question of whether they either ate or exercised differently; they had lower 1 hour blood glucose concentrations during a glucose tolerance test as well as lower fasting triglyceride concentrations – both suggesting they were less insulin-resistance than persons in the other 4 groups; and less than 25% were smokers, compared to >43% of smokers in all the other groups. Their reduced smoking prevalence alone could have explained their superior survival, independent of their other more favourable risk factor status.
In summary, the Helsinki Businessmen study found that reducing coronary risk factors by methods that included prescription drug treatment of hypertension and hyperlipidemia worsened outcomes in middle-aged men of high social class who were initially free of CHD.
That is the sole scientifically-valid conclusion from this study although the usual spin was placed on the study since (we, the anointed Keysian acolytes know) “the totality of evidence indicates that intervention on coronary risk factors will in fact decrease risk of cardiovascular disease” (90, p.1268).
Actually no! The best evidence from the most detailed study ever undertaken in the most at risk population – the MRFIT (4,6) – found that that altering coronary risk factors did not alter CHD mortality. That results of that study had been published already a decade before the editorial of Paul and Hennekens (90) but apparently they were unaware of it. Or at least, they had yet to comprehend its meaning – a state known as Wilful Blindness (91).
Others strongly criticized the authors’ conclusions. Bassler and Bassler (92) argued that “a trivial and temporary lowering of serum cholesterol concentrations doomed the subjects to long-term, increased mortality”. They continued: “Sixty million Americans have been targeted for cholesterol reduction. By projecting the Finland results onto this group, we estimate that 2 million lives could be lost if treatment is prescribed. And this increased mortality will not come cheap. The cost of cholesterol-lowering drugs for 60 million people will be about $120 billion each year – a poor use of health care dollars. These are issues that should have been addressed by the Editorial. Millions of lives and billions of dollars are at risk if we ignore the results from Finland” (p. 2183).
Newman et al. (93) were concerned lest the “striking increase in violent deaths should be swept under the rug” (93, p.2183). Instead they concluded that the relationship had to be considered causal since it has been repeatedly identified by randomized trials – “It is time to examine the implications of this consistently observered relationship for health policy rather than awaiting additional studies” (p.2183).
So if Paul and Henneken (90) were indeed guilty of Wilful Blindness and sweeping these inconvenient findings “under the rug”, would they also be able to ignore the evidence from the $700 million Women’s Health Initiative Randomized Controlled Dietary Modification Trial that was on the verge of being launched at the time of their editorial?
1993 – 2005. The Women’s Health Initiative Randomized Controlled Dietary Modification Trial (WHIRCDMT) is initiated
The goal of the Women’s Health Initiative Randomized Controlled Dietary Modification Trial (WHIRCDMT) (94) funded and directed by the NHLBI, was to determine whether a population of older women who adopted the USDA Dietary Guidelines for Americans would reduce their risks of colo-rectal and breast cancers and suffer less from coronary heart disease.
For the study 48 835 post-menopausal women were encouraged either to adopt the USDA Dietary Guidelines by reducing their fat intake and eating more vegetables and grains or to continue eating their usual diet. Women in the intervention group also received an “intensive behavioural modification program” comprising 18 group sessions in the first year followed by quarterly maintenance sessions for the next 7 years. The control group received only a copy of Dietary Guidelines for Americans. As a result any positive outcomes in the intervention group could not be ascribed purely to dietary change since the intervention group received additional interventions not shared by the control group. A similar criticism exists for the MRFIT study (4,6).
Clearly both trials were established to show that “throwing the kitchen sink” at humans in terms of all the “known” coronary risk factors might produce a positive outcome. Then the Diet-Heart acolytes would be free to conclude that it was solely the dietary change, and not the other simultaneous interventions, that produced the positive outcomes.
The first results of the effects of the intervention on CHD outcomes were reported in 2005 as described subsequently.
- Oliver MF, Geizerova H, Gyarfas I, et al. A co-operative trial in the primary prevention of ischaemic heart disease using clofibrate. Report of the Committee of Principal Investigators. Brit Heart J 1978;40:1069-1118.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 6. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-6-1960-1967
- Steinberg D. An interpretive history of the cholesterol controversy, part IV: The 1984 Coronary Primary Prevention Trial ends it – almost. J Lipid Res 2006;47:1-14.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 11. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-11-1979-1984
- Moore TJ. Heart Failure: A Critical Inquiry into American Medicine and the Revolution in Heart Care. New York, NY: Simon and Schuster, 1989.
- Noakes TD. It’s the insulin resistance, stupid. Part 9. CrossFit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-9
- Diamond DM, Alabdulgader AA, de Logeril M, et al. Dietary recommendations for Familial Hypercholesterolemia: an evidence-free zone. BMJ Evidence-Based Medicine Epub ahead of print: doi:10.1136/ bmjebm-2020-111412
- (Dr) AM Gotto MD “is on the board of Directors for Aegerion and Arisaph; has been a consultant for AstraZeneca, Janssen, Kowa, Merck, Pfizer and Roche; and has served on advisory boards for DuPont and Vatera Capital”. From: Fruchart J-C, Davignon J, Hermans MP, et al. Residual macrovascular risk in 2013: what have we learned? Cardiovasc Diabetol 2014;13:26
- “Dr. Antonio Gotto, Jr., former American Heart Association president, sent a letter promoting Puritan Oil to practicing physicians – printed on Baylor College of Medicine, The De Bakey Heart Center letterhead. The irony of Gotto’s letter is that De Bakey, the famous heart surgeon, co-authored a 1964 study involving 1700 patients which also showed no definite correlation between serum cholesterol levels and the nature and extent of coronary artery disease.10 In other words, those with low cholesterol levels were just as likely to have blocked arteries as those with high cholesterol levels. But while studies like De Bakey’s moldered in the basements of university libraries, the vegetable oil campaign took on increased bravado and audacity”. From: Enig MG, Fallon S. The Oiling of America. Part 1. Nexus December 1998-Jauary 1999:6(1):19-23;80-82. Available at: http://cruinthe.tripod.com/nexus/articles/oilingamerica.1.html.
- Kolata G. Heart Panel’s conclusions questioned. Science 1985;227:40-41.
- Steinberg D. The cholesterol controversy is over. Why did it take so long? Circulation 1989;80:1070-1078.
- Steinberg D. An interpretive history of the cholesterol controversy, part III: mechanistically defining the role of hyperlipidemia. J Lipid Res 2005;40:2037-2049.
- Steinberg D. The Cholesterol Wars. The Skeptics vs the Preponderance of Evidence. Academic Press, London, UK. 2007
- Anon. Lowering blood cholesterol to prevent heart disease. JAMA 1985;253:2080-2086.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 5. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-5
- Noakes TD. It’s the insulin resistance, stupid. Part 8. CrossFit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-8
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 7. The Noakes Foundation. https://thenoakesfoundation.org/news/blog/ancel-keys-cholesterol-con-part-7-1968-1970
- Ahrens EH, Blanckenhorn DH, Tsaltas TT. Effect on human serum lipids of substituting plant for animal fat in diet. Proc Soc Exp Biol Med 1954;86:872-878.
- Taubes G. Good calories bad calories. Fats, carbs, and the controversial science of diet and health. Anchor Books, New York, NY. 2008.
- Jacobs DR, Anderson JT, Blackburn H. Diet and serum cholesterol. Do zero correlations negate the relationship? Am J Epidemiol 1979;110:77-87.
- Sacks FM, Lichtenstein AH, Wu JHY, et al. Dietary fats and cardiovascular disease. A Presidential Advisory from the American Heart Association. Circulation 2017;136:e1-e23.
- DuBroff R, de Lorgeril M. Fat or fiction: The diet-heart hypothesis. BMJ Evidence-Based Medicine Published Online First: 29 May 2019. doi: 10.1136/bmjebm-2019-111180
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 10. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-10-1978-1979
- Ravnskov U, de Lorgeril M, Diamond DM, et al. LDL-C does not cause cardiovascular disease: a comprehensive review of the current literature. Expert Rev Clin Pharmacol 2018;11:959-970.
- American Heart Association Nutrition Committee. Rationale of the Diet-Heart Statement of the American Heart Association. Arteriosclerosis 1982;4:177-191.
- Keys A, Anderson JT, Mickelsen O, et al. Diet and serum cholesterol in man: Lack of effect on dietary cholesterol. J Nutr 1956;59;39-56.
- Ad Hoc Committee to Design a Dietary Treatment of Hyperlipoproteinemia. AHA Special Report. Recommendations for treatment of hyperlipidemia in adults. A Joint Statement of the Nutrition Committee and the Council on Arteriosclerosis. Circulation 1984;69:1067A-1090A.
- Gofman JW. Diet in the prevention and treatment of myocardial infarction. Am J Cardiol 1958;1:271-283.
- Frederickson DS. An international classification of hyperlipidemias and hyperlipoproteinemias. Ann Intern Med 1971;75:471-472.
- Noakes TD. It’s the insulin resistance, stupid. Part 2. CrossFit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-2
- Grundy SM. Effects of polyunsaturated fats on lipid metabolism in patients with hypertriglyceridemia. J Clin Invest 1975;55:269-282.
- Sturdevant RAL, Pearce ML, Dayton S. Increased prevalence of cholelithiasis in man ingesting a serum-cholesterol-lowering diet. N Eng J Med 1973;288:24-27
- King ME, Spector AA. Effect of specific fatty acyl enrichments on membrane physical properties detected with a spin label probe. J Biol Chem 1978;253:6493-6501.
- Carroll KK, Gammal EB, Plunkett ER. Dietary fat and mammary cancer. Can Med Assoc 1968;98:590-594; Carroll KK, Khor HT. Effects of level and type of dietary fat on incidence of mammary tumors induced in female Sprague-Dawley rats by 7,12 dimethylbenz (alpha) anthracene. Lipids 1971;6:415-420; Gammal EB, Carroll KK, Plunkett ER. Effects of dietary fat on mammary carcinogenesis by 7,12-dimethylbenz (alpha) anthracene in rats. Cancer Res 1967;27:1737-1742.
- Central Committee for Medical and Community Program of the American Heart Association. Dietary Fat and its relation to heart attacks and strokes. JAMA 1961;175:389-391.
- Willett WC. The role of dietary n-6 fatty acids in the prevention of cardiovascular disease. J Cardiovasc Med 2007;8(suppl 1):S42-S45.
- Ahrens – The Diet-Heart question in 1985: Has it is really been settled? Lancet 1985;1:1085-1087.
- Ahrens EH. Dietary fats and coronary heart disease: unfinished business. Lancet 1979;2:1345-1348.
- Oliver MF. Consensus or Nonsensus Conferences on coronary heart disease. Lancet 1985;1:1087-1089.
- Moore TJ. The cholesterol myth. The Atlantic.1989;264(September):37
- Cleeman JI, Lenfant C. The National Cholesterol Education Program. Progress and prospects. JAMA 1996;280:2099-2104.
- Cleeman JI. National Cholesterol Education Program. Overview and education activities. J Reproduct Med 1989;34:716-728.
- The Expert Panel. Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Arch Intern Med 1988;148:36-89.
- Schucker B, Bailey K, Heimbach JT, et al. Change in public perspective on cholesterol and heart disease. Results from two national surveys. JAMA 1987;258:3527-3531.
- Schucker B, Wittes JT, Cutler JA, et al. Change in physician perspective on cholesterol and heart disease. Results from two national surveys. JAMA 1987;258:3521-3526.
- Cleeman JI, Grundy SM. National Cholesterol Education Program recommendations for cholesterol testing in young adults. A science-based approach. Circulation 1997;95:1646-1650.
- Klag MJ, Ford DE, Mead LA, et al. Serum cholesterol in young men and subsequent cardiovascular disease. N Engl J Med 1993;328:313-318.
- de Isla LP, Alonso R, Mata N, et al. Predicting cardiovascular events in familial hypercholesterolemia. The SAFEHEART (Spanish Familial Hypercholesterolemia Cohort Study. Circulation 2017;135:2133-2144; Gaudet K, Vohl M-C, Perron P, et al. Relationships of abdominal obesity and hyperinsulinemia to angiographically assessed coronary artery disease in men with known mutations in the LDL receptor gene. Circulation 1998;97:871-877; Ravnskov U, de Lorgeril M, Kendrick M, et al. Inborn coagulation factors are more important cardiovascular risk factors than high LDL-cholesterol in familial hypercholesterolemia. Med Hypoth 2018;121:60-63.
- Grundy SM. Cholesterol and coronary heart disease. A new era. JAMA 1986;256:2849-2858
- Hill JA, Agewall J, Baranchuk A, et al. Medical misinformation: Vet the Message! Eur Heart J 2019;40:404–405.
- Frantz ID, Dawson EA, Ashman PL, et al. Test of effect of lipid lowering by diet on cardiovascular risk. The Minnesota Coronary Survey. Arteriosclerosis 1989;9:129-135.
- 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 2016;353:i1246..
- Teicholz N. The Big Fat Surprise. Why butter, meat and cheese belong in a heathy diet. Simon and Schuster, New York, NY. 2014.
- Turpeinen O, Karvonen MJ, Pekkarinen M, et al. Dietary prevention of coronary heart disease: The Finnish Mental Hospital Study. Int J Epidemiol 1979;8:99-118.
- Mozaffarian D, Micha R, Wallace S. Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat: A systematic review of meta-analysis of randomized controlled trials. PLOS Medicine 2010. Published March 23rd 2010. Available at: https://doi.org/10.1371/journal.pmed.1000252
- Halperin M, Cornfield J, Mitchell SC. Effect of diet on coronary heart disease mortality. Lancet 1973;302:438-439.
- Mietinnen M, Turpeinen O, Karvonen MJ, et al. Effect of diet on coronary heart disease mortality. Lancet 1973;302:1266-1267.
- Castelli WP, Doyle JT, Gordon T, et al. HDL cholesterol and other lipids in coronary heart disease. The Cooperative Lipoprotein Phenotyping Study. Circulation 1977;55:767-772.
- Gordon T, Castelli WP, Huortland MC, et al. High density lipoprotein as a proctective factor against coronary heart disease. The Framingham study. Am J Med 1977;62:707-714,
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 9. The Noakes Foundation. http://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-9-1976-1977
- Kannel WB, Castelli WP, Gordon T. Cholesterol in the prediction of atherosclerotic disease: New perspectives based on the Framingham study. Ann Intern Med 1974;90:85-91.
- Knopp RH, Walden CE, McCann BS, et al. Effects of alternating high carbohydrate, low fat diets on plasma triglycerides in hypercholesterolemic and combined hyperlipidemic men. Clin Res 1989;37:594A (abstr).
- McCann BS, Retzlaff BM, Dowdy AA, et al. Promoting adherence to low-fat, low-cholesterol diets: review and recommendations. J Am Diet Assoc 1990;90:1408-1414.
- Coulson AM, Hollenbeck CB, Swislocki ALM, et al. Deleterious metabolic effects of high-carbohydate, sucrose-containing diets in patients with non-insulin-dependent diabetes mellitus. Am J Med 1987;82:213-220.
- Coulson AM, Hollenbeck CB, Swislocki ALM, et al. Persistence of hypertriglyceridemic effects of low-fat high-carbohydrate diets in NIDDM patients. Diabetes Care 1989;12:94-101.
- Walden CE, McCann BS, Retzlaff B, et al. Alternative fat-restricted diets for hypercholesterolemia and combined hyperlipidemia: Feasibility, design, subjects recruitment, and baseline characteristics of the Dietary Alternative Study. J Am Coll Nutr 1991;5:429-422.
- Noakes TD. It’s the insulin resistance, stupid. Part 1. CrossFit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-1
- Retzlaff B, Walden CE, Dowdy AA, et al. Changes in plasma triacylglycerol concentrations among free-living hyperlipidemic men adopting different carbohydrate intakes over 2 y: The Dietary Alternatives Study. Am J Clin Nutr 1995;62:988-995.
- Knopp RH, Walden CE, Retzlaff BM, et al. Long-term cholesterol-lowering effects of 4 fat-restricted diets in hypercholesterolemic and combined hyperlipidemic men. The Dietary Alternatives Study. JAMA 1997;278:1509-1515,
- Ornish D, Brown SE, Schwerwitz LW, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-133.
- Grundy SM, Nix D, Whelan MF, et al. Comparison of three cholesterol-lowering diets in normolipidemic men. JAMA 1986;256:2351-2355.
- Brown GO, Whyte L, Gee MI, et al. Effects of two “lipid-lowering” diet on plasma lipid levels of patients with peripheral vascular disease. J Am Diet Assoc 1984;84;546-550.
- Walden CE, Retzlaff BM, Buck BL, et al. Lipoprotein lipid response to the National Cholesterol Education Program Step II Diet by hypercholesterolemic and combined hyperlipidemic women and men. Arteriosclerosis 1997;17:375-382.
- Bass KM, Newschaffer CJ, Klag MJ, et al. Plasma lipoprotein levels as predictors of cardiovascular death in women. Arch Intern Med 1993:153:2209-2216.
- Cobb M, Greenspan J, Timmons M, et al. Gender differences in lipoprotein responses to diet. Ann Nutr Metab 1993;37:225-236.
- Bush TL, Criqui MH, Cowan LD, et al. Cardiovascular disease mortality in women: results from the Lipid Research Clinics Follow-up Study. In: Eaker EE, Packard B, Wenger NK, et al., eds. Coronary Heart Disease in Women. New York, NY: Maymarket Doyma Inc; 1987:106-111.
- Walden CE, Retslaff B, Buck BL, et al. Differential effect of National Cholesterol Education Program (NCEP) Step II Diet on HDL cholesterol, its subfractions, and Apoprotein A-1 levels in hypercholesterolemic women and men after 1 year. The befit study. Arterioscler Thromb Vasc Biol 2000;20:1580-1587.
- Knopp RH, Retslaff B, Walden C, et al. One-year effect of increasingly fat-restricted, carbohydrate-enriched diets on lipoprotein levels in free-living individuals (445564E). Proc Exp Biol Med 2000;225:191-199.
- Knopp RH, Paramsothy P, Retzlaff BM, et al. Gender differences in lipoprotein metabolism and dietary response: Basis in hormonal differences and implications for cardiovascular disease. Curr Atheroscler Rep 2005;7:472-479.
- Knopp RH, Paramsothy P, Retzlaff BM, et al. Sex differences in lipoprotein metabolism and dietary response: Basis in hormonal differences and implications for cardiovascular disease. Curr Cardiol Rep 2006;8:452-459.
- Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79:8-15.
- Reardon MF, Nestel PJ, Craig IH, et al. Lipoprotein predictors of severity of coronary artery disease in men and women. Circulation 1985;71:881-888.
- Stefanik ML, Mackey S, Sheehan M, et al. Effects of diet and exercise in men and postmenopausal women with low levels of HDL cholesterol and high levels of LDL cholesterol. N Engl J Med 1998;339:12-20.
- Ginsberg HN, Kris-Ethertoon P, Dennis B, et al. Effects of reducing dietary saturated fatty acids on plasma lipids and lipoproteins in healthy subjects. The Delta Study, Protocol 1. Arterioscler Thromb Vasc Biol 1998;18:441-449.
- Li Z, Otvos JD, Lamon-Fava S, et al. Men and women differ in lipoprotein response to dietary saturated fat and cholesterol restriction. J Nutr 2003;133:3428-3433.
- National Cholesterol Education Program. Second report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). Circulation 1994;89:1335-1445.
- Strandberg TO, Salomaa VV, Naukkariner VS, et al. Long-term mortality after 5-year multifactorial primary prevention of cardiovascular diseases in middle-aged men. JAMA 1991;266:1225-1229.
- Strandberg TO, Salomaa VV, Vanhanen HT, et al. Mortality in participants and non-participants of a multifactorial prevention study of cardiovascular diseases: a 28 year follow up of the Helsinki Businessmen Study. Br Heart J 1995;74:449-454.
- Manson JE, Tosteson H, Ridker PM, et al. The primary prevention of myocardial infarction. N Engl J Med 1992;326:1406-1416.
- Paul O, Hennekens CH. The latest report from Finland. A lesson in expectations. JAMA 1991;266:1267-1268.
- Heffernan M. Wilful Blindness. Why we ignore the obvious at our peril. Simon and Shuster, New York, NY, 2011.
- Bassler TJ, Bassler TJ. Long-term mortality after primary prevention for cardiovascular disease. JAMA 1992;267:2183
- Newman TB, Browner WS, Hulley SB. Long-term mortality after primary prevention for cardiovascular disease. JAMA 1992;267:2183-2184.
- Rossouw JE, Finnegan LP, Harlan WR, et al. The evolution of the Women’s Health Initiative: perspectives from the NIH. JAMA 1995;50:50-55.
About the author
Professor Tim Noakes has dedicated his life to the pursuit of knowledge and undoing the last 50 years of ‘bad’ nutritional science. His aim is to fix the future outlook of human health, by changing the way people eat and the food policies to enable the change.
Prof. Noakes has published more than 750 scientific books and articles. He has been cited more than 19 000 times in scientific literature, has an H-index of 71 and has been rated an A1 scientist by the National Research Foundation of South Africa for a second 5-year term. He has won numerous awards over the years and made himself available on many editorial boards.