[2017] Lore of Nutrition: Challenging Conventional Dietary Beliefs

by Tim Noakes

  As will become apparent, the Naudé review and Volmink’s confident statement concluding the Centenary Debate are evidence linking these four professors (Rossouw, Volmink, Steyn and Levitt), the debate, and senior colleagues in the UCT and Stellenbosch medical faculties to the sequence of actions that culminated in the HPCSA hearing against me.

  Surprisingly, on 24 January 2013, I received a personally signed letter from Levitt and Steyn thanking me for making the debate ‘a very good evening of academia’. They continued: ‘The standard you set was very high and extremely important in South Africa at this point in time.’ However, these words were contradicted a week later with the publication of Bateman’s SAMJ report on the debate. I still had much to learn about the spiritual and ethical darkness into which my profession had descended in the past four decades.

  ‘Experts warn against the Noakes diet’. Under this headline, published within 10 weeks of the Centenary Debate, Health24 carried an article in which UCT Faculty of Health Sciences professor of medicine (now dean) and Volmink’s long-time friend Bongani Mayosi stated: ‘There will be no change in how students at UCT are taught about the risk factors for heart disease. It is clear to me that the diet-heart hypothesis is alive and well and we do not see any reason to change what we teach our students about the link between an unhealthy diet, high blood cholesterol and heart disease.’53

  Importantly, in a 2012 scientific publication, Mayosi had written that ‘Ancel Keys proved unequivocally that cholesterol causes heart disease’.54 Actually, Keys never proved anything of the sort. He could not, since most of his best-known studies are simple associational studies (see Chapter 17) that cannot prove causation. Worse, it appears that Keys may have resorted to dubious methods to ‘prove’ his associations.55 To provide definitive proof that cholesterol causes heart disease requires, as Mayosi should know, rather more complex science. So complex, in fact, that a causal link between cholesterol and heart disease has never been proven.

  In fact, Keys should have announced in 1973 already that his diet-heart hypothesis was false. But when he realised that the data from his first proper RCT to test the diet-heart hypothesis – the Minnesota Coronary Experiment (1968–1973), involving 9 423 Minnesotan mental patients56 – actually disproved it,57 he may simply have buried the data.58

  Speaking in 2016, the son of Dr Ivan Frantz, who headed up the Minnesota Coronary Experiment, recalled: ‘My father definitely believed in reducing saturated fats, and I grew up that way. We followed a relatively low-fat diet at home, and on Sundays or special occasions, we’d have bacon and eggs … When it turned out that it [a diet low in saturated fat] didn’t reduce risk, it was quite puzzling. And since it was effective in lowering cholesterol, it was weird.’59

  Steven Broste, who, as a student at the University of Minnesota at the time, had access to all the data for his 1981 master’s thesis, suggested to the Washington Post that

  at least part of the reason for the incomplete publication of the data might have been human nature. The Minnesota investigators had a theory that they believed in – that reducing blood cholesterol would make people healthier … So when the data they collected from the mental patients conflicted with this theory, the scientists may have been reluctant to believe what their experiment had turned up.

  ‘The results flew in the face of what people believed at the time,’ said Broste. ‘Everyone thought cholesterol was the culprit. This theory was so widely held and so firmly believed – and then it wasn’t borne out by the data. The question then became: Was it a bad theory? Or was it bad data? … My perception was they were hung up trying to understand the results.’60

  Perhaps, like Rossouw, Frantz and Keys also failed to grasp the significance of the null hypothesis.

  The truth about the Minnesota study was only exposed in 2016, when the original data was recovered and properly analysed.61 One of the lead authors of the re-analysis, Daisy Zamora, subsequently stated: ‘Had this research been published 40 years ago, it might have changed the trajectory of diet-heart research and recommendations.’62 Had that happened, it is probable that the 1977 USDA dietary guidelines might never have come into being and the subsequent obesity and diabetes epidemics would have been avoided.

  But Mayosi’s ringing endorsement of the diet-heart hypothesis in February 2013 would have provided welcome reassurance to the South African pharmaceutical industry. It confirmed that the UCT Faculty of Health Sciences would not be questioning the scientific basis for one of that industry’s most lucrative businesses, specifically the prescription of (largely ineffective) statin drugs to lower blood cholesterol concentrations in those believed to be at risk of developing heart disease or stroke.

  The same Health24 article outlined how, in the Centenary Debate, Rossouw had ‘presented extensive data collected over many decades that supports the current consensus that blood cholesterol occurs as “good” HDL cholesterol, which protects arteries by removing cholesterol from artery walls, and “bad” LDL cholesterol that carries cholesterol molecules into the artery walls’. Many of us continue to wonder how the same compound can be both good and bad. After all, cholesterol is simply a chemical; it lacks the character to be both good and bad, or indeed either.10

  The article quotes Rossouw: ‘This is true for men and women, people of all ages and those with diabetes and after suffering a heart attack. High levels of “bad” cholesterol can be treated with a group of drugs called statins, which reduce heart attacks in people using them.’

  I have to ask why Rossouw, who has spent his life studying the non-medical prevention of heart disease and heart attacks, felt compelled to mention the role of statins in lowering blood cholesterol concentrations. If he has spent his life studying the effects of a low-fat diet in preventing heart disease, why not use the example of dietary interventions to prevent heart attack? The obvious answer is that he knows there is no evidence, at least for the low-fat diet he tested in the WHIRCDMT. There is still no published evidence showing that lowering cholesterol with a low-fat diet reduces heart-attack risk, or anything else for that matter. Instead, replacing dietary saturated fat with manufactured polyunsaturated fat either has no effect or impairs health and may even increase the risk for developing cancer (see Chapter 7).

  Rossouw is further quoted by Health24 as saying: ‘The science is clear that if one follows an unhealthy diet high in fat, one increases one’s “bad” cholesterol levels causing heart attacks and angina.’

  This statement is also false. There is no RCT showing that people placed on high-fat diets develop more heart disease as a result. And there is no evidence that those with established heart disease eat more fat or saturated fat than those without the disease. Instead, it is more likely that they eat more sugar, refined carbohydrates and vegetable oils.

  The article continues to give an account of Rossouw’s presentation in the debate:

  Rossouw explained that the trend of increasing obesity could be attributed to factors like poor diet, lack of physical activity and the environment in which people live. ‘The association between overweight and diabetes,’ he said, ‘is clearly illustrated, as is the benefit of losing weight to prevent diabetes. The data shows that reduced calorie diets result in clinically meaningful weight loss, regardless of which foods are present in low calorie diets.’

  Again, why would he stray onto the topic of weight control and specifically mention the (failed) CICO model of weight control? Is it because this is the model favoured by those who teach dietetics in South Africa under the auspices of ADSA, the HPCSA and the International Life Sciences Institute, a known Coca-Cola front organisation?

  The article concludes its account of Rossouw’s argument with the following:

  Rossouw warns, ‘Noakes maintains that diabetics do not have higher blood cholesterol levels than other people, that half of all heart attacks occur in people with normal blood cholesterol, that glucose is the single most important predictor of risk and a high-fat diet reverses almost all coron
ary risk factors. Except for a very few close followers, the scientific evidence is clear and he’s flying against it. There is now absolute proof that high blood cholesterol is causal in heart disease.’

  The problem is that Rossouw is unable to move on; he continues to present the disproved beliefs of the 1950s as if they are still current, and despite his own $700-million WHIRCDMT disproving them. There is simply no ‘absolute proof that high blood cholesterol is causal in heart disease’. It is lipoproteins, not cholesterol, that are weakly linked, by association, to heart disease in those with IR and NAFLD, and who persist in eating high-carbohydrate diets loaded with polyunsaturated vegetable oils (see Chapter 17). The published scientific literature already supports this fact.

  In sticking to a belief that even his own research has disproved, Rossouw reminds me of what German theoretical physicist Max Planck once wrote: ‘A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.’63 Or, as sometimes paraphrased: ‘Truth never triumphs – its opponents just die out.’ In other words, as Planck also said, ‘Science advances one funeral at a time.’

  Perhaps Rossouw should heed the words of American physicist and Nobel Prize laureate Dr Richard Feynman, who said: ‘It doesn’t matter how beautiful your theory is, it doesn’t matter how smart you are. If it doesn’t agree with experiment, it’s wrong.’ And: ‘The first principle is that you must not fool yourself – and you are the easiest person to fool.’

  The Health24 article also quoted endocrinologist Dr Vash Mungal-Singh, then chief executive officer of the HSFSA:

  Any diet that is calorie and energy restricted will reduce weight. But caution should be used when generalising and saying that everyone should eat fats, no sugar and no carbs, especially when there is no long-term data on this, while there is extensive data on the conventional diet.

  Predictably, Mungal-Singh forgot to mention that the ‘extensive data on the conventional diet’ had proved it to be not only hopelessly ineffective, but also the direct cause of the global obesity/T2DM epidemic that is the principal contributor to heart attacks and strokes in South Africa and elsewhere. Or that Rossouw, the then president of the organisation that paid her salary, was one of the key drivers of the WHIRCDMT, which definitively disproved the diet-heart hypothesis.

  One wonders what continues to drive this cognitive dissonance in the HSFSA. In Mungal-Singh’s case, could it have been the sponsors of the HSFSA who effectively paid her salary? (The HSFSA receives funding and support from a wide range of organisations, as listed in Chapter 7, Table 7.1.) Or was it her misguided sense of loyalty to the ILSI, on whose board of directors she also serves?

  One of the great ironies of my journey is that in 1976 I was one of a group of five people, including two cardiologists, who started the Cape Heart Foundation, which would ultimately become the Heart and Stroke Foundation of South Africa. Naturally, we innovators were the first personally to adopt the 1977 USDA Dietary Guidelines for Americans that promoted LFHC diets based on ‘7–11 servings a day of cereals and grains’. It is my opinion that because of my genetic predisposition – my father suffered all the worst complications of T2DM, ultimately dying from it, while specialists advised him to eat a high-carbohydrate diet – it was this diet that caused my T2DM, even though I was (reasonably) active all my life and never obese (although my body mass index did increase gradually to reach an undesirable, overweight 28 kg/m2 in December 2010).

  Today, the HSFSA continues to promote low-fat diets as the cornerstone for the prevention of heart disease. It also fails to warn of the heart dangers of ingesting partially hydrogenated vegetable oils, especially those found in the margarines produced by one of the HSFSA’s most dedicated funders. The advice to follow the low-fat diet is a direct cause of the explosion of the most severe form of arterial disease, T2DM. Thus, far from preventing arterial disease (of the heart), the low-fat diet that the HSFSA promotes is the direct cause of the horrendous increase in diabetic arterial disease that will swamp global medical services within the next decade.

  In retrospect, I could not then appreciate the extent to which the Centenary Debate was the opening salvo of what I believe to have been a much wider campaign, the ultimate goal of which was to silence me through public humiliation. It is a well-known technique called refutation by denigration. My perception is that if the actions of my colleagues meant that my status as an A1-rated scientist, who had contributed greatly to the scientific and financial efforts of UCT’s Faculty of Health Sciences over 35 years, was destroyed, well, in their opinion, that was just too bad. According to their worldview, I was the architect of my own downfall.

  Only later, when I read Alice Dreger’s Galileo’s Middle Finger: Heretics, Activists, and One Scholar’s Search for Justice, did I begin to appreciate what I was really up against. Dreger’s book explores the unrelenting battle between scholars who put the pursuit of hard truths ahead of personal comfort and the social activists determined to silence them. She uses the voice of the social activist to explain what drives activists in their battles with empirical science and scientists:

  We have to use our privilege to advance the rights of the marginalized. We can’t let [scientists] say what is true about the world. We have to give voice and power to the oppressed and let them say what is true about the world. Science is as biased as all human endeavors, and so we have to empower the disempowered, and speak always with them.64

  The difference, of course, is that the activists I was facing, in my view, were not motivated to advance the voices of the oppressed and disempowered, but, either wittingly or by proxy, rather the opposite.

  In the face of this, what is the responsibility of those scientists who see their role as the pursuit of ‘truth’? Dreger’s answer is this:

  To scholars I want to say more: Our fellow human beings can’t afford to have us act like cattle in an industrial farming system. If we take seriously the importance of truth to justice and recognize the many factors now acting against the pursuit of knowledge – if we really get why our role in democracy is like no other – then we really ought to feel that we must do more to protect each other and the public from misinformation and disinformation …65

  We scholars had to put the search for evidence before everything else, even when the evidence pointed to facts we did not want to see. The world needed that of us, to maintain – by our example, by our very existence – a world that would keep learning and questioning, that would remain free in thought, inquiry, and word.66

  In the end, she concludes: ‘Justice cannot be determined merely by social position. Justice cannot be advanced by letting “truth” be determined by political goals.’67 Nor, I might add, can commercial interests be allowed to determine what is the ‘truth’.

  Dreger’s final message is this: ‘Evidence really is an ethical issue, the most important ethical issue in a modern democracy. If you want justice, you must work for truth. And if you want to work for truth, you must do a little more than wish for justice.’68

  As the media onslaught began, I did not understand that these academic activists seemingly did not care about the science. Neither did the tabloid journalists or Twitter trolls, including some medical colleagues, who at about the same time began to target me on social media. Were they also willing co-conspirators in the rush to silence my voice?

  1No scientist is God, therefore no scientist can ever be certain that s/he has designed the utterly perfect experiment to discover a novel, irrefutable truth. Scientists must assume that the single variable they are studying in their experiments, whatever it may be, will not make any difference to the experimental outcome. We call this the null hypothesis.

  2Modifying a theory to explain why an expected experimental outcome did not happen is known as post-hoc (after the fact) revision.

  3The null hypothesis in the WHIRCDMT was that eating a low-fat diet would
not produce a different outcome than would eating a higher-fat diet. When this was indeed found, the authors should have concluded that their experiment proved that the low-fat diet does not prevent heart disease. But they did not. Instead, they produced a series of post-hoc modifications to explain why their low-fat diet intervention did not work. One post-hoc revision was to conclude that the subjects did not modify their diets sufficiently and, as a result, their blood cholesterol levels did not drop enough to produce a favourable reduction in adverse heart outcomes. But that supposition can only be true if the authors implicitly believe that cholesterol-lowering reduces heart-attack risk – the very question they were studying. If they were so convinced that their theory was correct, why did they need to spend $700 million to prove their bias? Furthermore, $700 million should have been enough to ensure that the study was properly conducted.

  4A long-term trend that develops or progresses over many years.

  5Incensed by this patronising attitude of scientific denialism, the senior author of the Lyon Diet Heart Study, Dr Michel de Lorgeril, wrote a book, Cholesterol and Statins: Sham Science and Bad Medicine (2014; ebook, Thierry Souccar Publishing), expressing his disgust. He is also the author of a number of publications in the scientific literature addressing the absence of proven benefits of statin drugs. Clearly he also qualifies for the title ‘Cholesterol Denialist’.

  6If one measures only blood glucose concentrations to identify people with IR, it will be impossible to detect those who require elevated blood insulin concentrations (hyperinsulinaemia) to maintain normal blood glucose concentrations, because they are already insulin resistant. This is stage 1 IR/pre-diabetes. Stage 2 is when both blood glucose and insulin concentrations are elevated both when fasting and after eating. In stage 3, the IR is so severe that blood glucose concentrations rise to 8–9 mmol/L, causing glucose to appear in the urine. This is T2DM as traditionally described.