Written by Prof Tim Noakes
My critics have called me deluded and dangerous. In the South African Medical Journal in 2013, they said I have cherry picked, misinterpreted data, I don’t understand the science, I’ve lost my way, flouted the Hippocratic Oath, and I’m damaging patients and the population.
Last year, for the first time in the history of the University of Cape Town, no senior academic has ever been criticized as publicly as I was. Senior colleagues, including the Dean of the Medical Faculty at UCT (who has since moved upwards and onwards – a reward for his bravery perhaps) sent a letter to the Dean of all South African medical schools and to the press, saying:
“There is good reason for concern that this diet may rather result in nutritional deficiencies, increased risk for heart disease, diabetes mellitus, kidney problems, constipation, certain cancers and excessive iron stores in some individuals in the long-term.”
They said I was “making outrageous unproven claims about disease prevention, and maligning the integrity and credibility of peers who criticize his diet for being evidence-deficient and not conforming to the tenets of good and responsible science. This goes against the University of Cape Town’s commitment to academic freedom as the prerequisite to fostering responsible and respectful intellectual debate and free enquiry.”
The letter ended: “UCT’s Faculty of Health Sciences, a leading research institution in Africa, has a reputation for research excellence to uphold. Above all, our research must be socially responsible. We have therefore taken the unusual step of distancing ourselves from the proponents of this diet.”
The authors didn’t stop there. They ran a website in which they collected extraneous material, selected what they said was my argument, and threw in blogs by people unrelated to the topic, all without giving me the right of reply.
Clearly their goal was to prove that I’m deluded and dangerous. Rather than attacking the science, they attacked me personally and said I was practising junk science.
These are not ugly, horrible people. They were saying what they believed out of a deep sense of conviction.
Are they right, or am I right? We can’t both we right.
I will present the evidence to show that my opinion is scientifically based, does not break any of the rules of good science, that I have a right to that opinion, and that my critics are the ones who are practising junk science, and are endangering people’s health.
They don’t understand causation science, hazard ratios, insulin resistance as well as the special role of gluten, the leaky gut and non-coeliac gluten intolerance in human ill-health.
The key problem is that both sides believe the facts sit with them. At the first international low-carb, high-fat summit in Cape Town, we spent three days discussing evidence for and against low-carb, high-fat diets to treat insulin resistance.
We need to begin by looking at the quality of the evidence.
My critics say low-carb diets are proven not to work and saturated fat is proven to increase the risk of cardiovascular disease (CVD). They say we only have anecdotal evidence for low-carb, high-fat diets for the treatment of insulin resistance (IR) and diabetes.
That’s not correct. We have the evidence: all randomized, controlled trials (RCTs) either show that high-fat outperforms low-fat diets, or long-term RCTs show no evidence that a low-fat diet does any good.
My critics have ignored all the RCTs and other evidence that dispute their theories. In so doing, they have not practised good science according to the rules laid down by Sir Austin Bradford Hill, who is revered as the “father of medical statistics”.
Bradford wrote a series of article in 1937 in the Lancet describing the use of statistics in medical science. It was a completely new science. He was one man who really understood it, and he laid down levels of information for causation, starting from anecdote (case history), and including cross-sectional study, randomized, controlled, prospective, clinical trials, and finally systematic review and meta-analysis.
He explained that it was possible to prove causation from associational studies, but only if certain criteria were met. He listed nine such considerations, but I have focused in this paper on only two: coherence and strength of association.
He wrote in 1965: “Here then are nine different viewpoints from all of which we should study association before we cry causation. What I do not believe…is that we can usefully lay down some hard-and-fast rules of evidence that must be obeyed before we accept cause and effect.
“None of my nine viewpoints can bring indisputable evidence for or against the cause-and-effect hypothesis and none can be required as a sine qua non. What they can do, with greater or lesser strength, is to help us make up our minds on the fundamental question – is there any other way of explaining the set of facts before us, is there any other answer equally, or more likely than cause and effect?”
Bradford Hill designed and completed the first RCT in 1950 (Streptomycin in TB meningitis) and with Sir Richard Doll “proved” that smoking causes lung cancer initially from an associational study (which cannot prove causation). He found that the Hazard Ratio for lung cancer in smokers was 10-30 times higher than in non-smokers. Such a high value could only indicate causation in his opinion.
He wrote: “On the other hand the death rates from coronary thrombosis in smokers is no more than twice, possibly less, the death rate in non-smokers (ie Hazard Ratio of 2 or less). Though there is good evidence to support causation it is surely much easier in this case to think of some features of life that may go hand in hand with smoking – features that might conceivably be the real underlying cause or, at the least, an important contributor, whether it be lack of exercise, nature of diet or other factors.
“But to explain the pronounced excess in cancer of the lung in any other environmental terms requires some feature of life so intimately linked with cigarette smoking and with the amount of smoking that such a feature should be easily detectable. If we cannot detect it or reasonably infer a specific one, then in such circumstances I think we are reasonably entitled to reject the vague contention of the armchair critic ‘you can’t prove it, there may be such a feature’”.
Bradford Hill has since died, and many researchers have ignored his criteria and flipped into a model of junk science – the scientists who are more interested in getting funding and more work, not discovering how to make people healthier.
In science, the bar has dropped to the lowest level of scientific “proof” conceivable – so low that researchers now ignore what Bradford Hill taught and accept any Hazard Ratio above 1.0 as definitive evidence of causation. Poor Bradford Hill turns in his grave with every new publication.
That means just about anything can be proved to cause anything. The result is that have grown an entire discipline of nutritional science based on this improper understanding of Bradford Hill’s doctrines. And we wonder why we have got it all so very, very wrong.
So all the associational nutritional studies used to justify the 1977 USDA (low fat) Dietary Guidelines are based on studies with Hazard Ratios usually between 0.7-1.3. Bradford Hill would not have accepted any of these studies as evidence for causation, or allowed them to be used as the sole justification for novel global dietary guidelines. In fact studies using such feeble criteria for causation are simply scare-mongering, the ultimate junk science.
Since most such studies originate from departments of epidemiology that consider themselves at the forefront of hard science, author James Le Fanu proposes a simple solution: “Meanwhile the simple expedient of closing down most university departments of epidemiology could both extinguish this endlessly fertile source of anxiety-mongering while simultaneously releasing funds”.
Interestingly, a recent report – A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart disease – used four Bradford Hill criteria (strength, consistency, temporality and coherence) to evaluate all the evidence from dietary studies. The study showed strong evidence for vegetables, nuts, “Mediterranean” and high quality dietary patterns for “protective effects against CHD”. These are exactly the components that the Banting high fat diet promotes. They also showed strong evidence that trans-fatty acids and foods with high glycaemic index or load (ie high carbohydrate diets) as “harmful effects promoting CHD”. These are exactly the foods that the Banting high fat diet does not allow. The study found insufficient evidence for: Saturated and polyunsaturated fatty acids; total fat; alpha linolenic acid; meat; eggs; milk.
So if my critics were steeped in the science as they claim, they could only have concluded in their letter to the Deans of all the Medical Schools and to the media, that my advice was completely compatible with the most rigorous science currently available. That I am also correct to argue that there is no evidence that a high fat diet causes anything. Instead they concocted an argument based on evidence that Bradford Hill would have rejected. And this from leading scholars at a leading medical institution in this country.
In his book, The Rise and Fall of Modern Medicine, James Le Fanu says: “Bear in mind Sir Austin Bradford Hill’s insistence that statistical inferences by themselves have no meaning unless they are internally coherent, that is to say, when several different types of evidence for an association between an environmental factor and disease … are examined, they (must) all point to the same conclusion.”
“Put another way, no matter how plausible the link between dietary fat and heart disease might seem, just one substantial inconsistency in the statistical evidence effectively undermines it.” There is now more than one substantial inconsistency that should long ago have relegated the low fat diet to the dust bin of bad science.
My critics also like to say that we only have anecdotal evidence for the benefits of low-carb, high-fat diet. That’s simply not true. There is significant anecdotal evidence, and all of science begins with anecdote. But we also have a wealth of RCT evidence for the superiority of low-carb, high-fat diets of low fat, high carbohydrate diets.
A speaker at the summit, Canadian Dr Jay Wortman, told how he became diabetic 12 years ago, cut carbs from his diet (as a doctor he knows that carbs raise blood sugar), within days his symptoms had improved, and he has been without evidence of diabetes ever since.
Two of my clients attended the summit: Billy Tosh weighed 163kg in July 2012, was close to a heart attack, and had type 2 diabetes and hypertension. By March 2013, Billy had lost 84kg, his hypertension has gone, and he is free of symptoms of diabetes and hypertension. Brian Berkman weighed 153 kgs in July 2011, was diabetic, hypertensive and considering bariatric surgery. By January 2013, he had lost 82kg and is free of symptoms of diabetes, is no longer hypertensive, and avoided bariatric surgery.
Is that all just anecdote? Probably not. I don’t claim they are cured, but they don’t require medication and are without evidence for diabetes.
There are many other case histories of spontaneous recovery from type 2 diabetes, a condition that my profession teaches is irreversible, will require increasing medication for life and has, in essence, a hopeless outlook with the only certainty – more drugs, more illness, more disability.
How many subjects do you require in a trial to prove an effect?
In his book, The World Turned Upside Down, author Richard Feinman said: “It depends on how many people recover spontaneously.”
If there has never been a reported reversal of Type 2 diabetes mellitus (T2DM) in patients following conventional medical advice (which there has not), then a single case is not an anecdote. It is a black swan – in other words something that contradicts our previous beliefs, for example that all swans must be white to be classified as swans. The presence of a black swan requires the immediate funding of a proper scientific study – a randomised controlled clinical trial – to test whether it is possible to reverse T2DM with a low carbohydrate diet.
A key question has been: you can’t prove causation without randomized controlled trials (RCTs). But as I’ve already shown you can in fact surmise causation in cross-sectional study if you fulfill certain strict Bradford Hill criteria. And when applied to cross-sectional dietary studies, the Hill criteria support the health benefits of the low carbohydrate diet. But let’s first consider the evidence from RCTs.
One of my fiercest critics was involved in the most significant RCT: The Women’s Health Initiative Randomized Controlled Dietary Modification Trial, published in the JAMA in February 2006. It was a large trial of 48,836 post-menopausal women, followed over 8.1 years looking at the effects of low-fat eating and costing about $700 million. The study concluded: “Over a mean of 8.1 years, a dietary intervention that reduced total fat intake and increased intakes of vegetables, fruits, and grains did not significantly reduce the risk of CHD, stroke, or CVD in postmenopausal women and achieved only modest effects on CVD risk factors, suggesting that more focused diet and lifestyle interventions may be needed to improve risk factors and reduce CVD risk.” Another two very expensive low-fat RCTs have come to the same conclusions.
The WHI study authors should have designed their study as the test of a null hypothesis: Specifically, if you reduce your fat intake, you will reduce heart attacks and cancer rates. If the results don’t support the hypothesis, then the hypothesis is clearly wrong and must be abandoned (if one is practicing good rather than junk science).
Instead, when the authors discovered that the data did not support their original hypothesis, they simply added an ad hoc modification. So in a news release, Dr Elizabeth Nabel then head of the National Institute of Health – the statutory body that had funded the study with tax payers’ money – suggested that the findings “could have been due to chance”, and that the participants could still have been eating too much fat.
Actually no, Dr Nabel. That’s not science. The study disproves your hypothesis. When the hypothesis is disproven, you have to come up with a new one and then attempt to disprove it.
Albert Einstein said: “No amount of experimentation can ever prove me right;
a single experiment may at any time prove me wrong.”
The WHI study should have been considered the definitive disproof of the authors’ hypothesis that eating less fat will prevent heart disease. But instead the authors and the NIH marketed it as if it supported their hypothesis. That is not science. That’s science driven by industry or governments that are determined to find an outcome that supports their position, regardless of the facts. Why bother to do research if you “know” the outcome before the start? Of if you will interpret any outcome to support your ingrained prejudices?
But the finding that low-fat diets did not reduce the risk of cardiovascular disease is entirely predictable as a high carbohydrate diet produces a specific atheroma-generating metabolic profile in those who are metabolically vulnerable because they have insulin resistance.
For example, there is one study that looked at the progression of
coronary atherosclerosis (narrowing of the coronary arteries) in postmenopausal women by Harvard researcher David Mozaffarian and others, and published in the American Journal of Clinical Nutrition in 2004. The study concluded: “In postmenopausal women with relatively low total fat intake, a greater saturated fat intake is associated with less progression of coronary atherosclerosis, whereas carbohydrate intake is associated with a greater progression.” In other words the study found that higher intakes of “healthy” carbohydrates and “healthy” polyunsaturated fats was associated with more rapid disease progression, whereas women who ate the most saturated fat showed NO disease progression. Naturally this study has been buried, never to be heard of again. More recently, a study in the Journal of Nutrition in February 2015, confirmed that “ Dietary intake of saturated fat is not associated with risk of coronary events or mortality in patients with established coronary artery disease.”
But the strongest evidence against this fake hypothesis has been provided by Nina Teicholz in her riveting book, Big Fat Surprise: Why Butter, Meat and Cheese belong in a Healthy Diet. Teicholz reviews the absence of science behind the hypothesis that saturated fat causes disease. Dr Richard Smith, a former editor of the BMJ, had this to say about this book which should be required reading for all:
“The title, the subtitle, and the cover of the book are all demeaning, but the forensic demolition of the hypothesis that saturated fat is the cause of cardiovascular disease is impressive.
“Indeed, the book is deeply disturbing in showing how overenthusiastic scientists, poor science, massive conflicts of interest, and politically driven policy makers can make deeply damaging mistakes.
“Over 40 years I’ve come to recognise what I might have known from the beginning that science is a human activity with the error, self deception, grandiosity, bias, self interest, cruelty, fraud, and theft that is inherent in all human activities (together with some saintliness), but this book shook me.”
At the Old Mutual Health Summit, Zoe Harcombe presented more novel evidence against the fake hypothesis. She asked the question: What evidence was available from RCTs in 1977 and 1983 to support the adoption of those novel low fat guidelines. Her study reported in the BMJ Open Heart journal one week before the Summit concluded that “Recommendations were made for 276 million people following secondary studies of 2467 (ill) males with reported identical all-cause mortality. RCT evidence did not support the introduction of dietary fat guidelines.”
So now we know that there was no evidence available in 1977 to support the change in dietary advice that became the global standard. And we also know that no evidence has accumulated in the past 40 years to show retrospectively that those guidelines are correct and supported by the most rigorous science.
And more global experts are beginning to weigh in their support almost on a daily basis. Thus in an editorial in Diabetes Management in 2014, Dr Osama Hamdy, Medical Director of Joslin Obesity Centre’s Obesity Clinical Programme, wrote:
“It is clear that we made a major mistake in recommending the increase of carbohydrates load to >40% of the total caloric intake (especially for person with type 2 diabetes mellitus). This era should come to an end if we seriously want to reduce the obesity and diabetes epidemics. Such a move may also improve diabetes control and reduce the risk for cardiovascular disease. Unfortunately, many physicians and dieticians across the nation are still recommending high carbohydrates intake for patients with diabetes, a recommendation that may harm their patients more than benefit them.”
Yet my critics continue to ignore all the evidence that favors the health benefits of high-fat diets and the absence of evidence supporting low fat diets.
Could be due to the Upton Sinclair Theorem, which states: “It is difficult to get a man to understand something, when his salary depends upon his not understanding it”.
Perhaps the most important reason why the value of low carbohydrate diets are not yet properly appreciated especially by my profession is because we do not appreciate the importance of the condition of insulin resistance (IR) which is perhaps the single most important biological condition across the globe. So it is my argument that the global epidemics of obesity, diabetes, hypertension, gout, and atherogenic dyslipidaemia (high triglycerides, low HDL-C, increased number of small LDL-C particles, and increased triglyceride-rich remnant lipoproteins) and perhaps also cancer and dementia, are really the tip of the iceberg – the markers of an underlying biological predisposition that becomes apparent in those exposed to high carbohydrate diets and then presents as one of more of those conditions.
So it is my thesis that IR is the most prevalent biological condition in the world. It remains hidden as long as diets are not high in sugar and refined carbohydrates. But in the face of a high carbohydrate diet eaten for decades, the IR leads to all the common chronic diseases that we face today. But the problem is that IR is not taught in many medical schools or schools of dietetics and nutrition and this perhaps is the key problem. For if we don’t recognize the single most important factor predisposing to chronic ill health across the globe, then we are not likely ever to be able to cure or reverse those diseases. Especially if the cure is to remove the cause which is a high carbohydrate diet.
Ignoring IR undermines the modern practice of chronic medicine. If all the conditions linked to IR are caused ultimately by high carbohydrate diets – that is by a nutritional factor, as I believe they are, we don’t need medication, and the pharmaceutical industry that is designed to market its drugs to treat those conditions. For these are not conditions caused by the lack of a specific pharmaceutical chemical. They are caused by too many carbohydrates in the diet.
We fuel the fire with carbohydrates, and try to put it out with pharmacologic drugs that do not address the real cause.
Our critics’ views are based on the belief that all humans can metabolise carbohydrates equally. The condition of IR disproves that idea. Instead IR shows that for some even the smallest amounts of carbohydrate eaten for decades are enough to seriously damage our health in the long term.
In Black Holes and Baby Universes and Other Essays, Stephen Hawking says: “People are very reluctant to give up a theory in which they have invested a lot of time. They usually start by questioning the accuracy of the observations. If that fails, they try to modify the theory in an ad hoc manner. Eventually the theory becomes a creaking and ugly edifice.
“Then someone suggests a new theory in which all the awkward observations are explained in an elegant and natural manner.”
At the low-carb, high-fat summit in Cape Town, I believe that’s what we have done. We have exposed the creaking, ugly edifice of conventional wisdom on nutrition, and explained an alternative in an “elegant and natural manner”.