by Tim Noakes
The problem with Keys, Harcombe said, was his bias from the outset. One of the consequences of that bias was the statins industry. The diet-heart hypothesis created the conditions for statins to become the most prescribed drug in modern medicine, and earned the pharmaceutical companies that make them billions of dollars, despite compelling evidence of their inherent risks. In her testimony, Harcombe said that statins are ‘one of the biggest crimes against humanity that the pharmaceutical industry has unleashed’.
Perhaps Harcombe’s most startling contribution to Noakes’s defence was her investigation into the Naudé review, which had concluded that low-carb diets were no better for weight loss than conventional, LFHC ‘balanced’ diets. Harcombe told the hearing that she and Noakes had done their own analysis of the Naudé review, due for publication in the SAMJ in December 2016.5 They had found the review to be littered with errors, many of them material. While admittedly they had only re-examined one part of the Naudé review, given the amount of errors in that section alone, she and Noakes believed that it was unlikely that the conclusions were robust.
Noakes and Harcombe’s findings raised troubling questions that continue to swirl around the Naudé review to this day. How is it possible that so many researchers from top South African universities could make so many mistakes in one study? Were these honest mistakes, or was there ‘mischief’ behind the errors?
Harcombe said that, among other errors, the researchers had included studies that failed their own inclusion criteria; used invalid and subjective meta-analysis sub-groupings; and repeatedly extracted data inaccurately. One instance of data extraction was so erroneous that Harcombe called it ‘absurd’, another ‘inexplicable’.
A major limitation of the meta-analysis, Harcombe said, was that the authors claimed to have reviewed evidence for low-carb diets, but had not in fact done so. Put another way, she said, they could not judge low-carb diets because they did not study them. For example, the average dietary intake for 14 of the studies used was 35 per cent carbohydrate, 35 per cent fat and 30 per cent protein. That’s very different from the 5 to 10 per cent carbohydrate and 80 to 85 per cent fat of an accepted, effective LCHF diet to treat or prevent obesity, diabetes and heart disease.
The Naudé review also set isocaloric (having similar caloric values) as a criterion. This negated the satiety advantage of low-carb diets. Satiety (the feeling of fullness) is a key effect of the low-carb diet despite a reduced energy intake. The researchers would therefore have had to restrict the caloric intake voluntarily of subjects on the control diet to match this effect.
In the background to the study, the review’s authors referred to ‘some weight loss diets widely promoted through the media’, but named only one: the Atkins diet. They said that these diets ‘recommend a regimen greatly restricting carbohydrates, with increased protein and unrestricted total and saturated fat intake’. This is simply untrue. Most low-carb diets don’t advise unrestricted total and saturated fat intake; they cap it. ‘It is plausible that these low CHO diets could be harmful, especially over the longer term,’ the authors concluded in the background to the review. They gave references, but nothing that stood up to Harcombe and Noakes’s independent scrutiny.
Harcombe said that, based on their own data, the researchers could just as easily have concluded that a ‘balanced’ (high-carb) diet was no better than a low-carb diet for producing weight loss. If the researchers were to redo their research properly, she said, without all the errors, they would have to draw a very different conclusion – namely, that the low-carb diet worked better than the (LFHC) control diet for weight loss.
Harcombe was scathing about the media’s coverage of the Naudé review. While the review’s authors made no mention of LCHF, Banting or Noakes, media reports quoting a press release claimed that the Naudé review effectively ‘debunked the Banting diet’ – and Noakes. Some reports personalised the message, referring to Noakes as a ‘celebrity professor’ and stating that the review proved that ‘Noakes’s low-carb diet is not healthier’. Others quoted HSFSA CEO Dr Vash Mungal-Singh saying: ‘The current evidence means we cannot recommend a low-carbohydrate diet to the public.’
Harcombe said it was absurd, impactful and potentially harmful to claim that the review debunked Banting when it didn’t even study Banting diets. In her opinion, the entire episode – the Naudé review, the accompanying press release and the media coverage that followed – was ‘personal, unprofessional and, on examination, it was flawed’.
Noakes and Harcombe were not alone in raising questions about how the study got through peer review. Some said it signalled a deep rot in the process. Concerns about peer review are not new. In 2006, former BMJ editor Dr Richard Smith condemned the process in the Journal of the Royal Society of Medicine, saying that there is little evidence on the effectiveness of peer review, but considerable evidence on its defects. ‘In addition to being poor at detecting gross defects and almost useless for detecting fraud,’ he wrote, ‘it is slow, expensive, profligate of academic time, highly subjective, something of a lottery, prone to bias, and easily abused.’6
In cross-examination, Bhoopchand attempted to undermine Harcombe’s expertise. He suggested, for example, that she was not qualified to comment on statins, since she was neither a cardiologist nor any other kind of medical doctor. Harcombe easily deflected his objection, saying that part of the problem was that cardiologists and other doctors had little knowledge or training in nutrition. If they did, they would be better able to understand all the research pointing to the risks outweighing the benefits of statin use.
In particular, Bhoopchand tried to undermine the impact of Harcombe’s devastating critique of the Naudé review, likely on instruction from Senekal, one of the study’s co-authors and a consultant to the HPCSA in its case against Noakes. Senekal was present through all the defence’s evidence. When she wasn’t staring stonily in Harcombe’s direction, she was vigorously chewing gum and shaking her head. She scribbled notes throughout, which she passed to Bhoopchand, who referred to them repeatedly during cross-examination.
Bhoopchand’s questions thus revealed what was probably Senekal’s growing fury at Harcombe and Noakes. At one point, Bhoopchand asked Harcombe whether, as ‘a question of collegiality’, she had known that her critique of the Naudé review would cause ‘some embarrassment to the University of Stellenbosch’. He seemed to be suggesting that the researchers’ potential embarrassment should have stopped her and Noakes from going public with their findings. Bhoopchand asked Harcombe whether she should have ‘at least elicited some response from at least the first author to that particular study’.
Harcombe responded that she was not aware that the authors had given Noakes the same privilege, prompting Bhoopchand to ask her if it was a case of ‘an eye for an eye’. Looking him straight in the eye, Harcombe said: ‘No, not at all, sir. I looked at the [Naudé review] and it was not robust. So I reported it as such.’
Harcombe made no major concessions under cross-examination on any issue, and especially not on her critique of the Naudé review. Yet in their first response to Noakes and Harcombe’s re-analysis a week after its publication in December 2016, the review’s authors claimed that all the ‘numerous criticisms’ had been ‘addressed’ during cross-examination at the hearing in October, and that Harcombe had ‘conceded more than seven times’ that the errors ‘were in fact not material to the findings of [the] review’.7
It was as if they had attended a completely different hearing. I emailed the authors to suggest that someone must have given them the wrong information. They could look at the transcript to see for themselves, or they could ask Senekal for the facts, as she was present throughout. Celeste Naudé replied to say she was out of town and would get back to me. She never did, and I heard not a word from the other authors either.
In their official response to Noakes and Harcombe in the SAMJ in March 2017, Naudé and her co-authors basically rubbished the findings and suggested that Noak
es and Harcombe had themselves erred and shown a ‘lack of understanding of current methods in evidence synthesis’.8 The Naudé-review authors appeared to have taken a leaf out of US President Donald Trump’s strategy book. They were hiding behind alternative facts and creating fake news.
Noakes and Harcombe held nothing back in a rebuttal letter to the SAMJ in May 2017, titled ‘Naude et al. avoid answering the essential question: Mistake or mischief?’:
It is common cause that the Naude/Stellenbosch University/University of Cape Town meta-analysis played a decisive role in the multimillion rand prosecution of Prof. T Noakes by the Health Professions Council of South Africa … Without the ‘correct conclusion’ from this meta-analysis, it is possible that the HPCSA trial against Noakes might never have happened. Therefore, the importance of the Naude et al. meta-analysis extends far beyond any role purely as a neutral scientific publication.
Had we realised the disproportionate consideration given to this ostensibly innocuous publication in the HPCSA trial, we would have examined it sooner.9
Noakes and Harcombe concluded: ‘… we may never receive an answer to our research question: was this mistake or mischief?’ They were essentially accusing researchers from two of South Africa’s top universities of scientific fraud and colluding with ADSA and the HPCSA to silence him. If that accusation is not shocking enough, the response of all the universities involved, and at the highest levels, has been even more surprising and telling. They have closed ranks and protected their academics.
Bhoopchand had nothing to say about Harcombe’s vision of a way forward for public-health nutrition advice in future. She referred to the HPCSA’s own ‘Ethical rules of conduct for practitioners registered under the Health Professions Act, 1974’. Schedule 27A(d) states that a practitioner shall, at all times, ‘provide adequate information about the patient’s diagnosis, treatment options and alternatives … and any other pertinent information to enable the patient to exercise a choice in terms of treatment and informed decision-making pertaining to his or her health and that of others’ (Harcombe’s emphasis).
This, Harcombe said, brings practitioners back to ‘advising real food’. One obstacle is that many so-called experts still confuse saturated fatty acids with processed foods, she said, before listing the primary sources of saturated fat in the American diet as pizza, desserts, sweets, tacos and ready meals. These are not real foods, she said. If those preaching conventional ‘wisdom’ would just tell people not to eat processed, junk foods, ‘we could find agreement. But call it what it is. Stop calling it saturated fat, because it is predominantly junk.’
Harcombe also stated that dietitians’ associations globally should not have any compromising links with the food industry. ‘It is difficult to embrace wholeheartedly the concept of real food when your sponsors include cereal companies and sugar companies and makers of polyunsaturated products,’ she said. ‘It is inevitably going to lead you towards demonising real food and favouring those processed foods. But if we can get that conflict out of the way, surely we can agree on real food, and the only-then valid nutritional debate the world should be having is, what should that real food be?’
Harcombe concluded by saying that health professionals in South Africa have an ethical and professional duty to share the fact that ADSA’s advice is conflicted because of the association’s links with the food and drug industries. They also have a responsibility to share the facts that the most nutritious foods are of animal origin, and are naturally low in carbohydrates and high in healthy fats. Health professionals act professionally and ethically when they give patients all relevant information and options so that they can make informed choices, Harcombe said.
Nina Teicholz was up next. Her mission: to explode as many fat bombs as possible. In the course of her evidence, she dispelled myths and misconceptions about the role of carbohydrates and fats – especially saturated fats – in a healthy diet. Like Harcombe before her, she effectively undermined the foundations of the HPCSA’s charge against Noakes. Teicholz based her evidence on her ground-breaking book, The Big Fat Surprise, published in 2014. The fruits of nearly a decade of research into the surprisingly murky world of nutrition science, The Economist called it a ‘nutrition thriller’. It is indeed a fascinating forensic journey into a field influenced by politics and corporations, and characterised by scientific one-upmanship. In the book, Teicholz analyses the last 50 years of nutrition policy in the US as it relates to dietary fat and cholesterol, and reveals how an idea about fat and health became official policy despite all the evidence contradicting it. She exposes the policies, personalities, politics and industries behind the construction, implementation and maintenance of the US dietary guidelines, which have been emulated the world over.
The Big Fat Surprise focuses on the pillar of those guidelines – and the case against Noakes: the diet-heart hypothesis that saturated fat and cholesterol cause heart disease. As we have seen, this hypothesis is at the heart of ‘conventional’ nutrition advice, not just for cardiovascular health, but also for the treatment and prevention of obesity, diabetes and a host of other illnesses. In the book, Teicholz demolishes the belief that saturated fat causes heart disease, and shows the opposite to be true. She presented evidence that low-fat diets actually increase the risk of cardiovascular disease. In her testimony, Teicholz thus sabotaged an important part of the HPCSA’s case.
She also noted that the US Dietary Guidelines Advisory Committee had drawn attention to the harmful effects of the low-fat diet in its most recent report by warning that low-fat diets cause atherogenic dyslipidaemia (one of the major components of the metabolic syndrome, characterised by elevated levels of LDL and low levels of HDL cholesterol). The US dietary guidelines no longer include any official language about limiting total fat intake, and in the latest guidelines, cholesterol is ‘no longer a nutrient of concern’.
‘The low-fat diet is over,’ Teicholz declared. Not only does this diet appear to cause heart disease, but in clinical trials on more than 52 000 people, the low-fat diet was shown to be ‘ineffective in fighting any other kind of chronic disease’, she said. Yet the demonisation of saturated fat in South Africa’s official dietary guidelines continues.
Teicholz also presented evidence to show that low-fat diets deprive infants and children of the vital dietary fats needed to absorb vitamins and other nutrients during their most formative years. This effectively undermined the claim of the HPCSA’s expert witnesses that Noakes’s advice to a breastfeeding mother was potentially ‘dangerous’. It wasn’t difficult to work out who Teicholz believed was dishing out dangerous advice for infant weaning. She used robust science to finger the usual suspects, including Strydom, Vorster, Kruger and Dhansay.
Teicholz recounted for the hearing how her nutrition journey began: with an investigation into trans fats for food magazine Gourmet in 2004. This assignment introduced her to the ‘world of fat’, she said. ‘Fat is what we obsess most about in nutrition – how much fat to eat, what fat, good fat, bad fat, low-fat, non-fat.’ During her research, she began to realise that everything she thought she knew about dietary fat was wrong, and that US nutrition policy was ‘completely upside down and backwards as to what we should be eating’. She demonstrated that the same is true of South Africa’s official nutrition policy.
One of the strengths of her book, and thus of her evidence for Noakes, is that Teicholz did not rely on summary statements or review papers in her research. Instead, she went back to all the original papers, and sometimes to the original data. In many cases, she found that scientists had tried to hide their data, sometimes even publishing it in foreign-language publications to make it more difficult to access. Teicholz hunted those down too, and found professionals to translate them for her. She attended conferences and interviewed hundreds of top scientists from all over the world, as well as many leading food-industry executives.
Teicholz described how she was met with silence in unexpected places when asking certain que
stions. Interview subjects would say ‘I can’t talk about fat’ and abruptly end the conversation. She said that sometimes experts were so reluctant to talk that she felt as if she were ‘investigating the Mob’. The analogy is not inappropriate, as the case against Noakes has shown.
Besides showing the diet-heart hypothesis to be false, Teicholz’s book addresses how it became enshrined as truth in the public consciousness. Like any idea, the diet-heart hypothesis was born in ‘a moment in time’, Teicholz said at the hearing. It began in the US in the 1950s, when there was ‘rising panic over the increase in heart disease that had come from seemingly out of nowhere in the early 1900s’. Competing explanations for heart disease ranged from vitamin deficiency to vehicle exhaust fumes and an increasingly stressful lifestyle. Ancel Keys didn’t buy into any of these.
There are various theories about how history unfolds, Teicholz said. One is that history is controlled by sociological forces (‘guns, germs and steel’). Another, called the ‘Great Man’ theory, postulates that history is formed by powerful personalities. In the history of nutrition science, said Teicholz, Keys was one such person. As a man with a unique and forceful personality, an unwavering faith in his own beliefs and an ‘indomitable will’, Keys would virtually ‘argue people to death’. In her characteristically measured style, Teicholz said it was fair to say that Keys was always ‘more interested in being right than in being a good scientist’. Less diplomatic critics have called Keys a ruthless, arrogant bully.
By the mid-1950s, Keys was convinced that saturated fat caused heart disease. He believed that saturated fat and dietary cholesterol raised blood cholesterol levels, and that increased blood cholesterol clogged arteries, ultimately causing heart attacks. He called his theory the diet-heart hypothesis and endeavoured to make it a pillar of the US dietary guidelines. Teicholz described the circuitous route Keys took to get there. In 1950s America, the only group providing advice on lifestyle habits to avoid a heart attack was the American Heart Association. Fat did not feature among the foods to avoid. In fact, the AHA even warned against pulling the trigger on fat too soon, based on incomplete evidence.