Eagle-eyed readers may have noticed that the commonly-held belief that saturated fat cause heart disease appears to be waning. In the last few weeks alone we have seen the publication of two significant and weighty studies which have found no link between sat fat and heart disease. These come hot on the heels of another review last year that found the same. I covered these studies last Friday here. These recent studies did not get much airtime in conventional media, but they did at least appear to whip up some interest in the blogosphere.
Now, this might be a coincidence, but on Tuesday there was here in the UK a huge media splash instigated by the suggestion from a heart surgeon that we should ‘ban butter’. (See here for an example of this coverage). The surgeon in question, Mr Shyam Kolvekar, urges us all to cut back on saturated fat (sigh) to save our coronary arteries. Now, Mr Kolvekar’s message got widespread exposure (see here for an example), but he was not acting in isolation. He was aided and abetted by Unilever’s PR agency (KTB). Unilever, in case you are not familiar, is a food company that makes, among other things, highly chemicalised, non-food margarines touted as a ‘healthy’ alternative to butter.
There is no good evidence that I can find that supports the supposed ink between saturated fat and heart disease. So I emailed Mr Kolvekar asking for evidence that supports his assertions. I also asked him to clarify whether or not he had received payment for Unliever for this or any other work.
To his credit, Mr Kolvekar responded. He declared no commercial gain from any work with Unlever (I believe him). However, on the evidence, he was not so forthcoming. He referred me to the FSA (Food Standards Agency) and Department of Health websites. He stated, that what I read in the news was his own personal opinion.
I emailed him back, pressing him for actual studies. I wrote: Can I ask you to specifically cite the evidence (preferably human intervention studies, but epidemiological evidence too if you have it) that supports the assertions you have made regarding saturated fat and butter. Please do not refer me to websites, consensus statements or the like. It is actual studies I am interested in.
I did get a reply, which I suppose I should be thankful for. But it was not particularly fulsome. It actually contained just one citation, with no supporting text. Not even a ‘Dear John’.
I responded to Mr Kolvekar with the following email.
Dear Shyam
In response to my request for you to supply evidence that supports your assertions regarding saturated fat you provided a single reference [1]. This reference concerns what is commonly referred to as the North Karelia Project. As you may know, this project was an attempt to reduce the risk of heart disease in individuals living in the province of North Karelia in Finland. The project started in 1972. Outcomes were initially compared to a control group which was not subjected to any intervention. The intervention was deemed a success, and was subsequently adopted on a country-wide basis. The reference you offer refers to that country-wide intervention.
The first thing to note about the North Karelia Project is that is was multiple intervention in nature. In addition to attempts to have people eat less saturated fat, it also aimed to reduce smoking and blood pressure. It employed a number of strategies including health education, screening and intensification of treatment in those already being treated for cardiovascular disease. Because of the multiple intervention nature of the project, it is simply impossible to gauge which element(s) of it might have been responsible for any benefits seen from the interventions.
A closer look at the project itself reveals other issues, which have been highlighted before [2]. Shah Ebrahim and George Davey Smith point out that while the results of the North Karelia project have been held up as a ‘success’, the reality is that it’s results were quite unimpressive. Let me quote from this letter: CHD mortality trends over the period 1969 to 1995 show a greater decline in North Karelia than the rest of the country, but the 95% confidence intervals for the slopes overlap. An intriguing pattern of decline is hidden in the overall trend. First, North Karelia experienced an almost immediate and rapid decline, a rise and a fall in CHD mortality (1971″1975). Second, rates of decline were significantly greater in the country as a whole than in North Karelia (1976″1985). Finally, death rates tended to approximate to each other (1986″1995). Such trends do not provide unambiguous support for the hypothesis that the intervention was effective.
It is notable that one of the original North Karelia investigators subsequently wrote to the Lancet to express his own doubts regarding the ‘success’ of the project [3]. He expressed the opinion that he did not think it was possible to draw the positive conclusions that had been presented in the original paper.
With the one piece of evidence you have cited to support your assertions regarding saturated fat we have two fundamental problems:
1. It is a multiple intervention in nature, and it’s therefore impossible to draw conclusions about the effects of any specific intervention within it
2. The multiple interventions appeared to have little or no benefit anyway
My belief is that to hold this up as evidence for the benefits of eating a less saturated fat is scientifically untenable.
Against, this, I think it’s appropriate to take a wider view of the saturated fat/heart disease ‘link’. Last year, Mente and colleagues at the Population Health Research Institute in Hamilton, Canada, published a comprehensive review of the link between dietary factors and coronary heart disease [4]. They found that epidemiological evidence did not support a link, and neither were they able to identify appropriate intervention studies that supported any link.
A recent edition of the Annals of Nutrition and Metabolism was dedicated to reporting an expert consultation held jointly by the World Health Organization (WHO) and the Food and Agriculture Organization (FAO) of the US [5]. The consultation undertook a comprehensive review of the relationship between fats and health, and took place in late 2008. The report states with regard to epidemiological evidence that Intake of SFA was not significantly associated with CHD mortality and SFA intake was not significantly associated CHD events. With regard to interventional evidence, it was concluded that fatal CHD was not reduced by low-fat diets.
In addition, earlier this year saw the on-line publication of a meta-analysis of cohort studies assessing the relationship between saturated fat and heart disease [6]. This meta-analysis comprised 21 epidemiological studies (almost 350,000 people followed-up over a period of between 5 and 23 years). This meta-analysis found no significant association between saturated fat and risk of heart disease or stroke.
I do not doubt your motives and think it’s laudable that you would attempt to help inform individuals about healthy eating in an effort to help prevent cardiovascular disease in the public at large. However, it is important I think that such messages should be accurate and evidence-based. Otherwise, we risk doing more harm than good.
I’d be enormously grateful if you would perhaps give me your thoughts regarding the deficiencies of the North Karelia ‘evidence’, as well as the (in my view) compelling evidence which demonstrates that saturated fat does not cause heart disease.
Yours sincerely
John Briffa
References:
1. Puska P. Fat and heart disease: Yes we can make a change ” the case of North Karelia (Finland). Ann Nutr Metab 2009;54(suppl 1):33″38
2. Shah Ebrahim and George Davey Smith Editor’s Response ” exporting failure. International Journal of Epidemiology 2001;30:1496-1497
3. Salonen JT. Did the North Karelia project reduce coronary mortality? Lancet 1987;2:269
4. Mente A, et al. A Systematic Review of the Evidence Supporting a Causal Link Between Dietary Factors and Coronary Heart Disease. Arch Intern Med. 2009;169(7):659-669
5. Fats and Fatty Acids in Human Nutrition. Annals of Nutrition and Metabolism, 2009; 55 (1-3)
6. Siri-Tarino PW, et al. Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease Am J Clin Nutr 13 January 2010 [epub ahead of print publication]
48 hours on, I still have not had a response to this email. If I get one, I’ll post it. At this time, however, I think it’s fair to stay that when Mr Kolvekar embarked on his quest to save us from the ravages of heart disease, he was either unfamiliar with the science in this area, or had studiously ignored it. I am genuinely prepared to give him the benefit of the doubt, though, and imagine that Mr Kolvekar’s statements were just misguided and ill-informed. However, if he is unable to produce the evidence that supports his ‘personal opinion’, and now he has in front of him quite a stack of evidence to the contrary, surely the decent thing would be to issue a retraction?
I’m tempted to remind Mr Kolvekar of John Maynard Keyne’s quote I referred to here: When the facts change, I change my mind. What do you do, sir?
