The Seven Countries Study – Part 1

Response to The True Health Initiative White Paper

On 1st August 2017, a ‘white paper’ was published called “Ancel Keys and the Seven Countries Study: An Evidence-based Response to Revisionist Histories.” It was commissioned by an organisation called The True Health Initiative. It was written by Katherine Pett, Joel Kahn, Walter Willett and David Katz. Calling one’s work a white paper (typically the preserve of government) is an arrogant start. However, the business definition of ‘white paper’ is “A report that informs readers about an issue, which is often used to convey an organization’s philosophy and persuade potential customers.” On that basis, let’s review this white paper…

The True Health Initiative

David Katz is the founder of The True Health Initiative. Two other authors, Joel Kahn and Walter Willett, are members of the council of the organisation.

I personally agree with the vision and mission statements of The True Health Initiative:

“Our vision is a world where all people live long and healthy lives, free of preventable chronic disease.

“Our mission is to create a culture free of preventable chronic disease by demonstrating and disseminating the global consensus on the fundamental, evidence-based truths of lifestyle as medicine.”

I also agree on five of the six principles offered as a solution: Feet (regular physical activity); Fingers (avoiding smoking/excess alcohol); Sleep (adequate quality & quantity); Stress (management of); and Love (having meaningful supportive relationships). I also agree with part of the dietary advice: “Forks”. This is expanded upon as “A diet comprised mostly of minimally processed, generally plant-predominant foods in balanced combinations.”

When asked what should we eat? I answer i) Eat real food and ii) Choose that food for the nutrients it contains. The True Health Initiative and I would agree on (i) and thus we would agree on “minimally processed”. We would disagree on (ii). The foods richest in essential fats, complete protein, vitamins and minerals come from animals, not plants. The True Health Initiative was founded by a proponent of a plant-based diet, David Katz and it has many of the high priests of the vegan/plant-based world on its council: Neal Barnard; T. Colin Campbell (China Study); Garth Davis, Caldwell Esselstyn, Christopher Gardner, Joel Kahn, Virginia Messina and more. This is The Pledge. I was vegetarian for approximately 20 years until I studied the evidence base for nutrition. In my experience, being vegetarian/vegan is akin to religion. It is a belief system in itself – it is not the logical outcome of studying nutritional science.

The esteem with which the Keys Seven Countries Study is held by vegans/vegetarians has its roots in a long standing nutritional error, which prevails today. Keys used the terms “animal fat” and “saturated fat” interchangeably, as if they were the same. It was not until c. 1956 that Keys started to be accurate in his terminology (Ref 1). Animal fats do not equal saturated fats any more than plant fats equal unsaturated fats. All foods that contain fat – from animal and plant sources – contain all three fats: saturated; monounsaturated and polyunsaturated. Gram for gram, olive oil (Ref 2) contains seven times the saturated fat of a typical sirloin steak (Ref 3). The single greatest food source of saturated fat is the coconut – perfectly vegan. Meat, fish, eggs and lard all contain more unsaturated than saturated fat. The attack on saturated fat, falsely deemed synonymous with animal fat, is ignorant and unhelpful in the advancement of the “minimally processed” beliefs that vegans, vegetarians and omnivores can share. The primary sources of saturated fat are documented in the US as pizza, desserts, candy, potato chips, pasta, tortillas, burritos, tacos, hot dogs and other processed foods. Similarly in the UK, bread, cakes, buns, pastries, biscuits, cereals, confectionery and other processed foods are the primary sources of saturated fat. I thus proposed a way forward in this paper (Ref 4) that real food proponents should unite in our opposition to processed food, but stop demonising saturated fat (where found naturally in foods of animal, or plant, origin) in the name of processed food.

This response to the True Health Initiative white paper will be in two parts: Part 1) to review the allegations made in the white paper and Part 2) to present the real criticisms of the Seven Countries Study, which the white paper did not mention. This is Part 1. Part 2 will follow.

Part 1 – The White Paper

The abstract (p4) of the white paper sets out why it was written. The authors opened with the sentence: “Disparagements of the methods, intentions, and conclusions of the Seven Countries Study [SCS] are currently much in vogue.” They continued: “Critics frequently point out alleged flaws in the seminal study in order to contest its primary dietary finding, that saturated fat was correlated with heart disease.” They closed with: “Popular criticisms directed at the study, and the lead investigator, Ancel Keys, turn out to be untrue when the primary source material is examined.”

I have also examined the primary source material in great depth, as it was central to my Ph.D. in the evidence base for dietary fat guidelines (2012-2016) and research for my 2010 book The Obesity Epidemic. I too find some criticism of Ancel Keys untrue (and unfair), so I read the white paper with interest.

The white paper was focused on “four primary allegations”, which it claimed had been proposed to discredit Keys and which it claimed were wrong (I’ve taken the fuller allegations from p20 of the report, rather than the shorter version on p6-7):

1. “That Keys, et al. selected countries with a specific outcome in mind; that SCS had data from 22 countries but highlighted only seven.”

2. “France was deliberately excluded because the SCS research team was aware of the ‘French Paradox.’”

3. “That dietary surveys made in Greece were invalid because they occurred during Lent.”

4. “That an analysis of SCS data published in 1999 proved that sugar, and not saturated fat, was associated with coronary heart disease.”

I will address each of these in turn…

Allegation 1

“1. That Keys, et al. selected countries with a specific outcome in mind; that SCS had data from 22 countries but highlighted only seven.”

No specific references were given to clarify against whom this allegation was made. Instead the allegation was generally levelled at: “Speakers, bloggers, journalists, and other detractors of SCS” (p22). Notwithstanding the vagueness, there are two parts to Allegation 1 and I agree that both are made by a number of people. I agree that the second part (that SCS had data from 22 countries but highlighted only seven) is made in error. A number of people have confused the Keys’ six countries graph from the Mount Sinai presentation (1953) (Ref 5) with the Seven Countries Study (1956-1970) (Ref 6). I have always seen this as an easy test to see if someone knows what they’re talking about. Sadly, so many people get it wrong that I felt compelled to write a post about it

I disagree that the first part (That Keys, et al. selected countries with a specific outcome in mind) is made in error. I think that this criticism of Keys is fair. The six countries graph showed two important things:

i) Keys did exclude much of the available data to produce a graph of 6 countries (when data from 22 countries were available (Ref 7)) and this tells us something about his scientific integrity.

ii) Keys knew which countries to include in the Seven Countries Study. The six countries graph had Japan on the bottom left, USA on the top right and Italy firmly on the line. Keys could be confident, therefore, going into an expensive and lengthy study that his hypothesis on dietary fat and heart disease was likely to be supported.

The white paper devoted a number of pages (p13-16) to trying to justify the countries chosen. The telling comment was the final one: “Countries with willing researchers and ready funds were welcome.” The reference for this was “Blackburn H. personal communication. 2017.”

In my Ph.D. thesis, I also referenced the work of Blackburn for the rationale for the countries chosen…

“Blackburn acknowledged that “The study has been criticized for the method in which populations were selected for the study” (p2) (Ref 8). A book written in tribute to Keys for his 90th birthday documented that many of the countries were chosen because researchers were known to be examining similar issues to Keys: Martti Karvonen in Finland; Flaminio Fidanza and Vittorio Puddu in Italy; Ratko Buzina and Bozidar Djordjevic in what was known as Yugoslavia; Christos Aravanis and Anastasios (Andy) Dontas in Greece; Noboru Kimura and Hironori (Hiro) Toshima in Japan; and Frans Van Buchem and Louise Dalderup in The Netherlands (p11) (Ref 9). Keys had collaborated with Buzina, Fidanza, Karvonen and Kimura prior to the start of The Seven Countries Study.

“Zutphen was chosen by the Dutch researchers “because it offered practical advantages of local hospitals and physicians and had been the site of earlier health surveys” (p5) (Ref 10). The USA railroad industry was selected “because the pensions, death and disability benefits of the Railroad Retirement Board provide powerful incentives to maintain contact with the board” (p4) (Ref 10). The Rome railroad men were chosen “to provide a European sample to compare with the American railroad men” (p5) (Ref 10). For the selection of Crevalcore and Montegiorgio, “it was enough to have the assurance of our colleague Flaminio Fidanza that we would have good cooperation in those rural areas and that the cost of the work would be within our resources” (p6) (Ref 10). Similarly, “friendship with our colleague Ratko Buzina led to selection of areas of Dalmatia and Slavonia where he had friends to help and where, again, we could expect low cost and high cooperation” (p6) (Ref 10). In Japan “Noburu Kimura, who had once worked with us in Minnesota, offered enthusiastic help at little cost” (p7) (Ref 10). “Similarly, collaboration with Martti J. Karvonen, led to the choice of Finland.” “In Greece, friends A.S. Dontas and Christ Aravanis led us to Crete” (p7) (Ref 10).”

I concur with Blackburn and the white paper therefore that “Countries with willing researchers and ready funds were welcome” (P16). I’m sure they were. But if I planned a study today working with Dr Gary Fettke in Tasmania, Professor Tim Noakes in Cape Town, Dr Aseem Malhotra in Pioppi, Dr Jeff Gerber in Denver, Dr Caryn Zinn in Auckland, Dr Andreas Eenfeldt in Uppsala and Dr Trudi Deakin in Manchester, all of whom are known to share similar views on dietary fat, I’m sure I’d be (rightly) criticised.

Allegation 2

2. “France was deliberately excluded because the SCS research team was aware of the “French Paradox.”

I have never seen this “historical revision”, as the white paper calls it. No reference was given in the paper for this allegation. If it is an issue, it would be covered by Allegation 1 – France was one of 16 countries for which data were available, but it wasn’t included in Keys’ 6 countries graph. I would agree with the white paper that the “French Paradox” would not have been known about in 1956 (when the Seven Countries Study was in preparation). The data available for France, which was excluded from the six countries graph, would have suggested that France would have been an outlier. However, in 64 pages of the white paper there is no reference to any source of this allegation, so I don’t know why any effort has been taken to refute something that no one appears to have alleged. This doesn’t warrant any further consideration.

Allegation 3

3. “That dietary surveys made in Greece were invalid because they occurred during Lent.”
There is one reference for this allegation: The Big Fat Surprise book by Nina Teicholz. Teicholz is the only person, of whom I am aware, who has raised Lent as an interesting aspect of the Seven Countries Study. In a 480-page book, Teicholz devotes 3 paragraphs to the Greek cohorts in the Seven Countries Study and Lent. She does not allege that the dietary surveys in Greece were invalid. She questioned how Lent would have affected the dietary data and quoted a study conducted in Crete in 2000 and 2001 showing that saturated fat consumption halved during Lent (Ref 11).

I found no evidence for the claim, repeated three times, in the white paper: “Sampling during Lent, rather than being a researcher oversight, was a purposeful choice” (p37). The white paper noted the extract from the Den Hartog publication: “The Greek Orthodox rules are much more strict than those of the Roman Catholic Church and the ‘fasting’ period of eight weeks should, theoretically, have a major impact on the total diet for the year. However… actual practice is different from Church prescription” (p12) (Ref 12). Teicholz reported this similarly, quoting Keys as having said “strict adherence [to Lent] did not seem to be common.”

Teicholz, as a journalist would do, then reported the views of two other researchers, Katerina Sarri and Anthony Kafatos (Ref 13), who considered the Lent issue “a remarkable and troublesome omission.” Teicholz, as a journalist would do, then contacted Daan Kromhout, who directed the nutritional component of the Seven Countries Study. With reference to Lent, Kromhout acknowledged “In an ideal situation, we should not have done that.”

Having reviewed the 20 volumes of the Seven Countries Study published in Circulation and the Den Hartog dietary document in great detail for my Ph.D., the self-reported dietary survey information suggested that Lent didn’t make the difference that might have been expected. This is a reason why Teicholz reported in The Big Fat Surprise (p216-217) that Keys himself had concerns about the reliability of the self-reported dietary information. Keys’ concern about the validity of the Den Hartog data could also explain why it was published in relative obscurity.

Volume XVII of the Circulation publication reported the average proportion of the Corfu and Crete diets as 7% and 8% saturated fat respectively (fig. XVII.1, pI-168) (Ref 14). The primary Greek Island issue is not Lent, but that, with essentially the same saturated fat intake, coronary heart disease (CHD) deaths per 100,000 people were more than five times higher in Corfu than Crete. This alone undermined the saturated fat/CHD finding from the SCS.

Allegation 4

4. “That an analysis of SCS data published in 1999 proved that sugar, and not saturated fat, was associated with coronary heart disease.”

The white paper charged three people with this allegation: Teicholz again (for some reason, Katz is frequently personal, nasty and unprofessional when it comes to the author of The Big Fat Surprise); Dr Robert Lustig, for his book Fat chance; and journalist Ian Leslie, who wrote an article on sugar in The Guardian (7 April 2016).

The “analysis of SCS data published in 1999” is a reference to the paper by Menotti et al (Ref 15). Table 2 from this paper reported the linear correlation coefficients and mortality from CHD as 0.821 with sweets (defined in the paper as sugar products and pastries) and 0.798 with animal foods (defined in the paper as butter, meat, eggs, margarine + lard (although margarine could have a vegetable origin), milk (using the weight of its solid part) and cheese).

Teicholz reported this as: “In 1999, when the Seven Countries Study’s lead Italian researcher, Alessandro Menotti, went back twenty-five years later and looked at data from the study’s 12,770 subjects, he noticed an interesting fact: the category of foods that best correlated with coronary mortality was sweets. By ‘sweets’, he meant sugar products and pastries, which had a correlation coefficient with coronary mortality of 0.821…” “By contrast, ‘animal food’ (butter, meat, eggs, margarine, lard, milk and cheese) had a correlation coefficient of 0.798, and this number likely would have been lower had Menotti excluded margarine.”

In this passage, Teicholz has been completely faithful to the original article. She has not said that saturated fat was not associated with coronary heart disease. She has not used the word sugar, as accused.

Lustig’s book referred to the SCS once (p110-111). Lustig referenced the Seven Countries book (1980), which is much referenced by the white paper (Ref 10) and he quoted (verbatim) from p262 of this book: “The fact that the incidence of coronary heart disease was significantly correlated with the average percentage of calories from sucrose in the diets is explained by the intercorrelation of sucrose with saturated fat.” Lustig asked: “So which was it – the fat or the sugar?”

Lusitg has been completely faithful to the original publication. He has not said that saturated fat was not associated with coronary heart disease. He posed a question.

The relevant quotation from Leslie’s article is: “Years later, the Seven Countries study’s lead Italian researcher, Alessandro Menotti, went back to the data, and found that the food that correlated most closely with deaths from heart disease was not saturated fat, but sugar.” Leslie should have said “sweets”, not “sugar”, but he also did not say that saturated fat was not associated with coronary heart disease.

So one of the three accused used the word sugar, when he should have used the word sweets to be precise. Forgive me for being underwhelmed.

The white paper went off topic in this sugar section, into the row between Keys and Professor John Yudkin about fat vs. sugar. Ignoring this, to stick to the white paper’s own topic of ‘criticism of the SCS study’, the authors referred to the correlation coefficients in Table 2 of the Menotti et al paper (Ref 15) to argue that the sugar line alone has a correlation score of r=0.6. This is true, but cheese (one of the few foods with more saturated than unsaturated fat) has a score of r=0.4, so what’s the point?

Sugar, as a line item in dietary surveys, invariably means bought sugar (for baking and hot drinks) and this in no way reflects sugar intake in the diet. The “sweets” category is almost certainly a more accurate way to capture sugar than the “animal food” category is a way to capture saturated fat. Meat, eggs, margarine + lard all have more UNsaturated than saturated fat. Dairy products (milk, butter and cheese) are from the only food group with more saturated than unsaturated fat. Fish and oils both contain much saturated fat and both were inversely correlated with heart disease.

All of this misses a critical point. The fact that the Menotti et al paper, a review of the food intake in the SCS 25 years on, does not have a line item for saturated fat alone and a line item for sucrose alone tells us how useless the SCS was as a dietary survey. To be still using “sweets” or “animal foods”, as approximations for sucrose or saturated fat, is an admission that the SCS findings on saturated fat were not the outcome of accurately studying saturated fat.

There is another critical point. The Keys observation that sucrose and saturated fat were so closely correlated with each other (this was quantified as r=0.84 (Fig 3) (Ref 16)) suggests that the SCS observations were confounded by processed food. Sucrose and fat (saturated or unsaturated – no food has one without the other) rarely appear together in significant amounts in natural food. Exceptions would be nuts, seeds and avocados (and none of these could be described as sweet). Unless the SCS was a study of the impact of nuts, seeds and avocados consumption in seven countries, the observed correlation between sucrose and saturated fat (and heart disease) has been confounded by processed food. This was confirmed in the Seven Countries Study 1970 publication, volume XVII: “The check list, aimed particularly at sources of fat, included milk, eggs, butter, cheese, cake, ice cream, and so on.” (pI-164)(Ref 14).

In Summary

The white paper focused on four allegations, which the authors claim are made and are false:

1. “That Keys, et al. selected countries with a specific outcome in mind; that SCS had data from 22 countries but highlighted only seven.”

2. “France was deliberately excluded because the SCS research team was aware of the ‘French Paradox.’”

3. “That dietary surveys made in Greece were invalid because they occurred during Lent.”

4. “That an analysis of SCS data published in 1999 proved that sugar, and not saturated fat, was associated with coronary heart disease.”

I agree that the second part of Allegation 1 is a valid one. The evidence shows it to be a fair criticism. At least one blog already exists clarifying this. I agree that the first part of Allegation 1 is made, but I disagree that it is false.

My research suggests that Allegation 2 is invalid and the authors provide no evidence to the contrary.

Allegation 3 is invalid. The one reference given to support it did not allege that the dietary surveys made in Greece were invalid.

Of the three references given for Allegation 4, one should have used the word “sweets” instead of “sugar”.

This hardly warrants a white paper.

In Addition

There are three other points worthy of mention:

1) Other studies of the time.

On p5, the white paper stated: “Ultimately, SCS suggested a link between dietary intake, specifically saturated fat, and heart disease. This conclusion, which corroborated other clinical and epidemiological evidence at the time, generated numerous hypotheses and has since inspired countless clinical trials.”

This is not correct. The finding between saturated fat intake and heart disease was not replicated by other clinical or epidemiological evidence at the time (Refs 17, 18). As will be shown in Part 2, the SCS was quite alone in its finding on saturated fat.

2) The SCS subjects at baseline.

On P11, the white paper stated: “The Seven Countries Study is a type of observational study known as a “prospective cohort study.” For a cohort study to examine outcomes such as coronary heart disease, it must enroll participants who do not possess the condition at baseline and then collect data on these participants and cohorts over time.”

This is not correct. The Seven Countries Study did include men with coronary heart disease at baseline. This inclusion generated an important finding. At the first five year review, 588 of the 12,770 men had died – 158 from coronary heart disease (CHD). Of the 129 with a diagnosis of definite previous myocardial infarction (MI) at study entry, 27 died within 5 years. Of the 12,641 men not diagnosed with definite previous MI at study entry, 131 died within 5 years (pI-187) (Ref 19). The death rate from CHD for those with pre-existing MI was 20.9%. The death rate from CHD for those without previous MI was 1.0%. The most significant finding of The Seven Countries Study was that the biggest cause of mortality from CHD is having heart disease. People laugh when I say this at conferences, but it was an important finding – it was just not reported as such.

3) If all else fails, accuse people of denial.

On p51-52 of the white paper, the authors reveal their true colours… “Detractors invite false equivalence by allowing studies of smaller impact and dubious quality to be compared against large-scale, scrupulously conducted research. Current examples of the insidious dangers of such a process include climate change denial, and false allegations about the effects of childhood immunization…The public health impact of false or misleading narratives about nutrition research is potentially much larger even than that of vaccines.”

A ‘white paper’ is “used to convey an organization’s philosophy and persuade potential customers.”

I rest my case.

References
Ref 1. Keys A. The diet and the development of coronary heart disease. J. Chronic Dis. 1956.
Ref 2. United States Department of Agriculture ARS. Oil, olive, salad or cooking, 2013.
Ref 3. United States Department of Agriculture ARS. Beef, bottom sirloin, tri-tip, separable lean only, trimmed to 0″ fat, choice, raw [URMIS #2244], 2013.
Ref 4. Harcombe Z. Dietary fat guidelines have no evidence base: where next for public health nutritional advice? Br. J. Sports Med. 2016.
Ref 5. Keys A. Atherosclerosis: a problem in newer public health. J. Mt. Sinai Hosp. N. Y. 1953.
Ref 6. Keys A. Coronary heart disease in seven countries I. The study program and objectives. Circulation 1970.
Ref 7. Yerushalmy J, Hilleboe HE. Fat in the diet and mortality from heart disease; a methodologic note. N. Y. State J. Med. 1957.
Ref 8. Blackburn H. On the trail of heart attacks in seven countries. University of Minnesota, 1995.
Ref 9. Kromhout D, Menotti A, Blackburn H. The Seven Countries Study: A scientific adventure in cardiovascular disease epidemiology. The Netherlands: Brouwer Offset bv, Utrecht, The Netherlands, 1993.
Ref 10. Keys A. Seven countries: a multivariate analysis of death and coronary heart disease: Harvard University Press, 1980.
Ref 11. Sarri KO, Linardakis MK, Bervanaki FN, Tzanakis NE, Kafatos AG. Greek Orthodox fasting rituals: a hidden characteristic of the Mediterranean diet of Crete. The British journal of nutrition 2004.
Ref 12. den Hartog C, Buzina K, Fidanza F. Dietary studies and epidemiology of heart diseases. The Hague: Stichting tot Wetenschappelijke Voorlichting op Voedingsgebied, 1968.
Ref 13. Sarri K, Kafatos A. The Seven Countries Study in Crete: olive oil, Mediterranean diet or fasting? Public Health Nutr. 2005.
Ref 14. Keys A. Coronary heart disease in seven countries XVII. The Diet Circulation 1970.
Ref 15. Menotti A, Kromhout D, Blackburn H, Fidanza F, Buzina R, Nissinen A. Food intake patterns and 25-year mortality from coronary heart disease: cross-cultural correlations in the Seven Countries Study. The Seven Countries Study Research Group. European journal of epidemiology 1999.
Ref 16. Keys A. Sucrose in the diet and coronary heart disease. Atherosclerosis 1971.
Ref 17. Harcombe Z, Baker JS, Davies B. Evidence from prospective cohort studies did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review. Br. J. Sports Med. 2016.
Ref 18. Harcombe Z, Baker JS, Cooper SM, et al. Evidence from randomised controlled trials did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review and meta-analysis. Open Heart 2015.
Ref 19. Keys A. Coronary heart disease in seven countries Summary. Circulation 1970.