1993. Harvard University’s Professor Walter Willett MD invents the Mediterranean Diet.
When Ancel Keys began to promote his low fat diet to prevent CHD, he advocated replacing saturated fat with industrially-produced “vegetable” oils “rich in polyunsaturated fats”. Although he was aware that polyunsaturated fats had certain limitations – limitations fully exposed in the Joint Statement of the American Heart Association Nutrition Committee and the Council on Arteriosclerosis (but since conveniently forgotten) (1,2) – he was unaware of the problems posed by trans fatty acids, yet to be identified by Enig, Kummerow and ultimately Harvard University’s Walter Willett (3).
A safe edible fat to replace “vegetable” oils that would extend the life of the Diet-Heart Hypothesis would come from an unexpected source – from Greece and Italy in the form of olive oil. It is a story that would have remained hidden but for the critical investigative work of Nina Teicholz (4, p.179-224). Here I review just the most salient points she has unearthed, but her nearly 50 page expose needs to be read in full, better to understand more precisely the origins of the more recent medical love affair with olive oil and the Mediterranean diet.
Professor Antonia Trichopoulou of the University of Athens Medical School is the person credited with bringing the Mediterranean diet to the attention of the world. Some call her the “Godmother” of the Mediterranean diet (4, p.175). Intrigued by Keys’ finding that persons living in Crete seemed to enjoy protection from CHD, she wondered if there was something special in the Cretan diet that was protective. Something more than just eating less saturated fats and more cereals. Perhaps it was many different components of the diet that were protective. Or perhaps it was the Grecian love for olive oil. Soon she began to convene scientific meetings in Greece to advance the understanding of what would soon become known as the “Mediterranean diet (5).
At the same time Professor Anna Ferro-Luzzi was asking very similar questions at the National Institute of Nutrition in Rome, Italy, and was starting specific research studies. She directed a study (6) of the effects of replacing, with olive oil, foods like butter “and other natural foods rich in saturated fat and cholesterol” (6, p.1028) in the traditional diet of the residents of Cilento, a sea-side village 200km south of Naples.
The study found that in both men and women the switch to a diet “rich” in saturated fat and cholesterol significantly increased blood total cholesterol, LDL-cholesterol and Apoprotein-B concentrations. Somehow the authors failed to notice that blood HDL-cholesterol concentrations also increased on the higher saturated fat diet, especially in women. Which meant that the “Mediterranean” diet lowered the blood concentrations of what was then considered the “good” cholesterol especially in women. This conveniently ignored finding, especially as it effects women, was discussed at length in the previous column (6).
In a subsequent publication (7), Ferro-Luzzi and Sette attempted to define exactly what is a Mediterranean diet. They began by comparing the dietary patterns of persons living in countries adjoining the Mediterranean Sea (North Africa, Greece, Italy, Yugoslavia, France, Spain, Turkey, Syria, Lebanon, Israel, Malta, Albania – Southern Europeans) with those of Northern Europeans (including Iceland, Norway, Sweden, Finland, Denmark, United Kingdom, Ireland, Germany, Switzerland, Holland, Belgium and Luxemburg, amongst others).
The key differences were that the Northern Europeans ate more meat, fish, eggs, alcohol, milk and cheese, alcohol, fats, sweeteners and vegetables whereas the Mediterraneans ate more fruits and cereals (figure 1).
Legend to figure 1: Comparison of average individual yearly intakes in 1961-1963 of 9 different foodstuffs by inhabitants of ~12 Mediterranean countries bordering the Mediterranean Sea and of 12 Northern European countries. Reproduced from figure 1 in reference 7.
The percentage differences in the annual intakes of these 9 different foodstuffs between the Mediterraneans and the Northern Europeans are shown in figure 2.
Legend to figure 2: Percentage differences in annual intakes of these same 9 different foodstuffs (figure 1) by the Mediterraneans and the Northern Europeans in 1961-1963. Reproduced from figure 2 in reference 7.
Clearly the major differences are that at the time of the study (1961-1963), the Northern Europeans were eating substantially more meat, eggs, milk and cheese, fats and sweeteners and drinking more alcohol whereas the Mediterraneans preferred more fruits and cereals.
Next the authors evaluated the commonalities in the diets of these disparate communities bordering the Mediterranean Sea. Perhaps, they wondered, what is the probability that those living in North African countries touching the Mediterranean – Egypt, Libya, Algeria, Lebanon, Morocco, Syria and Tunisia – are eating the same foods as those living in Albania and Yugoslavia?
To determine the range of food choices eaten in 16 countries bordering the Mediterranean sea, the authors compared the consumption of cereals, fruit and vegetables, and olive oil (figure 3), and meats, milk and cheese and sweeteners (figure 4) in these countries.
Legend to figure 3: Variability in consumption of cereals, fruits and vegetables and olive oil, measured as % of total daily energy intakes, in 1961-1963 in 16 countries bordering the Mediterranean Sea. Reproduced from figure 3 in reference 7.
Legend to figure 4: Variability in consumption of meats, milk and cheese, and sweeteners, measured as % of total daily energy intakes, in 16 countries bordering the Mediterranean Sea. Reproduced from figure 3 in reference 7.
Figures 3 and 4 show that there is a wide range in the contribution of these different foodstuffs to daily energy consumption in the different Mediterranean countries. For example, Egypt has the highest rate of cereal consumption, providing close to 70% of daily energy (figure 3), and the lowest rate of meat consumption (figure 4) whereas France is the opposite.
On the basis of this evidence the authors decided to describe the Mediterranean Diet on the basis of what is commonly eaten in Spain, Italy, Yugoslavia, Albania, and Greece.
Figure 5 compares the diets of those living in these five Northern Mediterranean countries with that of the Northern European countries previously described. It confirms that the key difference is that cereal consumption provides twice the amount of energy in the Northern Mediterranean than in the Northern European diet whereas the opposite applies for meat and fish, and milk and cheese. Other differences were less marked.
Legend to figure 5: Comparison of percentage dietary energy derived from eight different foodstuffs for those living in five Northern Mediterranean countries compared to those living in Northern European countries. Reproduced from figure 4 in reference 7.
After considering all the evidence, the authors concluded that the following best describes the Mediterranean diet: “A high proportion of total energy provided by cereals and a low proportion by total lipids, a high contribution to total lipids of olive oil and a relatively low proportion of milk and dairy products. However, we observe that meat products as well as milk and dairy products have a much larger share in the diet than previously reported, and that cereals are relatively less important” (7, p.22). They also noted that the diet was evolving with time.
Accordingly the authors next evaluated changes in diet patterns in southern Italy in the two decades after 1960 (figure 6). The Italian diet was found to be very similar to that of the Yugoslavians and the Greeks and so was considered a reasonable representation of dietary changes in these specific “Mediterranean” countries.
Legend to figure 6: Percentage changes in the Italian dietary profile from 1960 to 1980. Reproduced from figure 8 in reference 7.
Figure 6 shows that in the 20 years between 1960 and 1980 the southern Italian diet changed in ways that suggest a move away from what is now widely considered to be the Mediterranean diet. In particular there were very large increases in the consumption of meat and cheese with somewhat smaller increases in fish, sugar, fats, eggs and milk consumption. At the same time there were reductions in consumption of cereals, vegetables and fruit, and alcohol.
Figures 7 shows the time course of this change in the Italian diet between 1960 and 1980. It also provides a comparison of the Italian diet in 1980 with that of the Northern European countries in 1983-1985.
Legend to figure 7: Results from five dietary surveys of the Italian diet at five different periods – 1960; 1961-1963; 1963-1965; 1969 and 1980 compared with the diet of Northern Europeans in 1983-1985. Reproduced from figure 9 in reference 7.
Figure 7 shows a marked reduction in cereal consumption by Italians between 1960 and 1980, with progressive increases in consumption of fruit and vegetables, meat and fish, milk and cheese, and eggs and fats.
With the result that by the 1980s the major remaining differences between the “Mediterranean” diet and that of the Northern Europeans is the greater cereal consumption by the Mediterraneans with much smaller differences in the consumption of meat and fish, milk and cheese, eggs and fats. The impression is that any remaining differences may be even less today and will likely become even smaller in the future.
The authors conclude that these data show that: “…we are witnessing a slow but progressive transformation of the dietary habits, with a levelling out of the earlier differences between the dietary profiles of North and South Europe. This transformation is characterized by reduction in the predominance of cereal products and a simultaneous increase in foods of animal origin, especially meat and meat products, milk and dairy products and eggs. Although separated fats do not appear to have increased markedly, there has been a striking increase in total fat (consumption) because of the higher consumption of meat and dairy products” (p.25).
They concluded that “describing the Mediterranean Diet, which was supposed to be a quite easy task, has turned out to be a demanding and almost impossible enterprise since data are lacking, incomplete, or too aggregated. It appears that there is insufficient material to give a proper definition of what the Mediterranean Diet is or was either in terms of well-defined chemical compounds, or even in terms of foods”. Rather “the all-embracing term ‘Mediterranean Diet’ should not be used in scientific literature until its composition, both in foods, nutrients and non-nutrients, is more clearly defined and the metabolic basis of its health-promoting virtues has been explained” (7, p.25-26).
As they continued their research to discover the soul of the “Mediterranean Diet”, Ferro-Luzzi and Trichopoulou both promoted the value of olive oil, but they differed in the amount of dietary fat they considered to be healthy. Whilst Trichopoulou promoted what would now be considered a high-fat diet with fat providing 40% or more of daily energy intake, Ferro-Luzzi’s argued that a healthy diet should contain only between 22-27% of daily energy as fat.
Ultimately, Teicholz reports, Trichopoulou’s opinion won out. But only because she received the support of two influential Americans, Greg Drescher and Harvard Medical School’s Professor Walter Willett. The immediate connection was Trichopoulou’s husband, Dimitros who, like Willett, worked as an epidemiologist at Harvard Medical School.
After Willett and Drescher had stayed with the Trichopoulouses in Athens, the quartet decided to join forces to promote a novel “higher-fat diet, with an appealing heart-healthy promise and wrapped in the bewitching beauty of Italy and Greece, (that) could potentially have a strong appeal in America. Together they were able to move the Mediterranean diet out of its academic-conference backwater into prime time” (4, p.184).
They began by concluding that that they would base their Mediterranean diet on the two countries in Keys’ Seven Countries study that had a similar diet – Crete and southern Italy; countries that also had low rates of CHD according to Keys’ findings (8,9). This despite the fact that we now know that the dietary data in Keys’ Seven Countries Study cannot be trusted (8,9).
In 1993, Willett and Drescher organized the world’s first international conference on the Mediterranean diet in Cambridge, Massachusetts. Teicholz writes that the 150 prominent nutrition experts from the US and Europe who attended, probably expected “the usual slew of dry scientific slides on HDL- and LDL-cholesterol cross-tabulated with various kinds of dietary fat”. But they were in for a surprise: “Instead to their delight, over the next few days they were regaled with stories about Italians, olive oil and rural life on the islands of Greece” (4, p.185-186).
On the third day Willett presented his novel “Mediterranean Diet Pyramid” (10) depicted in Figure 8. He would subsequently write a book extolling the virtues of this eating pattern (11). The Pyramid was modelled on the US Department of Agriculture Pyramid but with some significant changes as described in Table 1. These changes are purely the work of Willett himself.
Legend to figure 8: Professor Walter Willett’s 1993 Mediterranean Diet Pyramid. Reproduced from figure 1 in reference 12.
Table 1: Comparison of the dietary guidelines for the United States Department of Agriculture 1991 Food Pyramid and Professor Walter Willett’s 1993 Mediterranean Diet Pyramid.
|Dietary Guideline||USDA 1991 Food Pyramid||Willett’s 1993 Mediterranean Diet Pyramid|
|Foundation layer||Based on high carbohydrate intake from breads, cereals, rice and pasta.||Based on high carbohydrate intake from breads, cereals, rice and pasta.|
|Second layer||Fruits and vegetables||Fruits and vegetables|
|Third layer||Milk, yoghurt and cheese. Meat, poultry, fish, dry beans, eggs and nuts.||Beans, other legumes and nuts|
|Fourth layer||Fats, oils and sweets||Cheese, yoghurt and other dairy products|
|Fifth layer||Nil||Olive oil and olives|
|Seventh layer||Nil||Poultry and Eggs|
|Ninth layer||Nil||Lean red meat|
Perhaps the distinguishing difference between the two food pyramids described in Table 1 begins with the third layer.
There, most of the dietary fat and protein in the USDA Pyramid is provided by meat, dairy, poultry and fish. In contrast, Willett’s Pyramid advocates a vegetarian/plant-based option as the premier source of protein (with carbohydrate and some fat) in the form of beans, legumes and nuts. Willett’s Pyramid relegates dairy produce to the fourth layer with all animal produce being downgraded to the lowest possible rankings with the advice that eggs, fish and poultry should be eaten only once weekly, and meat only once a month.
So the fourth and final layer of the USDA Pyramid identifies fats, oils and sweets as foodstuffs that should be used “sparingly”.
In contrast the fifth layer of Willett’s Mediterranean Pyramid encourages the liberal use of olives and olive oil as the main source of added fat in his dietary plan.
Finally, whereas the bottom 5 layers of the Willett Pyramid should be eaten daily, the top 4 layers – fish, poultry/eggs, sweets and lean red meat – should be eaten “a few times per months or somewhat more often in very small amounts”. This underscores Willett’s belief that a more plant-based, vegetarian type diet that minimizes the use of animal foods is the healthiest option for both humans and for the planet (13).
In two other non-peer reviewed publications from the same conference (14,15), Willett and his co-workers presented the scientific findings that they believe explained why the Mediterranean diet, of the type described by Nestle at the same conference – also in a non-peer reviewed article (16) – is associated with low rates of heart disease and cancer. Note, as always, that association does not prove causation.
Willett and his colleagues advanced the following reasons for the allegedly protective effects of their Mediterranean diet:
- An abundance of plant foods, associated with protection from heart disease and cancer.
- This protection can be explained by high intakes of dietary fiber, potassium, antioxidants and folic acid.
- Daily consumption of low to moderate amounts of cheese and yogurt associated with reduced prevalence of bone fractures, colon cancer, hypertension and coronary heart disease.
- Low intake of meat. This is healthy because meat intake is associated with increased rates of coronary heart disease, colon cancer, prostate cancer, breast cancer, childhood cancers, as well as cancers of the pancreas, stomach and kidney.
- Wine in moderation and with meals reduces risk of coronary heart disease.
- Use of olive oil in place of dairy and other animal fats typical of Northern European countries. The resulting lower rates of saturated fat intake are associated with lower rates of coronary heart disease and of colon, breast and prostate cancer as well as overall (total) cancer rates.
- A lower fat intake has an equivocal influence on population obesity rates.
The real problem with this “evidence” is that it is based on associational epidemiological nutritional studies which cannot and do not prove causation. In the following column (17), I describe how some believe that this research is of so little value that it should all be scrapped. In addition recent studies, reviewed in that column, have given meat a clean bill of health; much to Willett’s displeasure. If meat is indeed safe then Willett’s Pyramid is unnecessarily restrictive of meat consumption.
Teicholz describes in great detail how this conference provided Willett with the platform to begin the promotion of his Pyramid (4, p.188-189). Professor Ferro-Luzzi told her that after Willett had given his presentation and his article had been published in that non-peer reviewed supplement to the American Journal of Clinical Nutrition, paid for by the olive oil industry: “We put out the Mediterranean diet pyramid which was rough, imprecise, but gave some connotation of what was compatible with good health. When you get into policy, you forget the minutia. You forget that the ground is not quite solid, a little shaky” (4, p.189). Indeed from that moment on Willett would promote his Mediterranean Pyramid as if it was grounded on the most rigorous science (10,11). Which it was, and is not.
Using the vehicle of 50 scientific conferences between 1993 and 2004, all organized by Willett’s friend, Greg Drescher, in his capacity as a founder member of the Oldways Preservation and Exchange Trust, and with Willett’s influential support, the belief that olive oil was a key to healthy eating would be promoted as established fact to the global academic community. The funding for these conferences was provided by industry, most especially the International Olive Oil Council (IOOC) with substantial support from the Italian, Spanish and Greek governments. In those three countries, olive oil is either the first or second most important agricultural product; 60% of the arable land in Greece, for example, is dedicated to growing olives (4, p.193).
Important to this process was the entrapment of a dedicated group of Olive Oil Ambassadors who, in exchange for free travel and accommodation at these lavish olive oil “academic” conferences, were expected to promote the value of olive oil in all their writings and lectures (4, p.193-199) especially in the US. These ambassadors brought to lipophobic North Americans who, for decades, had religiously starved themselves of dietary fat, the justification that adding more fat to their diets, in the form of olive oil, would enhance their heart health.
Harvard School of Public Health was an important beneficiary of this largesse with at least one member of Willett’s department attending each of the Oldways’ fifty conferences. During this period Willett’s department would publish more than fifty papers on the Mediterranean Diet (4, p.197).
Another captured scientist was Dr Scott Grundy who, historically, is guilty of overstating the importance of cholesterol as a risk factor for CHD and the benefits of cholesterol-lowering drugs (18,19). In one study Mattson and Grundy (20) compared the effects of different fats – saturated, polyunsaturated and monounsatured fats – on blood lipid levels in those with and without hypertriglyceridemia. Perhaps the most important findings, confirming that of Ferro-Luzzi and her colleagues (6), was that the replacement of dietary saturated fats with any other form of dietary fat, reduced blood HDL-cholesterol concentrations in men. The response would likely have been worse in women (2) but women were not included in this study. The potentially adverse implications of this finding were simply ignored. To Mattson and Grundy, it was as if the finding was never made.
The dietary fats tested in this trial were kindly provided by the Institute of Shortening and Edible Oils, likely a front organization actively involved in marketing the health benefits of “vegetable” oils, an opinion with which Grundy was clearly entirely comfortable.
But what is the evidence that olive oil is a healthy-fat option?
Teicholz (4) writes that in 2003 the North American Oil Association approached the Federal Drug Agency with all the evidence then available supposedly supporting the health benefits of olive oil. The FDA concluded that of seventy-three submitted studies only four, involving a total of 117 young men, was insufficient evidence on which to base any claims.
Indeed when he formulated his own pyramid, Willett had no scientific evidence from appropriately controlled randomized clinical trials to prove the efficacy of what he was proposing. In fact his long term colleague at Harvard Medical School and usually equally-ardent advocate of the meat-free, plant-based “Planetary” diet, Dr Frank Hu, wrote in 2003 in the prestigious New England Journal of Medicine: “The (Mediterranean) diet, however, has been surrounded by as much myth as scientific evidence. There is no single ‘Mediterranean’ diet. More than 15 countries border the Mediterranean Sea, and their dietary habits, the types of food produced, and their cultures vary considerably. Moreover, the differences that Keys observed in mortality from coronary disease could well be attributed to confounding by other lifestyle-related factors, such as physical activity” (12, p.2595).
Furthermore Teicholz exposes Willett’s error in basing his diet on Keys’ fallacious data supplemented by Willett’s deeply ingrained bias against red meat: “It therefore appears that in following the Mediterranean diet, we are relying on data collected by Keys in post-war Greece from a handful of men, partly during Lent, and then distorted by Willett’s team who, like so many experts, were biased against saturated fat (and meat – my addition). Cretans in the 1960s clearly drank more milk and ate more red meat than we’ve been led to believe. Even so, it’s curious that this (Willett – my addition) diet in its day, on Crete, was not widely beloved (by the Cretans – my addition)” (4, p.220).
Subsequently (17) I review the only randomized controlled trial (21) that has compared the effects of the Mediterranean diet with those of the standard US Dietary Goals for Americans diet, as well as Atkins’ low carbohydrate diet. Those findings clearly show that the Mediterranean diet is superior to the US Dietary Goals diet but, in almost all measurements, is substantially inferior to Atkins’ low carbohydrate diet.
But perhaps the strongest evidence against Willett’s Mediterranean Diet Pyramid as the ultimately healthy diet, in part because it places such a heavy restriction on the consumption of meat and other animal products, is provided by the real world experience in some Mediterranean and European countries that have dramatically increased their meat consumption in the decades after World War II. Recall that World War II produced significant rationing of foods of animal origin; such rationing continued in the United Kingdom until 1954, nearly a decade after the termination of World War II. There is no reason to suspect that the UK was alone in such rationing. The absence of variety in the UK diet produced by animal produce led British novelist Evelyn Waugh to write that: “The food gets drearier and drearier”.
It also inspired the writings of Englishwoman Elizabeth David (22) who wished to invigorate the dreary post-War British diet with the key elements of Mediterranean cuisine. Beside the oil, the spices, the wines, the vegetables and the fruits were “the great heaps of shiny fish, silver, vermillion or tiger-striped, and those long needle fish…There are, too, all manner of unfamiliar cheeses made from sheep’s or goat’s milk; the butchers’ stalls are festooned with every imaginable portion of the inside of every edible animal (anyone who has lived for long in Greece will be familiar with the sound of air gruesomely whistling through sheep’s lung frying in oil)” (22, p.5-6)
So in the real world of Mediterranean food choices, between 1960 and 1990 Italian men began to eat ten times more meat than during the war years (23). This was associated with an average increase in height of nearly three inches (4) but without any dramatic increase in rates of coronary heart disease. Today Italians are the world’s sixth longest living nation (24); the opposite of Willett’s prediction.
Similarly since 1976 there has been a progressive decrease in cardiovascular disease (CVD) and stroke mortality in Spanish men and women (25) associated with a substantial increase in life expectancy between 1972-1982 (26). The Spanish are now the sixth longest living nation behind Hong Kong, Japan, Macao, Switzerland and Singapore, followed immediately by Italy. By 2019, Spain, followed by Italy, Iceland, Japan and Switzerland, was ranked the world’s healthiest nation according to the Bloomberg Healthiest Country Index (27).
Since 1976, Spaniards have progressively increased their consumption of meat, dairy produce, fish and fruit with decreases in the consumption of olive oil, sugar and carbohydrate-rich foods including bread, rice and potatoes. In other words, the Spanish have done precisely what Willett’s Mediterranean Diet Pyramid said they must absolutely not do if they wish to become healthier and to live longer lives.
Yet the opposite has happened. In fact, the Spanish regions that recorded the highest increases in fat consumption displayed the lowest rates of CHD mortality. Currently Spain has the highest rates of meat consumption of all European countries (28).
The authors find it “paradoxical” that “although fat and saturated fat intakes increased, these changes were not accompanied by an increase in CHD mortality rates” (25, p.1351S). They conclude that this is likely due to better medical care rather than a result of eating a diet that contains more animal produce. As a result they suggest that Spaniards should continue to eat as they had before the marked decline in CHD rates had begun. In other words they should stick closely to the dietary advice contained in Willett’s Mediterranean Diet Pyramid. Such is the power of Wilful Blindness (29).
Yet an International Task Force on the Mediterranean Diet acknowledged that “current consumption trends (especially in European Mediterranean countries) do not correspond to the traditional concept of Mediterranean dietary patterns” (30, p.928) so that “certain types of meat, traditionally presented in a less favourable light, warrant reassessment of recommendations for these products” (p.928).
For example, two decades after they were last studied in Keys’ Seven Countries Study, the Cretans, on whose supposed eating patterns Willett had based his Pyramid, were eating 53% more saturated fat from animal produce and 41% more protein (31), yet their rates of heart attack remain very low (32). I return subsequently to these recent changes in the Cretan diet.
The overriding point is that the Italians, the Spaniards and the Cretans are not the exceptions.
In mainland Greece (as opposed to Crete), fat intake has increased from about 27% in the late 1950s (33) – at the time that Keys analysed, very poorly, the Cretan diet – to between 35 (34), 36 (33) or even 45-51% (35) so that the “overall picture from food balance sheets and household budget analyses points to a substantial increase in total fat intake in Greece over the last 50 years” (36, p.807). Interestingly, this increase has come mainly from sources other than olive oil (36) (figure 9) with the result that intake of saturated fatty acids has increased substantially perhaps as much as threefold during this time (36,37).
Legend to figure 9: Time course of changes in consumption of olive oil, vegetable oils and animal fats in Greece from 1961 to 1998. Note that the increase in total fat consumption is not due solely to an increase in olive oil consumption which increased from 1965-1980 before again falling. Since 1980 the increased fat consumption has come from “vegetable” oils and animal fats with a reduction in olive oil consumption. Reproduced from figure 4 in reference 36.
This increase in fat consumption has been associated with an ~7 cm increase in height and an ~4kg increase in weight in the Greek population associated with “marked” increases in blood cholesterol concentrations (33, p.1042).
Unlike other European countries in which this increase in consumption of animal produce has been associated with improved health, Greek researchers (33, 36) argue that this increase is perhaps the direct cause of rising rates of heart disease, diabetes and breast and colon cancer
Israel is another Mediterranean country that has undergone significant recent changes in their national diet. Thus: “The Jewish population in Israel has demonstrated one of the most striking examples of declining cardiovascular (and cerebrovascular) mortality during the 1970s” (38, p.72) despite “gradually increasing fat intake and reduced intake of non-refined carbohydrates, and no apparent change in cholesterol levels and cigarette smoking habits” (p.72). Interestingly the authors note that changes in way of life and in the usual coronary risk factors “have hardly taken place in Israel and constitute a totally unsuitable explanation for the declining death rates in Israel” (p.72). Once again the real value of the usual coronary risk factors is questioned. But importantly declining CHD death rates have occurred despite an increase in dietary fat intake and a reduction in carbohydrate consumption. And this has happened despite the claim that, when studied as part of the Lipid Research Clinics Program (39,40) in the 1970s the Israeli diet (41) was a mirror of the 1977 US Dietary Guidelines proving apparently that “the implementation of a fat-restricted diet is feasible in Western free-living populations, exercising a range of dietary and marketing options” (41, p.617). But apparently an increased dietary fat intake above that recommended by the US Dietary Guidelines can be eaten without preventing a sharp decline in CHD death rates. [More recently concerns have been expressed that an abnormally high intake of polyunsaturated fatty acids, especially linoleic acid, in the Israeli diet may be harmful (42)]
The French Paradox refers to the low rates of CHD in France despite high rates of saturated fat intake (43,44) due mainly to high rates of meat consumption (28). Blood cholesterol concentrations are not low in the French (45). In fact, “the prevalence of hypercholesterolemic subjects was comparable to that of the United States…as well as two cholesterol screenings performed in two large American cities in 1985 and 1986…(but) our prevalence was lower than reported in Finland” (46, p.699).
Figure 10 (left panels) shows that with age, blood cholesterol levels rise in both Parisian men and women so that by age >55 years, >80% of both men and women have either borderline or “high” blood cholesterol (figure 2 right panels). Yet their rates of heart disease remain lower than in most European countries.
Legend to figure 10: Left panels: Total cholesterol (top panel) and HDL-cholesterol (bottom panel) concentrations increase in Parisian men and women with increasing age.
Right panels: Percent of Parisian men (top panel ) and women (bottom panel) with either borderline high or high blood cholesterol concentrations with increasing age.
Reproduced from figures 1 and 2 in reference 46.
Perhaps, after all, it is not the “cholesterol” and the diet high in saturated fat that is not causing high rates of heart disease amongst the French.
In Switzerland, currently ranked 5th healthiest nation in the world (27) and with the fourth longest life expectancy (24), deaths from all forms of heart and circulatory diseases have fallen in both men and women since 1951 (figure 10) at the same time that dietary fat intake from a variety of food stuffs has increased (figure 11).
Legend to figure 11: Annual mortality from all forms or heart disease have fallen in Spanish men and women since 1951 with the largest decreases in both men and women in “non-rheumatic heart disease and hypertension” reflecting largely decreases in cardiovascular disease.
Reproduced from figure 1 in reference 47.
Legend to figure 12: Changes in consumption of different dietary fats in Switzerland between 1951 to 1977. Consumption of pork especially but also animal fats, vegetable fats, cheese and cream increased whilst eggs and butter consumption was relatively unchanged. Milk consumption fell by close to 30%.
Reproduced from figure 4 in reference 48.
The trends in Sweden, the world’s sixth healthiest nation (27) and twelfth in the global longevity ratings (24), have been equally interesting. Dietary fat consumption increased steadily in Sweden between 1939 when it provided just 32% of daily calories, and 1960 reaching 39% where it stabilized through the 1980s (48). From 1986 to 2000, fat intake then fell to about 36% of daily energy intake in men and to about 33% in women (49). This change was associated with an ~0.8mmol/L reduction in blood cholesterol concentrations (49), some of which might have been due to the increased prescription of cholesterol-lowering medications during this time.
The growth of the low-carbohydrate movement in Sweden after 2000 has seen an increase in dietary fat intake in the next decade (to 2010) to about 40% in men and to about 38% in women. Blood cholesterol concentrations have increased by about 0.4mmol/L since 2006 in both men and women. From 1992 to 2010, carbohydrate consumption has fallen from about 48% to 42% in men and from 50% to 44% in women (49). This change has been associated with improvements “in cardiometabolic risk factor profile” (50, p.1).
A study of Swedish nutrition students between 2002 to 2017 showed similar trends with carbohydrate intake falling from 47 to 41% of daily energy intake and fat increasing from 32 to 38% (51).
The small reduction (~3%) in dietary fat intake from 1986 to 2000 was associated with 53% and 52% reductions in CHD mortality in Swedish men and women respectively (52). If Keys’ Diet-Heart hypothesis is correct, the increase in dietary fat consumption after 2000 should have reversed this trend causing an increased CHD mortality.
However All-Cause Mortality in a population of men and women in Stockholm fell by 20-30% in 2005-2010 compared to 1987-1989. Reductions in mortality from acute heart attacks fell by even more (~70% in men and ~55% in women) during the same time period (53).
Clearly if any dietary factor is to be linked to the reduced CHD mortality in the Swedes, it would have to be the progressive reduction in carbohydrate consumption since the 1990s in both men and women (figure 2 in reference 49).
Although clearly neither a Mediterranean nor a European country, Japan has shown a precisely similar reduction in CHD mortality (54) despite increased consumption of animal products (55) and despite increases in average blood cholesterol concentrations (Figure 4 in reference 39). This was discussed previously in relation to the NiHonSan study of Japanese populations living in Japan or in Hawaii or California (39).
Teicholz’s conclusion is that whilst the Mediterranean Diet has been beneficial in some ways, the focus on olive oil as the panacea, has also had an important downside: “…one of the most disturbing aspects of the Mediterranean diet pyramid is that it has intensified America’s phobia about animal fats, accelerating our flight from those ancient foods to using vegetable oils instead. And this result may have harmed health in ways that appear serious but have not been well researched – because experts have for so long been focused on the supposed dangers of eating meat and dairy instead” (4, p.224).
The initial discovery that hydrogenated vegetable oils are harmful for health was covered previously (3).
But there is plenty of other evidence that Willett’s meat-free low-saturated fat Mediterranean diet is indeed a figment of his own fertile imagination.
In the first place, Table 2 lists the daily meat consumption in men and women in 10 European countries including Spain, Italy and Greece that would constitute Mediterranean countries. Documented meat intake is high in all three countries when compared to meat intakes of “health conscious” UK citizens who are likely vegan or vegetarian. Meat intake in these countries is similar to that in other European countries like The Netherlands, Germany, Denmark, Sweden, the United Kingdom and Norway that are not part of the Mediterranean diet mythology.
Table 2: Daily meat consumption (g/day), national life expectancy rankings and healthiest nation global ratings in 10 European Countries participating in the European Investigation into Cancer and Nutrition (EPIC) nutrition calibration study (based on 24-hour recall).
|Country||Meat consumption – Men||Meat Consumption – Women||Life expectancy ranking||Healthiest Nation Global Rating|
|Spain – north||200||113||5||1|
|Italy – Ragusa||132||90||6||2|
|Italy – north||130||89||6||2|
|Spain – south||128||77||5||1|
|UK – general population||110||71||27|
|Italy – Naples||?||69||6||2|
|UK – “Health conscious”||20||16||27|
Data for this table are from references 24, 27 and 28.
It is interesting that despite high rate of meat intake these different populations rate very highly in global ratings of longevity and global health with Spain, Italy and Sweden rating particularly highly.
Second, as part of her studies of French and Spanish co-inhabitants of the Cerdanya Valley in the Pyrenees, Dr Helen MacBeth (56-58) also evaluated their dietary patterns to determine whether the separate cultures ate an identical “Mediterranean diet”. She soon learned that the diets of the Cerdanya residents differed substantially from Willett’s fabricated Mediterranean diet: “Studies of the food habits of four adjacent French and Spanish populations on either side of the Franco-Spanish border…during the 1990s showed significant statistical differences in food intake between the populations, primarily according to nation state (58). Importantly, the studies also revealed a clear divergence from the popular anglophone concept of the Mediterranean diet. Meat was eaten as regularly as in northern (European) populations by all four study populations, and far more often than fish (59); consumption of potatoes exceeded that of pasta; beer was drunk as much as wine; a large bowl of milk was traditional in the mountains for breakfast; and cheaper vegetable oils were usually used instead of olive oil for cooking in Spain, and butter often in France. Cultural diversity in food habits clearly existed in close proximity even within southern Europe and the diets of these populations differed significantly from that suggested by those nutritional epidemiologists’ concept of the Mediterranean diet, as other studies have noted (7). At the time, little attention was paid to these results” (60, p.141).
Macbeth has also reported the results of a study (59) in which she collected 7-day dietary records from 371 French and 157 Spanish inhabitants of Catalonia in March 1995. She focused particularly on the consumption of meat and fish (Table 3).
Table 3: Seven day intake frequency of meat and fish by French and Spanish inhabitants of the Eastern Pyrenees (Catalonia)
|Food stuff||Mean intake frequency/7days||% subjects eating||Mean intake
|% subjects eating|
|Red Meat Total||2.47||84.7||2.96||90.8|
|Pig Meat Total||4.75||89.8||4.35||92.5|
Reproduced from Tables 2 and 3 in reference 59.
The key finding as it relates to Willett’s mythical diet is that meat was eaten by 100% of the participants included in the analysis and the average frequency of meat eating was 9-10 times per week. Recall that in Willett’s Mythical Mediterranean Diet (figure 1), meat may be consumed only once a month. Thus Willett’s proposed intake is exactly one-fortieth of the intake of a sample of real people eating the real, not the mythical “Mediterranean” Diet.
Dr Macbeth also questioned 102 adults in her home town of Oxford, UK (59). She asked them to list 10 food items that they considered to be characteristic of the Mediterranean Diet. The percentage of the sample who listed the 19 most frequently cited foodstuffs are shown in figure 13 in descending order by percentage.
Legend to figure 13: The foodstuffs that 102 adults living in Oxford, UK, considered to be most representative of the Mediterranean diet. Reproduced from figure 1 in reference 59.
The key point shown it figure 13 is that whereas 45-50% of respondents mentioned fish or shellfish as a typical dietary component of the Mediterranean diet, not one single respondent considered meat to be one of the 10 most typical Mediterranean diet foodstuffs. This conflicts absolutely with the real world evidence presented in Table 2.
Such has been the success of Willett’s illusion that the Mediterranean diet is a plant-based diet in which meat is seldom if ever eaten.
More about the diet of the Cretans.
Let us return, finally, to an analysis of what exactly the Cretans were eating in the past before the onset of World War II and the inadequate food analyses performed on their diet by Ancel Keys during Lent when their intake of animal foods would have been restricted (4, p.40-41).
The most complete study was performed by the Rockefeller Foundation at the request of the Greek government (61). The goal was to discover how “the knowledge and skills of industrial countries might best serve and least injure an underdeveloped area such as Crete” (16, p.1314S; 61).
Clearly there was no suggestion that the study was undertaken to discover why the Cretan diet might be the world’s healthiest. Rather the study was part of a paternalist attempt to teach the Cretans how they might be lifted out of their “underdevelopment”.
Table 4 lists proportion of the daily energy provided by the different foodstuffs in the Cretan diet in 1948, three years after the end of World War II.
Table 4: Proportion of the daily energy intake provided by the different foodstuffs in the Cretan diet in 1948.
|Food group||% Daily energy|
|Table oils and fats||29|
|Pulses, nuts and potatoes||11|
|Vegetables and fruits||11|
|Meat, fish and eggs||4|
|Sugar and honey||2|
|Wine, beer and spirits||1|
Edited from table 3 in reference 16.
There data are clearly more similar to Willett’s interpretation of the Mediterranean diet (10,11,14,15,62).
But unreported by Willett and others is that the Cretans considered this to be no better than a starvation diet. Indeed Allbaugh and colleagues (61) noted that “it is important to remember the role of cereals as the basis of simple cooking and as a weapon of daily survival, because of their ‘ability to fill’ reducing hunger pangs of poor classes” (p. 450).
Allbaugh and colleagues (61) quoted one family saying that they were “hungry most of the time” (p.105). Respondents to that survey listed the following foods in order of importance “as those most desired to improve their diets: meat, rice, fish, pasta, butter, and cheese. A large majority of the respondents (72%) listed meat as their favourite food”. On the basis of these interviews, Allbaugh et al. concluded that the diet of Crete could best be improved by providing more foods of animal origin” (16, p.1315S).
In other words, the one foodstuffs that the Cretans most desired to add to their diet was the exact foodstuff that Willett wished they and the rest of the world, should remove from their diets (in order to make them healthier).
Teicholz included some even more graphic descriptions of how the Calabrians – those living in the “boot” of Southern Italy and so the closest populations to those in Crete and Sicily – viewed the role of dietary meat (4). She quotes the work of a local historian Vito Teri who wrote that the local peasants and farm laborers “considered the diet to be the scourge of poverty and expressed relentless scorn for vegetables which were considered ‘not very nourishing’” (4, p.220). Teti concluded that they “considered the lack of food…almost entirely vegetarian, as the cause of…general mortality for cases linked to nutrition, the low stature of individuals, their physical weakness, their low ability to work and psychological debility” (4, p.221).
“Indeed”, Teicholz continues, “in the 1960s, 18% of men in southern Italy were of ‘low stature’ (under 5 feet 2 inches), compared to only 5% in the north, where more animal foods were eaten. Men from Calabria who were measured when they turned up for military service from 1920 to 1960, were the shortest men in the entire country. To improve their lot, the Calabrians, like the Cretans, desired mainly one thing, as Teti described: ‘Meat is what these peasants craved, above all else….the robust man, tall and ‘erotic’, was the man who had eaten meat’” (p. 221).
These comments would have resonated with Major General Sir Robert McCarrison who spent the bulk of his medical career in India studying the nutritional choices and health of the different Indian populations (63). He would later write: “Indeed, nothing could be more striking than the contrast between the manly, stalwart and resolute races of the north – the Pathans, Baluchis, Sikhs, Rajputs and Mararattas – and the poorly developed, toneless and supine people of the east and south: Bengalis, Madrassis, Kanarese and Tranvancorians” (64, p.68).
Importantly the Pathans, like the northern Italians (table 2) are meat-eaters whereas the Bengalis, Madrassis, Kanarese are, for the most part, cereal-eating vegetarians.
But the Cretans, perhaps like the Pathans, have subsequently followed their instincts. Between 1960 and 1979 they have increased their consumption of saturated fat by 53% and of protein by 41% (31) whilst reducing their intakes of oleic acid and monounsaturated fats by 20-30% (31). This suggests a quite large increased intake of animal produce (65,66).
A study of Cretan boys (67) found that 45% of their dietary energy came from fat of which olive oil provided more than half the energy; carbohydrate intake provided 44% of energy; a diet that was similar to their grandfathers studied by Ancel Keys. But surprisingly to the authors, the Cretan boys high intake of olive oil and low saturated fat intake did not seem to produce an ideal blood lipid profile; total and LDL-cholesterol concentrations were not lower nor were their HDL-cholesterol concentrations higher than values measured in boys in Europe and the US who do not follow the “Mediterranean” diet. The authors concluded that “there is no satisfactory explanation” for their unexpected finding that “our hypothesis that a typical olive-oil-rich Cretan diet causes a relatively high HDL-cholesterol to total cholesterol ratio is not supported by the present findings” (67, p.1121;1122).
Perhaps it is because Walter Willett’s olive oil claims truly are mythical.
Are the Cretans natural born heroes?
There is an assumption in all the studies trying to link specific diets to proneness to CHD, that other features of the studied populations plays no additional role in explaining any associational relationships that are uncovered. Malhotra and O’Neill address this in their study of the people of Pioppi, the town in southern Italy where Ancel Keys built his retirement villa and the town which now hosts the official Mediterranean Diet Museum.
In researching their Pioppi Diet (68), Malhotra and O’Neill concluded that what the Pioppians ate was only part of the reason for their extraordinary health and long life expectancy. Their diet, according to the Pioppians, referred to their complete lifestyle, not just to one specific eating plan. Indeed besides noting that Ancel Keys studied the Greek Orthodox Cretans during Lent when they were fasting and avoiding animal produce (69), Sarri and colleagues from the University of Crete have asked whether it is their diet, their olive oil intake or their “Mediterranean diet” that explain the excellent health of Greek Orthodox Christians living on the island of Crete (70).
They found that the overall diet contained about 38% fat and 47% carbohydrate and that during the fasting period carbohydrate consumption increased to 60% whilst fat consumption fell to 32%. Thus fasting “decreased energy intake, dietary cholesterol and the percentage of dietary protein, total fat, saturated fatty acids and trans-fatty acids while..it increased the percentage of total carbohydrates along with the intakes of fibre, folate and iron” (60, p.281). This confirms that Ancel Keys would have reported a lower fat and a higher carbohydrate intake by the Cretans than was their standard diet. Even then, the fat intake that Keys reported in the Cretans was still a wopping 31-42% (table 1 in reference 36), not precisely low-fat.
So what of the Cretans? Is it their diet that makes them healthy? Or are they simply another usual (normal) population identical to all others whose diets have been studied?
A largely unknown feature of the Cretans is that they played an extraordinarily important role in securing an Allied victory in World War II – a role far in excess of the relatively small island and its population situated far from the major battlefields of that war.
The importance of Crete was that it was a key island in the supply chain for German actions in North Africa. But when the German forces invaded Crete in May 1941, the local population rose up in unity as one, in resistance to the German occupation, which the Cretans maintained until the end of the war, four years later. Their effect of their resistance was to delay Hitler’s invasion of Russia by just enough time to cause his invading German forces to be overtaken by the harsh Russian winter of 1941.
A remarkable achievement of the Cretan resistance was the capture and removal from the island of the German commanding officer, General Heinrich Kreipe, who quite literally disappeared without a trace despite the intense military security that should have protected him. The story is retold in graphic detail by Christopher McDougall (71).
McDougall relates that the remarkable feats of endurance achieved by the Cretan resistance including that required to spirit Kreipe from the island in the face of 100 000 German troops searching for him and his captors, could not have been achieved if those involved had been eating Willett’s “heart-healthy” diet described in Table 4: “Those Resistance fighters couldn’t have gotten their calories from starch and sugar, because it wasn’t available. If they could only eat on the run, they needed food that would provide steady caloric energy all day. Greek battlefields didn’t have Gatorade stations. Fugitives couldn’t detour in search of snacks. Survival depended on two things: choosing slow burn food and adapting your body to use it” (71, p.286).
The frequent fasting requirements of their Greek Orthodox religion would have prepared their bodies to us fat as a fuel in preference to carbohydrate.
This would have allowed the resistance fighters to live off the land. And this meant eating animal produce high in fat, particularly, but also in protein. Not Willett’s beloved cereals.
McDougall relates how the original athletic diet, pioneered by Greek mathematician Pythagoras who lived in Croton in what is now southern Italy in the sixth century BC, was also an all “flesh” diet. “To intimidate his opponents”, Pythagoras’ most famous athlete, Milo of Croton, a six-time Olympic champion whose athletic career lasted 24 years, “would consume raw bull’s meat in front of his adversary and would drink raw bull’s blood for energy and vitality”. McDougall notes that whilst Pythagoras was advising the Crotan athletes they dominated Olympic competition. But when the Croton athletes no longer followed the Pythagorian diet, their success disappeared.
The original athletic diet pioneered in southern Italy in the sixth century BC and a diet that produced multiple Olympic champions, was high in animal produce and low in what would become the cereals and grains of Willett’s 1993 Mediterranean Diet Pyramid.
The reality is that there has never been a single Mediterranean diet. Or, at least, the Mediterranean diet of modern medical myth, as contrived by Walter Willett – high in olive oil and low in meat and other animal produce – simply does not exist in the real world.
The clear evidence is that meat is a central component of what people in the Mediterranean countries eat (Table 2) and that Mediterranean countries with high meat intakes also have populations that are extremely healthy and long-lived (Table 2).
It is time finally to bury the myth that only a plant-based meat-free diet, based on Willett’s pyramid, is healthy.
- Ad Hoc Committee to Design a Dietary Treatment of Hyperlipoproteinemia. AHA Special Report. Recommendations for treatment of hyperlipidemia in adults. A Joint Statement of the Nutrition Committee and the Council on Arteriosclerosis. Circulation 1984;69:1067A-1090A.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 12. The Noakes Foundation. https://thenoakesfoundation.org/news/the-ancel-keys-cholesterol-con-part-12-1984-1993
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 10. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-10-1978-1979
- Teicholz N. The Big Fat Surprise. Why butter, meat and cheese belong in a heathy diet. Simon and Schuster, New York, NY. 2014.
- Helsing E, Trichopoulou. The Mediterranean diet and food culture: a symposium. MacMillan, London (UK). 2015.
- Ferro-Luzzi A, Strazzullo P, Scaccini C, et al. Changing the Mediterranean diet: effect on blood lipids. Am J Clin Nutr 1984;40:1027-1037.
- Ferro-Luzzi A, Sette S. The Mediterranean Diet: an attempt to define its present and past composition. Europ J Clin Nutr 1989;43(suppl 2):13-29.
- Noakes TD. It’s the insulin resistance, stupid. Part 8. CrossFit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-8
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 7. The Noakes Foundation. https://thenoakesfoundation.org/news/blog/ancel-keys-cholesterol-con-part-7-1968-1970
- Willett WC, Sacks F, Trichopoulou A, et al. Mediterranean diet pyramid: a cultural model of healthy eating. Am J Clin Nutr 1995;61(suppl):1402S-1406S.
- Willett WC. Eat, drink and be healthy. The Harvard Medical School Guide to Healthy Eating. Free Press, New York, NY. 2005.
- Hu FB. The Mediterranean Diet and mortality – Olive oil and beyond. N Engl J Med 2003;248:2595-2596.
- Zimmerman E. Tell me about the Planetary Health Diet. An interview with the Planetary Health Diet’s Walter Willett. The Cut. February 8th 2019. Available at: https://www.thecut.com/2019/02/an-interview-with-the-planetary-health-diets-walter-willett.html
- Kushi LH, Lenart EB, Willett WC. Health implications of Mediterranean diets in light of contemporary knowledge. 1. Plant foods and dairy products. Am J Clin Nutr 1995;61(suppl):1407S-1415S.
- Kushi LH, Lenart EB, Willett WC. Health implications of Mediterranean diets in light of contemporary knowledge. 2. Meat, wine, fats and oils. Am J Clin Nutr 1995;61(suppl):1416S-1427S.
- Nestle M. Mediterranean diets: historical and research overview. Am J Clin Nutr 1995;61(suppl):1313S-1320S.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 14. Crossfit Health. Please complete.
- Grundy SM. Cholesterol and coronary heart disease. A new era. JAMA 1986;256:2849-2858
- Noakes TD. It’s the insulin resistance, stupid: Part 9. Crossfit Essentials. https://www.crossfit.com/essentials/its-the-insulin-resistance-stupid-part-9
- Mattson FH, Grundy SM. Comparison of effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J Lipid Res 1985;26:194-202. – olive oil
- Shai I, Schwarzfuchs D, Henkin Y, et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med 2008;359:229-241.
- David E. A book of Mediterranean Food. The Folio Society, London, UK. 2005.
- Ferro-Luzzi A, Branca F. Mediterranean diet, Italian-style: prototype of a healthy diet. Am J Clin Nutr 1995;61(suppl):1338S-1345S.
- Anon. Hong Kong Life Expectancy 1950-2020. Macrotrends. Available at: https://www.macrotrends.net/countries/HKG/hong-kong/life-expectancy
- Serra – Majem L, Ribas L, Tresserras R, et al. How could changes in diet explain changes in coronary heart disease mortality in Spain? The Spanish paradox. Am J Clin Nutr 1995;61(suppl):1351S-1359S.
- Chenet L, McKee M, Otero A, et al. What happened to life expectancy in Spain in the 1980s? J Epidemiol Community Health 1997;51:510-514.
- Welch A. The world’s healthiest countries, ranked. CBS News. February 25th 2019. Available at: https://www.cbsnews.com/news/the-worlds-healthiest-countries-ranked/
- Linseisen J, Kesse E, Slimani N, et al. Meat consumption in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohorts: results from 24-hour dietary recalls. Publ Health Nutr 2002;5:1243-1258.
- Heffernan M. Wilful blindness. Why we ignore the obvious at our peril. Bloomsbury, USA. 2012.
- Serra-Majem L, Trichopoulou A, de la Cruz JN, et al. Does the definition of the Mediterranean diet need to be undated? Publ Health Nutr 2004;7:927-929.
- Aravanis C. The classic risk factors for coronary heart disease: experience in Europe. Prev Med 1983;12:16-19.
- Lionis CD, Koutis AD, Antonakis N, et al. Mortality rates in a cardiovascular ‘low risk’ population in rural Crete. Fam Pract 1993;10:300-304.
- Trichopoulou AD, Efstathiadis PP. Changes of nutrition patterns and health indicators at the population level in Greece. Am J Clin Nutr 1989;49:1042-1047.
- Van der Vijver LPL, Kardinaal AFM, Couet C, et al. Association between trans fatty acid intake and cardiovascular risk factors in Europe: the TRANSFAIR study. Eur J Clin Nutr 2000;54:126-135.
- Trichopoulou AD, Lagiou P. Healthy traditional Mediterranean Diet: an expression of culture, history, and lifestyle. Nutr Rev 1997;55:383-389.
- Ferro-Luizzi A, James WPT, Kafatos A. The high-fat Greek diet: a recipe for all? Eur J Clin Nutr 2002;56:786-809
- Kafatos A, Mamalakis G. Changing patterns of fat intake in Crete. Eur J Clin Nutr 1993;47 Suppl 1:S21-S24.
- Goldbourt U, Neufeld HN. Trends in coronary heart disease mortality and related factors in Israel. Cardiology 1985;72:63-74.
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 8. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-8-1970-1974
- Noakes TD. Ancel Keys’ Cholesterol Con: Part 11. The Noakes Foundation. https://thenoakesfoundation.org/news/ancel-keys-cholesterol-con-part-11-1979-1984
- Kaufmann NA, Dennis BH, Heiss G, et al. Comparison of nutrient intakes of selected populations in the United States and Israel: the lipid research clinics prevalence study. Am J Clin Nutr 1986;43:604-620.
- Dubnov G, Berry EM. Omega-6/omega-3 fatty acid ratio: The Israeli Paradox. World Rev Nutr Diet 2003;92:81-91.
- Ferrieres J. The French Paradox: lessons from other countries. Heart 2004;90:107-111.
- Artaud – Wild SM, Connor SL, Sexton G, et al. Differences in coronary mortality can be explained by differences in cholesterol and saturated fat intakes in 40 countries but not in France and Finland. A paradox. Circulation 1993;88:2771-2779.
- Ducimetiere P, Richard JL, Cambien F, et al. Coronary heart disease in middle-aged Frenchmen. Comparisons between Paris Prospective Study, Seven Countries Study, and Pooling Project. Lancet 1980;1:1346-1350.
- Laurier D, Chau NP, Segond P, et al. Cholesterol and other cardiovascular risk factors in a working population in Ile-de-France (France): first results of the PCV-Metra Study. Eur J Epidemiol 1992;8:693-701.
- Guberan E. Surprising decline of cardiovascular mortality in Switzerland: 1951-1976. J Epidemiol Comm Health 1979;33:114-120.
- Becker W. Fat consumption in Sweden – recent trends illustrated by food balance sheet data. Naringsforskning 1990;34:70-73.
- Johansson I, Nilsson LM, Stegmayr B, et al. Associations among 25-year trends in diet, cholesterol and BMI from 140,000 observations in men and women in Northern Sweden. Nutr J 2012;11:40.
- Winkvist A, Klingberg S, Nilsson LM, et al. Longitudinal 10-year changes in dietary intake and associations with cardio-metabolic risk factors in the Northern Sweden Health and Disease Study. Nutr J 2017;16:20.
- Bergstrom M, Hakansson A, Blucher A, et al. From carbohydrates to fat: Trend in food intake among Swedish nutrition students from 2002 to 2017. PLoSONE 2020;15(1):e022820.
- Bjorck L, Rosengren A, Bennett K, et al. Modelling the decreasing coronary heart disease mortality in Sweden between 1986 and 2002. Eur Heart J 2009;30:1046-1056.
- Journath G, Hammar N, Elofsson S, et al. Time trends in incidence and mortality of acute myocardial infarction, and all-cause mortality following a cardiovascular prevention program in Sweden. PLoSONE 2015;10(11):e0140201.
- Sekikawa A, Miyamoto Y, Miura K, et al. Continuous decline in mortality from coronary heart disease in Japan despite a continuous and marked rise in total cholesterol: Japanese experience after the Seven Countries Study. Cardiovasc Epidemiol 2015;1614-1624.
- Yoshiike N, Matsumura Y, Miura K, et al. National nutrition survey in Japan. J Epidemiol 1996;6:S189-S200.
- Macbeth H, Salvat M, Vigo M, et al. Cerdanya: Mountain valley, genetic highway. Ann Hum Biol 1996;23:41-62.
- Salvat M, Vigo M, Macbeth H, et al. Seasonality of marriages in Spanish and French parishes in the Cerdanya valley, eastern Pyrenees. J Biosoc Sci 1997;29:51-62.
- Macbeth H. The Cerdanya, a valley divided: biosocial anthropology in a research project. In: Boyce AJ, Reynolds V. Human Populations: Diversity and Adaptation. Oxford University Press, Oxford, UK, 1995:233-251.
- Macbeth H. Concepts of ‘The Mediterranean Diet’ in UK and Australia compared to food intake studies in the Catalan coast of the Mediterranean. Rivist di Anthropol 1988;76(suppl):307-313.
- Macbeth H, Medina FX. New views about the Mediterranean diet. Nature Med 2020;1:141-142.
- Allbaugh LG. Crete: a case study of an underdeveloped area. Princeton, NJ: Princeton University Press, 1953.
- Willett WC. Diet and Health: What should we eat? Science 1994;264:532-537.
- McCarrison R. Nutrition and Health. The McCarrison Society, London UK. 1982; McCarrison R. Studies in Deficiency Disease. Henry Frowde and Hodder and Stoughton, London, UK. 1945.
- Sinclair HM. The Work of Sir Robert McCarrison. Faber and Faber Ltd, London, UK. 1953.
- Kafatos A, Verhagen H, Moschandreas J, et al. Mediterranean Diet of Crete: Food and nutrient content. J Am Diet Assoc 2000;100:1487-1493.
- Kafatos A, Kouroumalis I, Vlachonikolis I, et al. Coronary-heart-disease risk-factor status of the Cretan urban population in the 1980s. Am J Clin Nutr 1991;54:591-598.
- Aravanis C, Mensink RP, Maralias N, et al. Serum lipids, apoproteins and nutrient intake in rural Cretan boys consuming high-olive-oil diets. J Clin Epidemiol 1988;41:117-1123.
- Malhotra A, O’Neill D. The Pioppi Diet. A 21-Day Lifestyle Plan. Penguin Random House, London, UK, 2017.
- Sarri K, Kafatos A. Letter to the Editor: The Seven Countries Study in Crete: olive oil, Mediterranean Diet or fasting? Publ Health Nutr 2005;8:666
- Sarri KO, Linardakis MK, Bervanaki FN, et al. Greek Orthodox fasting rituals: a hidden characteristic of the Mediterranean diet of Crete. Brit J Nutr 2004;92:277-284.
- McDougall C. Natural born heroes. The lost secrets of strength and endurance. Profile Books, London, UK. 2015.
About The Author
Professor Tim Noakes has dedicated his life to the pursuit of knowledge and undoing the last 50 years of ‘bad’ nutritional science. His aim is to fix the future outlook of human health, by changing the way people eat and the food policies to enable the change.
Prof. Noakes has published more than 750 scientific books and articles. He has been cited more than 19 000 times in scientific literature, has an H-index of 71 and has been rated an A1 scientist by the National Research Foundation of South Africa for a second 5-year term. He has won numerous awards over the years and made himself available on many editorial boards.