Stunning: Saturated fat and the European paradox

Saturated Fat and the Pan-European Paradox

Wow. This is mindblowing.

Have you heard about the French Paradox? French people traditionally eat a lot of saturated fat, like butter – yet they generally have less heart disease than other populations. A lot of brainpower has been wasted to explain this – does perhaps the red wine protect them?

But maybe it’s not really a paradox.

That’s it’s been seen as a paradox in the past is probably because older observational studies have shown weak associations between saturated fat consumption and increased risk of heart disease. However, there are many explanations for these findings, such as “the healthy user effect.” People who adhere to other health-related behaviors are also likely to avoid saturated fat because they’ve been told it is “unhealthy.” But there’s no way to know if their good health is due to avoiding saturated fat or is the result of all of those other behaviors – or caused by something else entirely.

That’s because observational studies can only show associations; they can’t show cause-effect relationships. There may be other factors besides dietary choices or health-related behaviors that cause the outcomes seen in these older studies.

I was just shown the diagram above, recently published in the journal Nutrition. It’s based on WHO and FAO statistics over the average intake of saturated fat in 41 European countries in 1998 (the latest available data), and the age-adjusted risk of dying from heart disease. I added some explanations.

More saturated fat, less heart disease

It’s a stunner. The French paradox is actually a French-Swiss-Icelandic-Swedish-German-Austrian-etc.-paradox!

  1. France eats the most saturated fat and has the lowest rate of heart disease deaths in all of Europe.
  2. Switzerland eats second-most saturated fat and has the second-lowest mortality.
  3. The countries eating more saturated fat have less heart disease, period.

Less saturated fat, more heart disease

And the countries eating less saturated fat? Like Georgia, Moldavia, Azerbaijan etc.? Well, they seem to have the highest mortality from heart disease in Europe.

It’s a Pan-European paradox now.


No need to hold the butter?

What does it mean?

Correlations between populations, like these, are known as ecological data. It doesn’t really prove anything. In other words, the diagram above does not prove that saturated fat protects you from heart disease. There are obviously many other differences between these populations, not just the intake of saturated fat.

But a diagram like this can provide a counter-argument to the observational studies mentioned above. It’s unlikely that saturated fat would be a major cause of deaths from heart disease, when European populations stuffing themselves with it have fewer deaths from heart disease, without exception.

Can this possibly be a weird coincidence? Can saturated fat still possibly be bad? What do you say?


When I recently interviewed professor Loren Cordain about our hunter-gatherer ancestors, his guess was that they, on average, got about 15 percent of their calories from saturated fat.

If that’s true, it means that our genes should be well adapted to eating about 15 percent saturated fat. That’s more than twice as much as the maximum set in the advice from the USDA and others. Yet it is about as much as consumed by the populations in Europe today with the best heart health. Coincidence?

More: The Paleo Diet Explained

1 2 3 4


  1. Jeff

    Read Daniel Steinbergs Cholesterol Wars: The Cholesterol Skeptics vs the Preponderance of Evidence

    Average is not necessarily normal, and people who maintain cholesterol levels under 150mg/dl consistently most of their lives do not need statins. In order to get under this threshold most people would need to consume high fiber, close to zero cholesterol diet, and since no one is doing that in the Western world, statins are needed. Ornish proved that not even statins are needed when aggressive dietary interventions are being applied. Esselstyn uses statins, because he deals with more high risk patients, he comes soon with a new paper with big sample size. The results are the same if not better the in the first one.

    All the big statin merchants have lost exclusive licenses, so it's not a probitable business, anymore.

    In the time frame of 40 years, only 5-6 people with cholesterol levels under 150mg/dl (3,88mmol/l) in the Framingham data set have died in heart disease, it's suspected that all of these had some kind of vascular anomalities. So, people with "normal" cholesterol levels do not get atherosclerosis if these normal levels are maintained. People having TC cholesterol under 150mg/dl do not get atherosclerosis (Castelli et al. Circulation, May, 1977)

    The interview of the principal scholar of Framingham study, William Castelli

    Heart Disease Risk: Cholesterol and Lipids in 2011
    What Do We Really Know?

    And Ted, no one beside those who have written a low-carb book takes seriously the "additional benefits of statins"-theory. This is just poor propaganda propagated by the loosers in the business.

    1) Low-density lipoprotein cholesterol reduction and prevention of cardiovascular disease

    "There was little question after the first major statin trials that the reduction in CVD was related to lipid lowering and was totally consistent and supportive of the lipid hypothesis. However, stimulated by funding from the pharmaceutical industry, in which competition was fierce for market share and was driven mainly by the efficacy of lowering LDL-C levels, manufacturers of less-effective agents for lowering LDL-C levels helped propagate “beyond LDL-C” theories; these theories were that statins reduced CVD events by means other than lipid reduction, often termed pleotropic effects, usually shown in in vitro laboratory studies or small, poorly standardized surrogate marker trials. This belief culminated in an RCT by a pharmaceutical company that was designed to show that more LDL-C reduction with a competitor's statin achieved no greater benefit.13 However, the results of that study clearly and convincingly showed otherwise, with additional reduction in CVD events with the drug that lowered LDL-C levels more. Even with this evidence, and perhaps with an even more powerful statin about to be approved, the investigators suggested that the reduced events were due to pleotropic effects of the more efficacious statin. However, the trial was soon followed up with results from another head-to-head RCT, with the same drug at different LDL-C lowering doses,14 which eliminated the pleotropic potential and rein-forced that lower is better"

    2) A meta-analysis of 19 studies (81,859 participants in all) that lowered LDL and measured clinical disease or death, including 5 diet trials, 3 bile sequestrant trials, 1 surgery trial and 10 statin trials concluded that: "The pleiotropic effects [effects apparently unrelated to lowering LDL] of statins do not seem to contribute an additional cardiovascular risk reduction benefit beyond that expected from the degree of LDL-C lowering observed in other trials that primarily lowered LDL-C."

    Robinson JG, Smith B, Maheshwari N, Schrott H. Pleiotropic effects of statins: benefit beyond cholesterol reduction? A meta-regression analysis. J Am Coll Cardiol. 2005 Nov 15;46(10):1855-62

  2. Jeff
    Here's the new paper I was referring to. It was not funded by statin merchants. Statins are so safe that you get low-dose statins over-the-counter in the UK. The authors of the paper concluded in an additional file that all people over 50 need to take statins. Clinicians are prescribing them too little and too late. And my mouth is almost where my money is, my mother takes statins too. I don't never need to.

    The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials

    "In individuals with 5-year risk of major vascular events lower than 10%, each 1 mmol/L reduction in LDL cholesterol produced an absolute reduction in major vascular events of about 11 per 1000 over 5 years. This benefit greatly exceeds any known hazards of statin therapy. Under present guidelines, such individuals would not typically be regarded as suitable for LDL-lowering statin therapy. The present report suggests, therefore, that these guidelines might need to be reconsidered".

    But incase you do not want to you use statins, you need to buy Esselstyns book and read carefully the food recipe section. Vegan, ultra-high carb, low-fat diet. In the supplementary section Esselstyn shows an angiogram picture of his fellow-surgeon in the same Cleveland clinic who managed to do a 100% regression of his atherosclerosis without the statins. He suffered a stroke and was diagnosed with atherosclerosis, despite he was in his 40's, lean and athletic, his TC cholesterol was just 156mg/dl, but it was too high. Thanks to the diet, in a matter of months his TC cholesterol went to 89mg/dl. Your cholesterol cannot be too low. It has never happened, not even among those who carry the genetic polymorphism which gives the LDL of just 15-20mg/dl, even these people do just great.

    The angiogram shot

  3. Jeff
    Sorry, the fellow-surgeon of the Cliveland clinic did not do 100% regression, but 100% reversal of his atherosclerosis. English is not my first language.
    Reply: #144
  4. Bill42
    Hi Jeff,

    You should question why Esseltyn's fellow surgeon had a stroke and had atherosclerosis in the first place, before attributing it to a "too high" TC level of 156mg/dl.

    Again, perhaps you'd be kind enough to explain why plaques form in the first place, and also why approximately half of all CVD occurs in people who do not have "high" TC levels?

  5. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    The comments are a place for discussion. They are not for pages after pages of copy-pasting the same stuff over and over. That's called spamming (unless it's your own blog).
  6. Bill42

    You quote in (1) "But you know they found these villages in China where you couldn't get a heart attack or you couldn't get diabetes and the women couldn't get breast cancer and you know their total cholesterol were 127." This does not show any causal link. Question: why only in isolated villages, and not throughout the 880m people involved in the study?

    You quote in (2) "although not clearly established at this time, to prevent atherosclerotic plaques, the serum LDL cholesterol must be <70 mg/dL" and also "exactly what HDL cholesterol level is required to prevent plaques is unclear at this time". This is then followed by "In summary, the connection between cholesterol elevation and atherosclerotic plaques is clear and well established." Please explain this jump from "not clearly established" to "clear and well established" without any apparent supporting evidence.

    You then quote "If one has elevated cholesterol, has an elevated blood pressure, smokes cigarettes, or has an elevated blood sugar, these additional factors serve to amplify the cholesterol damage but they by themselves do not produce atherosclerotic plaques!" So the cause of plaques is not elevated cholesterol, as cholesterol by itself it does not produce atherosclerotic plaques.

    (3) There is no causal link to show that the low TC prevented deaths from CVD. Also, the Cornall-Oxford China Project was a study way back in 1983 and was on the effects of diet on cancer, not CVD, and was designed accordingly. Can you provide a specific reference to where in the study it states that not a single death occurred from CVD in 250,000 people over 3 years? It does not say that there were no CVD events, only no deaths.

    In (4) you say "Thanks to Seven countries study we know that in the 1960's out 1000 Japanese 5 died due to fatal heart disease, during the 10 year follow-up. The mean serum lipids were 160mg/dl in Japan during that time." but again this does not show a causal link between cholesterol and heart disease.

    There are many peer-reviewed published works that have shown that there is no causal link between TC levels and CVD.

  7. Juhani M
    Jeff/Richard/Peter, since you keep your arguments very consistent from one nutrition blog to another, why you're not keeping the same consistency with your name?
  8. You couldn't pay me a million dollars to take a statin.
  9. PhilT
    R^2=0.339, so nothing to get excited about !
  10. Stansmom
    But we see on this graph that the French, who eat the most saturated fat, eat only a little more than 15% of their total energy in the form of saturated fats. So is it possible that saturated fat consumption has bad effects when it rises much above this rather modest level of about 15%? Likewise, it is possible that these bad effects are increased when overall fat consumption is also quite high? I would very much like to know if there is recent research that addresses these questions. (Yes, I am aware of the Inuit's diet, but perhaps they are a very special case. Also, I know of no longevity figures for the Inuit, especially figures that control for deaths by infection, accident, etc., so that you could isolate the effect of the diet per se.) Thanks for any help you might be able to give.
  11. Where does that figure come from? The data that I have seen (does not include info from the former USSR) show very strongly the opposite. The data is from Keys which is somewhat suspicious but it is in the textbooks (p. 416 in Willett's Nutritional Epidemiology) showing very strong correlation between CHD and saturated fat intake for different countries. As Willett points out the countries with high incidence are more industrialized and richer and had many other factors that might contribute.
  12. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    There's a link in the post to the statistics:

    The Keys graphs had cherry-picked countries, right? Also they looked at the statistics decades earlier.

    I find in conceivable that rich countries used to eat more sat fat AND sugar and have more heart disease back then (e.g. in the 50s).

    Processed carbs and sugar used to be more common in developed richer countries. But now it's everywhere.

  13. Dick
    "The Keys graphs had cherry-picked countries, right? Also they looked at the statistics decades earlier"

    Just because this just so-story is perpetuated in every low-carb book doesn't mean it is actually true. The original 22-country study had nothing to do with the seven-country study. These studies were complete different studies with completely different data-sets. The data for the 22-country study was of very poor quality (Mexico f.ex did not have even proper mortality statistics at that point). Keys knew this and had not yet made his mind about whether it is fat in general or just saturated fat that leads to increased heart-disease risk. Seven country study was a prospective cohort study which meant that data collectors were sent to each country to properly collect analyze the data (questionares and blood samples), this enabled very high quality data. After controlling for every possible co-founder weight, age, blood pressure etc the link of saturated to increased heart disease mortality remained. Even after this study Keys wasn't sure about the case and proceeded to conduct metabolic-ward type, extremely controlled feeding studies.

    The seven country study is by far the best ecologic study ever done. It comes from an era were people did not use statins nor tried to loose weight. It has much less potential co-founders as opposed to studies that come these days, usually just measuring up homogenously eating high risk cohorts from a single country.

    For those interested, the whole Keys story along with the fabrications made by low-carbers are nicely presented in this high-quality 4-serie youtube video:

    Primitive Nutrition 36: The Infamous Ancel Keys? Part I

  14. Dick
    "I find in conceivable that rich countries used to eat more sat fat AND sugar and have more heart disease back then (e.g. in the 50s).

    Processed carbs and sugar used to be more common in developed richer countries. But now it's everywhere".

    The sugar link to heart disease doesn't really fit to the epidemiologic school of thought (not to even mention biologic school of thought since, unlike with the SAFA-case, we have no known biological mechanism about the allaged sugar induced heart-disease risk, other besides the weight-gain mechanism maybe).

    Countries like Costa Rica, Cuba, Ecuador, etc have had traditionally very high (table) sugar consumption and yet low heart-disease rates.

    Again from the same primitivenutrion -youtube channel, very illustrating and high quality cover up:

    The Futility of Cholesterol Denialism, Part 3: A Process of Elimination

    and before some even throws the Masai-card:

    1) The Masai of East Africa: some unique biological characteristics

    2) "In contrast to white-americans who have a limited maximal absorption capacity of 300mg of cholesterol, the Masai could absord more than 650mg cholesterol. Compared with the 25% suppression of synthesis found in white Americans, the Masai could suppress 50% of their endogenous cholesterol synthesis".

    Cardiovascular disease in the tropics. IV. Coronary heart disease

    3) Lipid intakes of Maasai women and children

    "Cholesterol intakes were below 220 mg/day. Energy intakes were inadequate and were between 65% and 80% of the recommended daily intakes based on body weight, or 50% and 60% of the RDI based on age/physiological status. It is suggested that this is the reason why serum lipid levels are not high".

  15. Dr. Andreas Eenfeldt, MD Team Diet Doctor

    "The Keys graphs had cherry-picked countries, right? Also they looked at the statistics decades earlier"

    Just because this just so-story is perpetuated in every low-carb book doesn't mean it is actually true. The original 22-country study had nothing to do with the seven-country study.

    What 22-country study? The first one, much-discussed by low-carbers and sometimes mixed up with the 7CS, contained only six countries ("Atherosclerosis: A problem in newer health" from 1954).

    Anyway, moving on to the later seven-country study: this too concerns cherry-picked countries. So there's seven instead of six, so what? Whether the data is prospective or not does not have anything to do with the selection.

  16. Dick

    Keys referred to a six countries in his speech and the paper that followed it in 1953 (The 7CS came in 1970). Why would he have referred to the other countries? Yerushalmy and Hilleboe did actually analyze the whole 22-country data set in 1958 and found a whopping 0,756 correlation for animal protein to deaths from atherosclerosis (see Yerushalmy and Hillboe paper 1958, figure 15 with correlation co-efficients from various dietary components to death rates), correlation of animal fats to atherosclerosis was 0,684. Carbohydrate intake did not correlate with heart disease mortality at all.

    Yerushalmy and Hilleboe only wanted to highlight that Keys had, in his own interpretation of the data, overlooked animal protein. This is pointed out in numerous textbooks. I find it ironical that Ravnskov, Taubes, Richard Feinman & Volek all give attention to the paper by Y & H. It's sad though that they provide a glimpse of what actually happened in that paper. Feinman thinks protein provides metabolic advantages yet he refers to authors that saw (animal) protein as a main culprit for cardiovascular disease. Mark Sisson certainly did not understood this when he brought the "original evidence" at the table, he didn't even understood what paper he was referring to and confused them all. Atkins himself though the 7CS came in 1950's....phew. When do folk realize what a mess this low-carb crank crew is :)

    Anyways, Keys knew better than Y & H and did not include the other countries in the 1953 paper since the data was skewed due to the war and in many ways of poor quality. Sweden, Norway, the Netherlands, Denmark, etc had just recovered from the war era and the period of instability that had occurred prior and after the war years. So, when the data showed that these countries had quite high level of SAFA consumption and low levels of death from cardiovascular disease, Keys was aware that cardiovascular death had drastically plummeted in these countries during these difficult years as had the consumption of animal foods as well. The German troops had confiscated the dairy and meat farms in continental Europe. The new diet high in animal foods that was recently adopted in these countries probably did not yet reflect in the statistics. It takes years for diet changes to manifest themselves, especially in regards to cardiovascular mortality. And, as said, Mexico did not have even proper death certificate system during that time (1950s). Why would Keys had included Mexico in his 1953 paper? It would not have made any sense. Besided, even if he had included all the countries, the SAFA link to heart disease mortality would have remained, although overshadowed by animal protein.

    The data from the 22-countries was from FAO, the data for 7CS came from a different sources, as I explained. The 7CS was top-snotch stuff, it came not only from 7 countries from many contrasting regional cultures within these countries. Indications that Keys cherry-picked these countries and regions is just plain non-sense.

    References to Y & H paper including the own words and "comments" by Y & H (together with lot more).

    Primitive Nutrition 37: The Infamous Ancel Keys? Part II

    Cardiovascular Disease Epidemiology: A Journey From the Past Into the Future

  17. Dr. Andreas Eenfeldt, MD Team Diet Doctor

    The 7CS was top-snotch stuff, it came not only from 7 countries from many contrasting regional cultures within these countries. Indications that Keys cherry-picked these countries and regions is just plain non-sense.

    Except you just said yourself that he cherry-picked the countries, choosing not to include those (Sweden, Norway, the Netherlands, Denmark etc.) that didn't fit his theory.

    What you're describing is not a random selection. It's simply rationalization for cherry-picking.

    Besides, an ecological study should hardly be described as "top-notch" with all its many weaknesses. Maybe it once was, back in the seventies, but science has advanced a bit since then. The diet-heart hypothesis has since crashed and burned in proper RCTs.

  18. Dick

    1) Netherlands was included in the 7CS study. Again, the idea that Keys picked the data doesn't hold water. Sweden and Denmark would have fallen nicely on the regression line in the 1960's. The study included over 18 regions. What part did Keys cherry-picked, he even included Crete which showed quite high SAFA intake and low level of cardiovascular mortality. Infact the correlation would have become much more stronger had he replaced Crete with Sweden and Denmark. The countries selected were selected purely out of convenience; and the technical challenges of conducting surveys across cultures by national teams, often working under difficult field conditions.

    Overview: The Seven Countries Study in Brief´

    "The Seven Countries Study is the prototypical comparison study of populations, made across a wide range of diet, risk, and disease experience. It was the first to explore associations among diet, risk, and disease in contrasting populations (ecologic correlations). Central chemical analysis of foods consumed among randomly selected families in each area, plus diet-recall measures in all the men, allowed an effective test of the dietary hypothesis. The study was unique for its time, in standardization of measurements of diet, risk factors, and disease; training its survey teams; and central, blindfold coding and analysis of data.".

    2) RCT's are designed for pharmaceutical purposes. The idea that they would a new triump-card for the diet-disease sphere is simple flawed. Atherosclerosis takes decades to manifest itself, how would RCT properly express this? We have not a single clinical trial that would indicate that cigarette consumption or asbest exposure is dangerous, we only have the totality of evidence including metabolic-ward studies and understanding of biological mechanisms thanks to laboratory work, and the totality of the evidence points out that animal fats are harmfull to human.

    "The randomized clinical trial (RCT) is NOT the "trump card" of all diet and health studies, as claimed. RCTs were meant for the testing of pharmaceutical agents and would-be pharmaceuticals called neutriceuticals (i.e., nutrients), not for assessing lifestyle nutrition. RCTs have undoubtedly distorted the concept of nutrition and caused more confusion in diet and health research than almost any other scientific endeavor in recent years".

    Colin T Campbell

    Fatty acids in cardiovascular health and disease: Comprehensive update

    "Food is extraordinarily complex; thus, it is unlikely that randomized controlled trials assessing dietary interventions will be able to determine definitively the effects of altering intakes of various fatty acids on CVD risk. To make dietary recommendations, we will have to rely on epidemiologic evidence coupled with controlled clinical trials on surrogate markers, along with an evolving understanding of the pathophysiology of CVD"

    3) Besides what is with the idea that diet-heart theory does not receive support from the RCTs?

    Abstract 3610: Comparative Effects of 3 Popular Diets on Lipids, Endothelial Function and Biomarkers of Atherothrombosis in the Absence of Weight Loss

    "In the absence of weight loss, the high fat Atkins diet is associated with increased LDL-C, reduced endothelial vasoreactivity and increased expression of biomarkers of atherothrombosis. As such, these data suggest that isocaloric conversion to the Atkins diet may negatively impact cardiovascular health.."

    Meta-analysis of RCTs finds that increasing consumption of polyunsaturated fat as a replacement for saturated fat reduces the risk of coronary heart disease

    4) I think Daniel Steinberg makes a good case in his book, Cholesterol Wars, while people who accepted the lipid-theory were relying on not only in Darwinian biomedical foundation inlcuding animal models but on bile-acid studies, laboratory work on biological mechanisms, surgical studies, metabolic-ward studies, epidemiology, diet-trials, etc. People who opposed (these days lipid-theory is universally accepted by the people who have credentials to say something constructive about it) were simply relying on diet-trials from which the evidence for diet-heart was strong but but "airtight" as Steinberg puts it.

    5) But, let's not pretend it, opposing diet-heart theory (lipid-theory) is not opposing the theory per se. It implies that the one who opposes the theory succumbs her/himself to religious ideas of some extent. It would mean that the our Darwinian paradigm is flawed, a system that is foundation of our current understanding of biology. Elevated LDL implies higher heart disease risk in every animal and insect. It's pretty much a law compared to gravity. Cats, lions and dogs do not get elevated LDL due to saturated fat and dietary cholesterol, but immeadiately when we discovered a way to raise their LDL (manipulating thyroid function) an increased heart-disease risk came into realization. The fact that SAFA and dietary cholesterol elevates the serum lipids of man and apes is a metabolic fact that has been proved in hundreds of extremly tightly controlled metabolic-ward studies. Maybe this is the fact why there's no expert in lipid metabolism who opposes diet-heart theory anymore, they understand the implication of denying it, and do not want to a end up as being perceived as creationists for the rest of their lives.

    Jeremish Stamler on Siri-Tarino paper (all the authors of this paper paid by the Dairy Counsil embrace the lipid-theory and warn against the SAFA consumption).

    ”…To neglect this fact in a review about humans is to imply that the Darwinian foundation of biomedical research is invalid and/or that there is a body of substantial contrary evidence in humans. Neither is the case. Dietary cholesterol (as well as SFA) adversely influences human serum lipid concentrations, per cited equations”

  19. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    Colin Campbell, now there's the cherry-picker to rule them all! Not surprising that he dislikes higher-quality science that points in another direction than his cherry-picked observational data.

    Dismissing gold-standard RCTs and hailing weak ecological studies is just sad. Talk about being in love with your theories.

    Besides, if you do trust in ecological data like the 7CS, what about the graph above?

  20. Yoly

    In the face of contradictory evidence: Report of the Dietary Guidelines for Americans Committee


    Concerns that were raised with the first dietary recommendations 30 y ago have yet to be adequately addressed. The initial Dietary Goals for Americans (1977) proposed increases in carbohydrate intake and decreases in fat, saturated fat, cholesterol, and salt consumption that are carried further in the 2010 Dietary Guidelines Advisory Committee (DGAC) Report. Important aspects of these recommendations remain unproven, yet a dietary shift in this direction has already taken place even as overweight/obesity and diabetes have increased. Although appealing to an evidence-based methodology, the DGAC Report demonstrates several critical weaknesses, including use of an incomplete body of relevant science; inaccurately representing, interpreting, or summarizing the literature; and drawing conclusions and/or making recommendations that do not reflect the limitations or controversies in the science. An objective assessment of evidence in the DGAC Report does not suggest a conclusive proscription against low-carbohydrate diets. The DGAC Report does not provide sufficient evidence to conclude that increases in whole grain and fiber and decreases in dietary saturated fat, salt, and animal protein will lead to positive health outcomes. Lack of supporting evidence limits the value of the proposed recommendations as guidance for consumers or as the basis for public health policy. It is time to reexamine how US dietary guidelines are created and ask whether the current process is still appropriate for our needs.


    Effects of saturated fat: Science is inaccurately represented or summarized

    The conclusion of the DGAC Report suggests that the replacement of SFA with monounsaturated fatty acids or PUFA creates unequivocally positive cardiovascular risk factor outcomes; this is not the case. Studies cited by the DGAC Report demonstrate increases in atherogenic lipoprotein levels or triacylglycerols, decreases in high-density lipoprotein cholesterol, and varied metabolic responses to lowered dietary SFA in subpopulations [45], [46], [47], [48]. These controversies and uncertainties with regard to SFA are not included in the DGAC Report.

    Citing a meta-analysis by Jakobsen et al. [49] as evidence of “a significant inverse association for PUFA (with 5% substitution for SFA) and coronary events” (p. D3-16), the DGAC Report misrepresents the actual findings of the meta-analysis. The NEL summary shows this association was weak overall and significant only for women younger than 60 y. The meta-analysis [49] also showed that for all men in the study and for women at least 60 y of age, there was no significant association between substitution of PUFA for SFA and risk of coronary events or coronary death. An accurate summary of this meta-analysis and the additional controversies and uncertainties in the science suggest that evidence associating dietary SFA with increased risk of CVD is inconclusive.

  21. Yoly
    Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease1,2,3,4,5

    Design: Twenty-one studies identified by searching MEDLINE and EMBASE databases and secondary referencing qualified for inclusion in this study. A random-effects model was used to derive composite relative risk estimates for CHD, stroke, and CVD.

    Conclusions: A meta-analysis of prospective epidemiologic studies showed that there is no significant evidence for concluding that dietary saturated fat is associated with an increased risk of CHD or CVD. More data are needed to elucidate whether CVD risks are likely to be influenced by the specific nutrients used to replace saturated fat.

  22. Dick

    thanks for a good debate.

    In my previous post I screwed up:

    "People who opposed (these days lipid-theory is universally accepted by the people who have credentials to say something constructive about it) were simply relying on diet-trials from which the evidence for diet-heart was strong but but "airtight" as Steinberg puts it"

    I meant to write that Steinberg addressed that the evidence from diet-trials is strong in favour of lipid-theory, but not "airtight". He pinpointed the totality of evidence, statin-trials, lab work, metabolic-ward feeding experiments, bile-acid research, animal models, epidemiology, etc. Campbell made essentially the same point.

    The point made by Campbell has made in tens of similar kind of studies from China. The idea that Sir Richard Peto, the inventor of meta-analysis concept and professor in Oxford, manipulated the data is just ridiculous. He was the man behind the China Study raw data fact checking.

  23. Yoly,0,5464425.story

    "Fat is not the problem," says Dr. Walter Willett, chairman of the department of nutrition at the Harvard School of Public Health. "If Americans could eliminate sugary beverages, potatoes, white bread, pasta, white rice and sugary snacks, we would wipe out almost all the problems we have with weight and diabetes and other metabolic diseases."

    It's a confusing message. For years we've been fed the line that eating fat would make us fat and lead to chronic illnesses. "Dietary fat used to be public enemy No. 1," says Dr. Edward Saltzman, associate professor of nutrition and medicine at Tufts University. "Now a growing and convincing body of science is pointing the finger at carbs, especially those containing refined flour and sugar."


    And the nation's levels of obesity, Type 2 diabetes and heart disease have risen. "The country's big low-fat message backfired," says Dr. Frank Hu, professor of nutrition and epidemiology at the Harvard School of Public Health. "The overemphasis on reducing fat caused the consumption of carbohydrates and sugar in our diets to soar. That shift may be linked to the biggest health problems in America today."

  24. I offered an earlier link to Edward Saltzman in my "fad" paper,,

    suggesting who it is that fed us that line. The conclusion was:

    "Very low carbohydrate diets may be effective for some, but their long-term benefit remains questionable. In addition, while no overall ill effects on blood lipids or other cardiovascular risk factors have thus far been observed, further research is needed to determine if certain individuals may be more susceptible to the nergative effects of high-fat and other diets that are traditionally associated with increased cardiovascular risk.

  25. Maybe the Soft Drink Post is out of date but I was TAKING A LITTLE POLLand got only one answer although I got three on a Facebook page.

    Probably useful here since we're talking saturated fat.

    So here's the question:

    For general health, should you change the type of carbohydrate or replace the carbohydrate with fat (any natural fat, no trans-fat)?

    It's a thought experiment (not a real world situation with subtleties). You only get three choices: For general health (no change in calories):

    A. Change type of carbohydrate
    B. Replace carbohydrate with fat
    C. It doesn't matter

    A., B. or C. ?

  26. Yoly
    B. Replace carbohydrate with fat
  27. Traditionally The french diet (exept south east) is rich in saturated fat (butter, meat, goose and duck grease, eggs, raw milk cheese) but also rich in white bread and starchy foods (bread comes with the 3 meals). And even if they to make us cut down on butter and fatty meats (foie gras !!) we try to resist to all fad diets. Yet, more and more are replacing butter with olive oil and are pushed to reduce their overall fat intake. I have no idea whether it is changing our health.
  28. Daniel
    A low carb vegetarian diet would probably refer to a restricted number of blood-sugar spiking carbohydrates, which excludes primarily fiber but also some other forms of carbohydrates with low glycemic index.

    The post says in itself that the graph doesn't prove anything; it should say something though, that predicted results are not observed

  29. Just throwing this out there... Why do we think we can conclude anything about saturated fat from this correlation? At best, it looks like deaths from heart disease correlate to geographic location... or how about stress? Happiness?

    Don't get me wrong - I'm a fan of saturated fat... but I'm not convinced that this graph is scientifically valid. It sure is emotionally valid.


    Reply: #130
  30. However at some polnt we have to accept there isn't a statistically significant association between SFA and CVD risk.Dietary saturated fat and fibre and risk of cardiovascular disease and all-cause mortality

    and also relevant to this discussion it's worth noting
    Ruminant-produced trans-fatty acids raise plasma HDL particle concentrations
    There is a difference in the effect of INDUSTRIALLY PRODUCED trans fats from partially hydrogenated vegetable oils and those NATURALLY produced by cows and found in butter.

  31. I would agree with you Ted...

    I'd just hate to fall into the trap of using a neat graph to conclude our own biases.

  32. I agree. The graph is not a support for the idea. Otherwise we would have to accept the very strong correlation country-by-country of the effect of saturated fat on colon cancer. For a correlation to imply causality, it must have 1) good correlation which this does have, as the colon cancer association does, 2) Must be strong underlying mechanism from basic chemistry which neither of them really have, 3) NOT be strongly correlated with a third parameter: the saturated fat-colon cancer correlates with economic factor (I think GNP) and I suspect this does to.

    The real force of the graph is that it doesn't go the other way. So, whereas a correlation doesn't necessarily imply causality, a LACK of correlation (or going the other way) is strong evidence of LACK of causality. More generally, if saturated fat were as bad as they say, no correlation should fail, not one, no big clinical trial should fail, not one, no small test should fail, not one. In fact, you could fill a book or at least half a book with all the ones that do failr (Good Calories, Bad Calories, The Cholesterol Myths, The Cholesterol Con, Big Fat Lies, etc.)

    So, yes, the graph doesn't prove the case. Still, you gotta' love it.

  33. And I do... love it that is.

    As a geologist by training... I tend to take data like this and wonder how it could apply to the underlying bedrock mineralogy which would/could make it into their bodies. This would strongly correlate to their local diets.

  34. I agree that saturated fat as a causative agent in CVD is dead.

    Of course, as in a court of law, you can only be found not guilty. You cannot be found innocent. If you are found not guilty of child molesting, people still won't let you move into their neighborhood. Saturated fat should never have been indicted in the first place. In fact, whatever effects it does have will be hard to find because of all the bad science.

    That is also my critique of fructophobia. Guilty until proved innocent. In addition to all the other bad effects, we will never actually find out the real biochemistry of fructose because of all the recommendation-driven bad science and the hysteria which I still give DietDoctor a little bit of a hard time on.

  35. Daniel
    Ironically, in France you are guilty until proven innocent. And according to their national statistics, Sat Fat has been found innocent.
  36. ...and I think my family has little heart disease because they left byelaruss.
  37. Vive le code de Napoléon (at least in this case).
  38. 1 comment removed
  39. Isn't this blog moderated?
  40. Zepp
    Yes it is, but not so often.. I just klicked the triangel to make somone to react of this!
  41. If this nutty, earthy tea does not seem to your satisfaction then why not try other
    calming, stress relieving teas like peppermint, chamomile or chilli chai tea.

    Tea has long been revered as a health elixir in the East
    and now Western science is supporting those claims.

    The antioxidants in honeybush are also known for being helpful in
    treating women's menopausal symptoms.

  42. Kirk
    The eastern European countries shown in your graph do not have the same level of sophisticated medical procedures and access to cholesterol reducing drugs as the western European countries you list. Any simpleton knows that in the U.S., which has the highest rate of animal product consumption in the world has also seen the death rate from heart disease drop since the 1960's. This does NOT mean that heart disease has declined. It just means we have gotten better at preventing people from dying from it. Take a look at the rates of heart attacks, angioplasty and open heart surgery in the U.S. and in these western European nations over the last 40 years and you will see how they have skyrocketed.
  43. I'm from Ireland and heart disease has always been one of the biggest killers. In Scotland its even worse. However, I would not place the blame solely at door of saturated fat.

    There are a lot of other determining factors such as high levels of alcohol and smoking which is still a big problem in Northern Europe and the Eastern Bloc countries.

    In saying that my grandparents who lived on butter and milk lived well into their 90's.

  44. Rita
    Hi Jeff, So sorry to get in here so belatedly. I and my family all have familial hypercholesterolaemia (my mother is now 82 and hers was only "discovered" in her sixties. The statins give ALL of us rhabdomyolises. So we're stuck on fibrates (still with myopathy). Without drugs our numbers soar into the 13's. I very "bravely" went onto keto dieting on 1 February 2012 and astonished my doctors. After a few weeks for the first time ever in my life my numbers plunged to almost what you doc's would call normal. Even ON high dosages of whatever we cannot get our total chol. numbers below 8 ever. And mine was just over 6. I KNOW in my knower, that I can never go off this diet of saturated fats, olive oil, coconut oil (definitely hard for me to ingest this) lots of eggs, etc. and enough vegetables. I have proof of this since I've been monitored over 30 years. My sons are now both on Keto diets and though it's not a cure all, it's done something stellarly grand to my cholesterol numbers. The blood spoke by itself. I don't like to be contrary, but I feel for us it was a matter of life. Cheers
  45. Magarietha
    Though I do believe in a keto diet (for me and my my family specifically) I think those eastern European countries have some of the highest smoking rates???
  46. MargaretRC
    @Dick, "Meta-analysis of RCTs finds that increasing consumption of polyunsaturated fat as a replacement for saturated fat reduces the risk of coronary heart disease." Not according to this study in the same journal from which you quoted. Your article is dated 2010. This article is dated 2013 and addresses some data that was missing when the article you quote was written and was later found. "In this cohort, substituting dietary linoleic acid in place of saturated fats increased the rates of death from all causes, coronary heart disease, and cardiovascular disease. An updated meta-analysis of linoleic acid intervention trials showed no evidence of cardiovascular benefit. These findings could have important implications for worldwide dietary advice to substitute omega 6 linoleic acid, or polyunsaturated fats in general, for saturated fats."
  47. MargaretRC
    Time to put the Lipid Hypothesis to bed. Seriously. As both Dr. Eenfeldt and Dr. Feinman and others put it, it's not what this graph proves. It's what it DISproves. It disproves that there is any causal relationship between dietary saturated fat and CVD. Time to investigate other causes.

    @Stansmom, I believe studies have been done that show that in the context of a low carbohydrate diet, considerably higher sat fat intake than 15% still results in improvement in those lipid profiles that are considered indicative of CVD risk.

  48. Ted Hutchinson
    Self‐reported whole‐fat dairy and butter intakes have strongest associations with phospholipid fatty acid 15:0.

    So when Phospholipid 15 is measured we can see what happens to cardiovascular risk factors in those with the highest consumption of whole‐fat dairy foods such as butter, cheese, and whole‐fat milk.

    If you believe dairy saturated fat consumption is linked to heart disease you would expect to find those with the highest proven consumption of dairy fat (as demonstrated by higher phospholipid 15 levels), will have the highest heart disease risk factors and highest cvd incidence?

    The full text of this paper is freely available here
    Biomarkers of Dairy Fatty Acids and Risk of Cardiovascular Disease in the Multi‐Ethnic Study of Atherosclerosis

    The results show an inverse association between plasma phospholipid 15:0, a biomarker of dairy fat, and CVD risk factors and CVD incidence.

    That means those with the HIGHEST measured dairy fat intake from whole fat dairy (butter/cheese full fat milk) had the LOWEST risk factors for heart disease.and lowest CVD incidence.

  49. Firebird
    This is what I know. My cholesterol is around 300, but my ratio of HDL/LDL falls in the excellent range. My triglycerides and insulin levels fall into the excellent range. My doctor, going into a total panic, sent me for a stress test. An MRI showed ZERO BLOCKAGES in my arteries. I eat low carb, high fat and protein (maybe a little higher than normal as I am a weight lifter). The true tale sign of heart disease is blood pressure, and mine falls in the 110/80 range. So, no heart disease for me.

    Still, for all of that, my doctor wants me on a statin, eating less protein (because now he is worried about my kidneys, which according to an ultrasound, function perfectly) and even fewer fats.

  50. Richard Feinman
    I would challenge your doctor to go for an MRI himself.
1 2 3 4

Leave a reply

Reply to comment #0 by

Older posts