Repair Your Metabolism For a Better Life.

What Every Doctor Should Know About Ancel Key’s Experiments

Share

If you’ve been reading the news, you probably know that eating delicious foods like butter and eggs is no longer thought to increase risk of heart attack or stroke. This post discusses new information has come to light suggesting Ancel Keys suppressed evidence that polyunsaturated fats are more harmful than trans fat!

For those of you who aren’t fully convinced that butter and eggs are healthy, I’ve devoted the first half of this article to highlighting why, when your doctor recommends that you swap out saturated fats in foods like butter and eggs for polyunsaturated fat in products like Smart Balance and packaged breakfast cereals, it’s largely thanks to Ancel Keys and his misleading, even dishonest, public statements.

Back in the 1950s and 1960s, Ancel Keys designed a series of highly influential experiments that changed the course of American dietary history. Before Keys, Americans enjoyed traditional foods like butter, eggs, and bacon without worrying about their health. After Keys made the cover of Time magazine on Jan 13, 1961, the American public was introduced to the idea that saturated fats were clogging their arteries, and that idea ultimately led to a sea change in the foods we eat. Real foods would increasingly replaced by processed, and the era of obesity and chronic disease would begin.

 

Sources of fat in the 20th century

Sources of solid fat in the 20th century, from Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th Century, Taha Et Al. Shortenings are made from hydrogenating vegetable oils, as are margarines.

Keys’ First Tactic was to Blind Doctors With Science

In the Time magazine article Keys is paraphrased as stating “Americans eat too much fat. With meat, milk, butter and ice cream, the calorie-heavy U.S. diet is 40% fat, and most of that is saturated fat—the insidious kind…that increases blood cholesterol, damages arteries, and leads to coronary disease.”

Keys based his claim that saturated fat increases blood cholesterol on the experiments he’d done. But the experiments were not performed using real foods rich in saturated fat. His experiments used saturated fats that had been artificially created from vegetable oils using a process called hydrogenation. Hydrogenated vegetable oils contain not just saturated fat but also trans fat and a whole host of other unnatural molecules. Trans fat does indeed raise blood cholesterol, as well as lead to heart attacks and strokes, and has been since shown to be so unhealthy cities like Manhattan  and San Francisco have outlawed its use in restaurants.

Keys fooled doctors using the term “hydrogenated fats.” He rightly guessed nobody would bother to look up either his research or the meaning of the word. 60 years later, most doctors are still confused.

 

Keys wasn’t trying particularly hard to hide this deception. One publication he authored was titled “Hydrogenated fats in the diet and lipids in the serum of man,” which should have been a red flag that he was not using real food in his experiments. But it wasn’t. Although doctors do learn some biochemistry, it’s not enough, and what little we do learn is not put in the context required to become relevant. Bottom line is I didn’t learn what hydrogenated fats were in medical school, and the doctors of Keys’ day would have had no idea either. So when Keys said saturated fat from butter and eggs caused atherosclerosis, he was intentionally misleading everyone by implying that he was speaking of the results of his experiments, when he was not.

This is how Keys managed to pull the wool over the eyes of generations of doctors–because he understood chemistry better than they did.

Recently, thanks to authors of blockbuster books like Good Calories, Bad Calories, The Big Fat Surprise, and The Primal Blueprint, the reading public has come a long way in our understanding of the health harms of trans fat, and the lack of harm from natural saturated fat. This information has trickled down to influence the thinking of some doctors as well, and a few have lifted their restriction on butter and eggs.

But Keys did more than fool physicians into believing eating butter and eggs would kill us.

 

Keys also convinced doctors that eating polyunsaturated fats would help people live longer.

In spite of progress in the realm of saturated fat, doctors still have a long way to go in understanding the health consequences of consuming polyunsaturated fats from industrially produced vegetable oils. Nutrition science as a whole has made very little progress on this front in large part because Ancel Keys appears to have done everything in his power to keep us from understanding just how unhealthy polyunsaturated fats and vegetable oils can be.

 

Medical investigators discovered data Keys did not publish. This data would have changed the course of American dietary history had Keys' shared it.

Medical investigators discovered data Keys did not publish. This data would have changed the course of American dietary history had Keys’ shared it.

 

Medical detective work by Christopher E. Ramsden, medical investigator at the NIH, who published an article called “Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73)” implicates Keys in a scientific cover up, perhaps the biggest cover up of important health data ever committed in American history.

The Minnesota Coronary Experiment was a massive undertaking involving thousands of adult residents of six state mental hospitals and one nursing home. The study subjects were divided into two experimental groups and fed two different experimental diets for between 3 1/2 and 4 1/2 years, depending on the hospital. One experimental diet was high in saturated fat, the other high in polyunsaturated fat. The high saturated-fat diet contained lots of margarine and shortening, both loaded with trans fat, and not much if any butter or natural animal fats (some of the details are lost to history). The diet high polyunsaturated-fat diet contained lean meats and skim milks and cheeses to which corn or other vegetable oils were added, along with some margarines, and no butter or natural animal fats.

The Minnesota Coronary Experiment was composed of two parts, each designed to test the validity of a different hypotheses:

1) That a diet high in polyunsaturated fat would lower cholesterol.

2) That a diet high in polyunsaturated fat would reduce atherosclerosis and deaths from heart attack and stroke.

The first hypothesis turned out to be true. Sort of.

 

Part One of Keys’ Study:

Polyunsaturated Fats do Lower Cholesterol…

After one year, the study subjects were retested and those in the polyunsaturated fat diet group had cholesterol levels 30 points lower, on average, than those in the saturated fat group. And that’s the reason doctors still to this day recommend increasing your intake. But remember, this reduced blood cholesterol from polyunsaturated fat is not compared to actual foods containing saturated fat. It’s compared to artificially made saturated fats from hydrogenated vegetable oils, and that means they would also contain massive amounts of trans fat.

…But Keys Only Compared Polyunsaturated Fats to Trans Fats

Trans fats have been shown to raise cholesterol. And foods containing artificially made saturated fat, like margarine and shortenings, contain lots of trans fat. These foods have been clearly shown to elevate cholesterol when added to a diet that was previously free of these foods.

But Keys studies never included a diet composed of actual foods rich in saturated fat.

When compared to actual food, the difference is slight and inconsistent.

Years later, others compared butter to both margarine and polyunsaturated fat, and found that polyunsaturated fat did indeed reduce cholesterol compared to butter–but not by much.

Taken together, Keys’ research plus the research showing slight differences between butter and polyunsaturated fat was enough to convince doctors that polyunsaturated fats do indeed produce reductions in serum cholesterol. That’s why products like Smart Balance can claim to reduce your cholesterol, and it’s why thousands of other products claim to be heart healthy.

But are they?

 

Part Two of Keys’ Study:

The second part of the experiment, the results that Keys suppressed and that Dr. Ramsden published, showed that the study group that got more polyunsaturated fats were not only not protected from heart attacks and strokes by the diet, but appeared to be at slightly increased risk for both.

The graphs below show the results of statistical analysis of various subgroups. The blue lines (labeled Diet) represent the diet high in polyunsaturated fats, and the red line (labeled control) represents the diet high in saturated and trans fat from margarines and shortenings. The higher the line, the more deaths. What we see is that the blue line is higher than the red line in all the groups no matter how you divide up the study participants (but especially in women over the age of 65), and that means the diet high in polyunsaturated fat caused more deaths than the diet high in saturated fats.

This is huge.

 

 Risk of death from any cause by diet assignment in full MCE cohort and prespecified subgroups (Kaplan Meier life table graphs of cumulative mortality). Graphical depiction of cumulative mortality in full MCE cohort (n=9423) and prespecified subgroups in 1981 Broste thesis7 showed no indication of benefit and suggested possibility of unfavorable effects of serum cholesterol lowering intervention among participants aged ?65. Patient level data needed to repeat this analysis were not recovered

Risk of death from any cause by diet assignment in full MCE cohort and prespecified subgroups (Kaplan Meier life table graphs of cumulative mortality). Graphical depiction of cumulative mortality in full MCE cohort (n=9423) and prespecified subgroups in 1981 Broste thesis (Broste S. Lifetable Analysis of the Minnesota Coronary Survey.University of Minnesota, 1981.) showed no indication of benefit and suggested possibility of unfavorable effects of serum cholesterol lowering intervention among participants aged ?65.

Polyunsaturated Fats INCREASE the Risk of Heart Attack and Stroke

We know that Keys’ so-called saturated fat diet was also loaded with trans fats, and we know that trans fats cause atherosclerosis. The Nurses Health Study showed that for each 2% increase in trans fat in the diet, the risk of heart disease is doubled. And if this trans-fat loaded diet performed BETTER than a diet high in polyunsaturated fats, that suggests polyunsaturated fats from processed vegetable oils, when consumed in the amounts consumed in the study, are more unhealthy than trans fat.

The medical investigator and study’s lead author was not able to quantify exactly the amount of trans fat in the saturated (here labeled “control”) diet, but it was likely to be quite a lot higher than the amount of trans fat in the polyunsaturated (here labeled “diet”) diet.

The amount of polyunsaturated fats in this experimental diet was not provided. Instead, the authors provide the amount of a single type of polyunsaturated fat that is also the most common of all polyunsaturated fats in vegetable oils, called linoleic acid. The study diets contained between 8 and 12% of calories from linoleic acid. Today’s average American consumes about 10 % of calories from linoleic acid, which puts us right in the ballpark of the average study participant. In other words, the average American is on a diet very much like Keys’ high polyunsaturated fat diet that does lower cholesterol but actually increases a persons risk of heart attack and strokes compared to a diet high in trans fat.

In other words, the American public is trapped in Keys’ medical experiment.

And we know what the results will be: more heart attacks and strokes. American doctors who still recommend polyunsaturated fats are unknowingly putting their patients at risk of death or disability. This is why every doctor should learn the truth about Ancel Keys’ experiments.

Unfortunately, when faced with potential reason to change our practice behavior, most doctors behave like herd animals in that we instinctively follow the recommendation of authorities in order to avoid slaughter by predatory lawyers.

And our authorities show no sign of changing their direction, in spite of this additional body of evidence that Keys’ did not have best intentions at heart.

Harvard’s Nutrition Department Chair, perhaps the ultimate authority figure on the topic of diet and health, still celebrates Keys’ legacy, and continues to promote polyunsaturates. In an interview  response to the publication of Dr. Ramsden’s article, the department chair and co-author with Keys on numerous articles, Dr. Walter Willet, expressed his disregard for the potential value of these newly released findings: “This is an interesting historical footnote that has no relevance to current dietary recommendations.”

Hopefully, Dr. Willet will retire and be replaced with someone less invested in the status quo.

There’s one more twist to this story. A twist that gets to the core of what’s wrong with the idea that cholesterol causes heart attacks. And if you have been wondering, how does a diet that reduces cholesterol increase heart attacks and strokes, you are already thinking in the right direction.

The reason they lower cholesterol has not been systematically examined. What little evidence we do have suggests they reduce cholesterol by causing a fatty streak, the first stage of atherosclerosis, and I review that evidence in this video. It has to do with disruptions in the function of lipoprotein, and I go into more detail in Chapter 7 of Deep Nutrition (Chapter 8 in the original version), entitled Good Fats and Bad.

Related Content on DrCate.com :

Canola Oil: The blob that ate butter, olive oil, coconut oil and peanut oil threatens American cuisine

Practical Lipid Management for Low Carb Practitioners

Related content from the web:

A 2014 article defending Keys reputation that was, at the time, being publicly called to question because of how he covered up data in his famous 7-countries study. This was before the 2016 article came out implicating Keys in an entirely different cover up.

The 7-countries study controversy, covered by Authority Nutrition with a cute video.

A brief movie about Keys. At minute 22 we gain some insights into how Keys public humiliation at a scientific meeting in 1955 may have lead to his obsession with proving fat intake was the problem. Rather than being driven to find the truth, he was driven by a desire to redeem his idea, “I’ll show those guys.”

Harvard’s rebuttal of Dr. Ramsden’s study by Dr. Willet, who defends the status-quo and blames the increased deaths in the polyunsaturated arm of the study on the lack of supplemental omega-3. This suggestion seems absurd, since the polyunsaturated corn oil would have contained more omega-3 than the hydrogenated oil diet that outperformed it. He also states the study was not long enough. It was, in reality, one of the longest if not the longest study to date.

Share

About Author

Dr. Cate

With over two decades of clinical experience and expertise in genetic and biochemical research, Dr. Cate can help you to reverse metabolic disease and reshape your body.

  • Melissa

    Hi Dr. Cate,
    I am reading your book Deep Nutrition and really fascinated on how how you healed your knee issues with nutrition. I’m only 34 and have chrondromalcia patella and have been trying to do an anti inflammatory diet and recently added bone broth to my diet. I was wondering if your dietary recommendations would still be the same for someone who doesn’t need to lose weight? I would actually like to build some more lean muscle, but I’m actually not a big carb fan, however I am a protein eater. I was just wondering if your guidelines would be the same, I’m not super active anymore with the knee issues but still workout praying healing will come. Thank you for your time.

    • For building muscle it’s helpful to get 2-3 meals per day with 20-30gm+ high quality (real food, animal) protein. The lysine content seems to stimulate muscle protein synthesis.

  • Sam

    Dr. Cate,

    Are you familiar with Dr. Greger? He has a completely opposite opinion in regards to nutrition.

    nutritionfacts.org

    • brad

      Dr. Greger, nutritionfacts.org, is the best (if not only) source of objective, scientifically-based, information that is non-profit and totally free of commercial interests. His brief research-based nutrition videos are a wealth of information (quite entertaining) and available at no cost.

      • Brad that’s not entirely true. Dr Gregor is not totally free of commercial interests. His income, his employer, is the humane society and although they do great work for animals, when it comes to human health, they do have a powerful pro-vegan agenda. Check it out.

      • Sam

        I’m sorting thought the sources.. It seems Dr. Greger is totally against animal based products, meat, butter and such. and I found your site and Mark Sisson’s is contrary to his views, pro-anumal products. Dr. Greger is way pro- bean diet and Mark is against it (not sure your position, thought)

        It seems like two different and opposite camps.

  • Karen McClintock

    Hi Cate,

    Not sure if you will get to read this but I would love your help. I recently listened to you interview with Dave Asprey. Loved all of your knowledge on veg oils. I am currently compiling my notes for my own podcast on veg oils, and was wondering if you had any links to scientific studies on two aspects you mentioned. One being that veg oils damage epithial cells, and one that they gobble up anti-oxidants. I’d love to cite some studies in my notes.
    Many thanks

    Karen

    • HI Karen,
      Dave and I discussed that vegetable oils interfere with something called endothelial function, which is how your body regulates blood flow. So we weren’t talking about epithelial damage. I explain why endothelial function affects your ability to concentrate, among other things, and discuss the antioxidant issue in Chapters 7 and 8 of Deep Nutrition and include citations to support all claims.

    • HI Karen, Dave and I were referring to “endothelial function” and the study I mentioned is here http://www.sciencedirect.com/science/article/pii/S0735109798006810 Many more in Ch 7 of Deep Nutrition!

  • Brett Graham

    I’m sorry, but Keys’ didn’t cover up anything in the sever countries study. It was a multi-country, multi-cultural epidemiological longitudinal study spanning many years and involving many international institutions. What you’re talking about is his paper from 1953, where he plotted 6 counties fat consumption against heart disease. He was accused of leaving out 16 countries data in that paper, but he gave good reason; the data was poor and unreliable.

    • Oops I’ve mixed up the different tactics Keys’ used to prove his point in the 6 versus the 7 countries studies. Thanks for pointing that out. In the 6-countries study he excluded entire countries when the data didn’t match what he wanted to prove. In the 7-countries study, he used entirely more complicated statistical cheats, delineated by both Gary Taubes and Nina Planck in their books. I’m a fan of trying to be fair, but Key’s made it clear that he had an agenda and was financially backed by the sugar industry. Plus, he didn’t even practice what he preached as far as not eating red meat because somewhere along the way he wrote or was quoted as saying that he and his only ate brisket (or some specific cut) for breakfast 3 times a week….I think it was in his book on the Mediterranean diet that I read that….Not sure anymore as it was over a decade ago.

  • Jay Wortman MD

    Good article. The recently published revision of the Sydney Diet Heart Study, also by Ramsden, reported similar findings which were also ignored in the initial article. These are major revelations and should be sending a shock wave through the nutritional science establishment. It is fascinating how they are so wedded to their dogma that they cannot conceive that it is wrong even when the very foundations are clearly shown to be the reverse. Another great folly of medicine like Semmelweis, bloodletting, etc. How many more will be harmed before we right the ship?

  • Dana Mulvany

    Dr. Cate, could you develop graphs showing the increase in heart attacks and strokes over time in relationship to the change in consumption of fats? The relationship wouldn’t be immediate but there should be a delayed one.

    A graph like that should then be persuasive to other doctors.

    Thanks for developing this article!

    • Interesting idea. It would be heavily confounded by smoking rates, which also paralleled a lot of this stuff. See my graph posted in response to Kurt Lass, the day prior. What do you think?

  • Chris Gorczynski, MD

    Very well presented. Over the past couple of years I significantly increased my dietary fat intake, especially saturated fat, while dramatically restricting refined carbohydrate intake. The result? Effortless 25 pound weight loss, 3 inch waistline decrease, reduced cravings, and improved mood/energy level. I write about this at https://www.yourorthomd.com/orthopedic-dietnutrition/why-would-an-orthopedic-surgeon-care-about-diet

  • Kurt Lass

    While Keys was certainly responsible for the original development of the K Ration, he is unlikely to be responsible for the cigarettes. Tobacco and cigarettes were included in US military field rations long before the K Ration was developed…

  • I want to hear from you! Do you agree with the arguments here? Any ideas about how we get more doctors, dietitians and other who are in the ‘system’ (by which I mean they accept health insurance) on board?

    • Interested Party

      I have been following the literature in this field for over 4 years, ever since reading Gary Taubes’ epic volume Good Calories Bad Calories. It has been and still is great experience.
      I totally agree with what you presented, and I admire your leadership in this regard, because it is clear that many in the health profession do not want to publicly dissent with the conventional wisdom. As we all know, there are strong political and Industry forces at work to discredit dissenters and make their life difficult. Just look at Dr. Noakes, Dr. Fettke, and also several other doctors or dieticians. Just the fact that the American Dietetics Association clearly states that any dissent from what the association guidelines state by a member is punishable by exclusion from the association. That is a serious deterrent, and should not be supported by law. So we can see that the establishment is fighting back, as big money is at stake, as well as scientific reputations. The only way I see the situation turning around is through medical leaders such as you who stand up for what is right and push ahead in spite of all the politically supported road blocks. The internet make is all possible, and I am very grateful for that. Please keep your good work.

    • zoe harcombe

      Hi Cate,

      While I welcome any blog pointing out that real foods, like butter and eggs, are healthy and omega-6 vegetable oils are not, as you’ve said “I want to hear from you! Do you agree with the arguments here?” there are some points that could do with clarification…

      1) Keys was sometimes erratic in his fat terminology (describing cakes and ice cream as “saturated fats” as late as 1970 in The Seven Countries Study) but he knew what he was doing in terms of experiments.

      There is indeed a paper called “Hydrogenated fats in the diet and lipids in the serum of man”, but it is about hydrogenated fats! The paper describes experiments involving 23-27 men where safflower oil was compared with hydrogenated safflower oil and corn oil was compared with hydrogenated corn oil. (For interest, consumption of the hydrogenated vegetable oils resulted in higher total cholesterol than the non-hydrogenated versions). Examination of this paper, or others of Keys at the time, show his lipid knowledge (dietary and serum) to be impressive.

      Another paper, from four years earlier (1957) “The effects of different food fats on serum cholesterol concentration in man”, with the same author team, administered butter, olive oil, corn oil, coconut oil, sunflower seed oil, sardine oil and cottonseed oil in separate tests to see what impact these had on serum cholesterol levels. For each fat, Keys, Anderson and Grande documented the saturated, monounsaturated and polyunsaturated content – something not enough people are aware of today.

      The main finding, which Keys presented against saturated fat, was in the summary volume of the Seven Countries Study. Here he claimed: “The average serum cholesterol values of the cohorts tended to be directly related to the average proportion of calories provided by saturated fats in the diet” and “The CHD incidence rates of the cohorts are just as closely related to the dietary saturated fatty acids as to the serum cholesterol level.” Not terribly assertive on either point.
      In this same study, Keys reported “Serum cholesterol averages and CHD incidence rates were not found to be related to the percentage of calories provided by polyunsaturated fatty acids in the diet” (this quotation continued: “and were only slightly related to total fat calories.”) This was when Keys changed his view that total fat was a factor in heart disease. This is not widely enough known.

      I’m not aware of any claim made by Keys that eating PUFAs would help people live longer.

      2) The Ramsden re-evaluation of the Minnesota Coronary Survey (MCS) was a brilliant piece of work, but I’m not sure it was necessary. The MCS started in 1968, but the main paper was published in 1989 (“Test of effect of lipid lowering by diet on cardiovascular risk. The Minnesota Coronary Survey”). Why did it take so long? You rightly praise Gary Taubes for his work in this area. In Good Calories Bad Calories, GT reports “When I asked Frantz in late 2003 why the study went unpublished for sixteen years, he said, ‘We were just disappointed in the way it came out.'”

      The 1989 MCS publication concluded: “For the entire study population, no differences between the treatment and control groups were observed for cardiovascular events, cardiovascular deaths, or total mortality.”

      This study simply didn’t support the lipid hypothesis. Plus, I’m not sure it’s fair to attribute this study to Keys. There were seven authors on the 1989 paper and Keys wasn’t one of them. Frantz may have been in Keys’ department and I’m aware that Ramsden refers to Keys as “co-principal investigator” (based on the research grant application?), but I’m not convinced that this can be pinned on Keys.

      3) Here’s how vegetable oils (plant sterols to be specific) can lower cholesterol by the way – while being associated with increased risk for CHD (http://thescipub.com/abstract/10.3844/ojbsci.2014.167.169)

      4) It is common, but wrong, for people to mix up the 1953 Keys Mount Sinai presentation – where data for 6 countries were presented while data for 22 countries were available – with the 1970 Seven Countries Study, which was always about 7 countries (16 cohorts). Tom Naughton is right in his video (he doesn’t mention the source of the graph); Authority Nutrition is wrong and so is your link across.

      I think that Keys does have a lot to answer for and I’ve criticised him myself a number of times. However, we’re taking on conventional ‘wisdom’ when we write blogs like this, so we need to be accurate as well as challenging.

      Best wishes – Zoe

      • Thank you for your thoughtful writing. Your statements are valuable. These concepts are complex and easy to misread so thanks for pointing out what I could clarify.

        1) My point was not that Keys lied about his use of hydrogenated fats, but rather that used our ignorance against us. When it came to speaking about the results of his studies to reporters he never explains what hydrogenated fats are, and typically brought up butter and animal fats as if to imply that’s what he studied. As I hoped the cartoon of the mouse and the cat would clarify, doctors didn’t know what the heck hydrogenated fats were, but they did know what butter was. So it was easy to confuse us.

        2) The Ramsden study was invaluable because he published never before published results of the autopsies. In addition, it helps to again highlight, as I did here, that the “control” diet was loaded with trans fat and the PUFA diet performed worse than that, which I think it a key detail we needed to know. And no doubt Keys didn’t want his name on a disappointing study, eh?

        4), I am not mixing up Keys’ studies. My point, and quite possibly Naughton’s is well, is that his 7-countries study was done on those countries that he knew ahead of time (because he’d done the 22/6 countries study by that point, along with other surveys) would ultimately support his data.

Dr. Cate’s Books

Special Book Offer from Dr. Cate

If you signed up before Jan 31, to access your free ebook Click Here

Got Pain? Free Summit, happening NOW!

** Book of the Month **

 

Health & Nutrition Articles

Latest Health & Nutrition Videos