Ancel Keys Study Data Disproving Cholesterol Hypothesis Hidden for Decades
I am not a doctor, scientist, or expert in anything. This content should not be construed as advice or recommendation, but is intended for entertainment and informational purposes only.
Ancel Keys is the father of the diet-heart hypothesis that vilified saturated fat and cholesterol, and it turns out that he hid study data from the public that would have disproven his hypothesis decades ago. This is actually oldish news that I hadn’t heard about until recently, and based on the continuing guidelines regarding saturated fat intake and cholesterol medications I’m guessing many others are not aware of this revelatory study either.
But first, who is Ancel Keys?
As mentioned above, Ancel Keys is the guy we can thank for getting us to eat more seed oils and egg whites, and less red meat, egg yolks, butter, cheese, and milk. His Seven Countries Study detected a correlation between serum cholesterol levels and cardiovascular and coronary heart disease (CVD/CHD). The countries involved in the study were the United States, Finland, the Netherlands, Italy, Yugoslavia, Greece, and Japan.
From the early stages of this study there were disagreements between researchers as to the actual cause of heart disease. Keys had obviously hitched his ride to the saturated fat/cholesterol wagon yet a stronger correlation between refined carbohydrate and CVD/CHD was ignored by Keys.
For whatever reason, health authorities in the U.S. and other western nations bought into Keys’ hypothesis, and began changing dietary guidelines to lower saturated fat and cholesterol intake and increase polyunsaturated fats (seed oils) and more carbohydrate.
Let’s consider some of the potential flaws of this study. First, correlation is not causation. Second, there is a big difference between an observational study and an interventional study. Observational studies, while inexpensive and easier to conduct, are highly prone to confounders and bias that can easily skew results, and any data gleaned from these types of studies must be taken with more than a few grains of salt.
Third, why choose only seven countries when there were numerous other countries from which to gather data. What was he afraid of discovering?
Anyhoo, the health authorities were sold on this idea, and Big Ag and Big Food loved it too because that meant that more of their heavily processed, industrial grain and seed oil products would be consumed.
The problem with this hypothesis is that it doesn’t stand up in the real world. So let’s get back to the study I referred to at the opening of the post.
The Minnesota Coronary Experiment was an interventional, randomized controlled trial (RCT) study conducted from 1968-1973. While not perfect by any stretch, RCTs are considered the gold standard because they remove many of the potential confounders and biases that affect observational studies. RCTs are great when they are well constructed because they do a better job of identifying any potential effect of the intervention being studied rather than just a possible correlation.
The reason why this study is so unique is that it was conducted on participants who were patients in six state mental hospitals and one nursing home in Minnesota. While consent was determined to not be necessary since they could opt out of participation once the project was explained to them, they were essentially captive participants so controlling for variables was much easier than trying to control for variables outside of an institutional setting.
The study involved over 9,000 people, with over 2,300 who were on the diets for more than a year. The intervention arm received significantly higher amounts of linoleic acids from corn oil and corn oil based margarine. The control arm continued eating higher amounts of saturated fat, but was also given a higher amount of linoleic acid than they had received at baseline.
Both the intervention arm and control arm saw a decrease in serum cholesterol levels, but the intervention arm had a much more significant reduction of 13.8% on average while the control arm only saw a 1.0% reduction on average. Yay for less cholesterol, right?
Maybe not.
It turns out that the reduction in serum cholesterol had no impact on all-cause mortality for those under the age of 65, but in those over 65 there was an inverse effect. The more serum cholesterol was reduced, the higher the incidence of all-cause mortality in this older age group.
The number, probability, and proportion of deaths also increased with greater reductions in serum cholesterol levels including those in the control group.
The authors of the study also reviewed autopsy data that had been recovered. 41% (31/76) of those in the intervention group who died and had an autopsy performed had clinical evidence of at least one myocardial infarct (MI). Only 22% (16/73) of the control group participants who died and were autopsied had a MI. The participants in the intervention arm also did not have less coronary or aortic atherosclerosis than the control arm.
The premise of Keys hypothesis can be summed up as follows:
Higher saturated fat intake increases serum cholesterol levels = more heart disease and higher incidence of death. Therefore lowering serum cholesterol levels will lead to lower incidence of heart disease and death. Polyunsaturated fats high in linoleic acid which are found in seed oils mistakenly called vegetable oils lower serum cholesterol which should improve cardiovascular health and mortality rates.
What this data shows us instead is that polyunsaturated fats do indeed seem to lower serum cholesterol levels, but do not lead to less cardiovascular disease or lower all-cause mortality. In fact, we see the inverse effect in populations over the age of 65. What might be even more stunning is that the control arm diet had fairly high amounts of industrially produced trans fats which are incredibly terrible for heart health.
How much greater would the disparity in results have been had Keys not increased consumption of linoleic acid in the control arm, and if trans fats had also been removed from the control diet?
This is not the only study that contradicts the popular narrative around the cholesterol/saturated fat causes heart disease myth. The Sydney Diet Heart Study conducted around the same time as the Minnesota Coronary Experiment was a smaller study, but was also a RCT that evaluated 458 men aged 30-59 years with a recent coronary event.
The intervention group replaced saturated fats with linoleic acid from safflower oil and safflower margarine. The intervention group had higher rates of all-cause mortality as well as higher rates of cardiovascular and coronary heart disease deaths.
So was Keys full of shit all along, and knew from day one that his hypothesis was bogus?
Keys is long dead so we can’t ask him, but based on accounts from some of his colleagues Keys truly believed that saturated fat and cholesterol were the enemy. He also was apparently a bit of a bully, and often got his own way. In a word, he was human and just as fallible as the rest of us.
Keys couldn’t override his ego and just follow the data that said, “You’re wrong” as a scientist should do if he or she is being true to the scientific method. He had made his name and career as the guy who decreed that saturated fat and cholesterol cause heart disease, and he was bound and determined to prove that out regardless of what the data showed.
Unfortunately, this is how “science” often goes. Much of what passes for science more closely resembles dogmas and religion based on beliefs instead of cold, hard, objective facts.
What exacerbates the problem in Keys’ case is that government bodies bought into his hypothesis and began making widespread recommendations. Big Ag and Big Food companies that would benefit financially from these recommendations couldn’t wait to spread the good word that their products would help lower cholesterol making them “heart healthy.”
Pharmaceutical companies joined the act by developing cholesterol lowering medications that don’t improve health outcomes, but which are some of the most highly prescribed drugs today and are insanely lucrative.
The fact that the data from these two studies finally saw light of day in 2016 and 2013 respectively show the power of narratives and the cult-like fervor that often accompanies them. This information should have been bombshell news spread across every media outlet causing those in the health community to take pause, yet nothing came of this because too much money and too many careers are at stake.
I’m sure that 99%+ of cardiologists are not familiar with these studies, or at least not familiar with the full results. That might be an interesting topic of conversation at your next appointment if you have a cardiologist.
You might print these studies off and share them with him or her, and ask if they are aware of the contradiction regarding lower cholesterol and worse health outcomes. I would love to be a fly on the wall if he or she actually took the time to review the data.
Cholesterol, much like animal protein, is a critical nutrient that the body needs more as we age, but the cholesterol is bad train will take decades to stop and only after an incalculable amount of damage has been done to the collective health of the people who follow these recommendations.
To be clear, I’m not making any recommendations whatsoever. But it would be wise to do your own homework, especially when it comes to your health, and be wary of any recommendations whether or not they are accompanied by scientific/expert backing.