FAT CONTROVERSY: HOW BIASED SCIENCE LED TO THE DEMONIZATION OF FAT AND FATTENING OF AMERICA

In this article I am going to present the history and (pseudo)science behind the demonization of fat. What I am going to show is how a researcher’s bias can play into the results of a study (or experiment). What followed from the conclusions of that study is the fattening of America (and the rest of the world) and the current state of obesity and heart disease around the world.

Early Work

It all began in 1912 by a Russian military doctor by the name of Nikolai N. Anichkov, who was the pioneer in establishing the role of cholesterol in the formation of plaque in the arteries. His theory stated that cholesterol alone was responsible for the physical changes in the arterial walls that led to the subsequent development of plaque.

His early work on Arteriosclerosis involved putting rabbits on a diet of bacon and eggs and studying the changes in their arterial walls from a period of 2 hrs to 1 yr.  This work titled “Inflammatory changes in myocardium: apropos of myocarditis,” formed the crux of his doctoral thesis which he successfully defended on 21^st April, 1912. This preliminary work in the experimental model of atherosclerosis was considered a pioneering “classic” and was used by a great number of subsequent researchers in the field.

The early work of Anichkov was published in the Russian literature but his first English medical literature publication was a chapter in Cowdry’s Arteriosclerosis in 1933[1]. Worldwide recognition of Anichkov’s work came only after a science [2] publication by Dr. John Gofman and his associates in 1950. Using Anichkov’s experimental techniques they confirmed the fact that introducing cholesterol in a rabbits diet rapidly led to atherosclerosis. They went a step further and were able to use an ultracentrifuge (which was unavailable to Anichkov during his time), and separate out the hypercholesteremic serum samples of their cholesterol-fed rabbits into 2 distinct groups. The fraction that floated to the surface of the serum sample was designated as low-density lipoprotein (LDL) cholesterol, and the fraction settled at the bottom was designated high-density lipoprotein (HDL) cholesterol. They further went on to show that low-density lipoprotein cholesterol is responsible for the rapid progression of atherosclerosis. This created a great interest in the worldwide medical community and a spate of research in cholesterol-induced atherosclerosis followed.

Follow up work of Ancel Keys and the Lipid Hypothesis:

Following Anichkov’s work, in 1952 a biochemist by the name of Ancel Keys came up with the lipid hypothesis, which, simply stated that the amount of fat in ones diet was directly correlated to the the incidence of coronary heart disease (CHD). Seems pretty common sense, right? So in order to back-up his “hypothesis”, he looked at epidemiological data from several populations across the globe to correlate dietary fat intake to the incidence of CHD. This led to what was called the ‘7 countries study’ which was a landmark study that changed the course of health for million to follow. The study involved looking at dietary habits of populations from 7 countries (Japan, Italy, Canada, Australia, USA, England and Wales) and the incidence of heart disease over period of a decade. The study didn’t account for any other lifestyle variables like, smoking, exercise habits, pollution levels, stress levels etc. Figure 1 shows the results of this study:

Figure 1. Correlation between CHD Vs. total fat intake (%) in Ancel Keys 7 countries study. Red line is a linear (least-squares) fit to the data.

From these results it’s pretty obvious that the more fat one consumed the more the likelihood for developing CHD. There is no disputing this data. But it didn’t end there. Ancel keys further went on to ‘show’ with his data that the amount of cholesterol in ones diet was directly correlated to the amount of dietary fat and again to the incidence of CHD. This earned him the title of ‘monsieur cholesterol’ and put him on the cover of Time magazine in 1961. He showed us the light and from there on began a crusade to put an end to all fats especially animal (saturated) fats as they were a rich source of cholesterol.

This landmark finding was hailed as the most important discoveries in the quest to end CHD. In fact it was so important that 211 prominent experts in the medical community’s were gathered in a conference in 1978[3] and asked if they thought cholesterol was the biomarker of CHD and 90% of them answered with a resounding ‘YES’! The National Institute of Health (NIH) held a conference in 1984 where they asked a panel of 14 experts if they thought blood cholesterol was the cause of CHD and weather reducing it would prevent heart disease, during which they voted with a unanimous “yes”. The panel further concluded [4]:

It has been established beyond a reasonable doubt that lowering definitely elevated blood cholesterol levels (specifically, blood levels of low-density lipoprotein (LDL) cholesterol) will reduce the risk of heart attacks caused by coronary heart disease...

 But little did they know that this was just the beginning of CHD! This was just the tip of the iceberg!

The Trans-Fat Revolution

So from the time Ancel Keys started his land mark study till now, deaths from CHD grew 600-fold, from 1 in a million to 600 in million! In the mean time the banishment of all fat and cholesterol from diet led to the trans-fat revolution and the low-fat fad. The gold rush had just begun!

The vegetable oil companies that were looking to expand in the human consumption market (before this most of the applications for vegetable oils was in the paints industry) found a gold mine in this study. They then started promoting hydrogenated fats which were vegetable oils with an extra hydrogen atom to make the oil solid at room temperature. This gave the consumer the ‘taste’ of butter (which was the fat of choice prior to this study) without having any of the ‘bad’ cholesterol. The other beneficiary of the results of this study was the pharmaceutical industry that found another gold nugget, selling drugs to lower lipid cholesterol levels. A $27 billion gold nugget to be exact (and it continues to grow!). This started a whole new low-fat food revolution that created factory foods where the calories from fat were replaced by calories from simple carbs and sugars. This brings forth the third beneficiary of the gold rush which was the American Heart Association (AHA) which started selling the AHA label endorsements to the low-fat industry, certifying low-fat products.

So while the pharmaceutical and vegetable oil industry and the AHA were reaping the benefits of this gold rush, the consumer was losing all along. The consumer didn’t reap any of the purported health benefits of the low-fat, cholesterol free products and drugs these entities were peddling. Furthermore, the US Department of Agriculture (USDA) was mandated with coming up with the recommended dietary guidelines for the general population and they came up with the food pyramid in the 1980’s which relied heavily on simple carbs from grains (at the bottom of the pyramid) and put fat in the prohibited category (on top of the pyramid). This was the turning point in the obesity epidemic, in that from there on the general population started getting fatter and sicker. Americans today are 30 lbs heavier than they were before the USDA guidelines.  The graph in figure 2 shows the trends in obesity and the adoption of carbs (over fat) over the years following the 7 countries study.

Figure 2: Trends in obesity and carb consumption from 1960 (courtesy Donald W. Miller, Jr., MD)

Analysis and Discussion

So what went wrong? Shouldn’t Ancel Keys’ study have unraveled the mystery of CHD? Or did he completely misinterpret the data and some other variable confounded the results of his study? Actually it was none of these. The whole thing started with a wrong experiment and later on was reinforced by incomplete data and poor science and then consensus was built around the faulty notion by the so called medical “experts” and from there on this so called “fact” propagated time and again until it reached epic proportions as we now see it. A quote from Joseph Goebbels aptly sums it up:

“If you tell a lie big enough and keep repeating it, people will eventually come to believe it...”

There were three main reasons why this whole thing was wrong to begin with. And it wouldn’t take a rocket scientist to figure out these inconsistencies in experimentation that led to this sorry state of affairs (at at least for the consumer’s health). Let’s delve into the simplistic early models and the methodology of the 7 countries study. But before we go any further let me bring forth a few common sense principles that all of us engineers and scientists have been taught to follow during the course of any experimentation. It will all fall in place once you hear me out.

1)    The results of an experiment are only as good as the model it is based upon.

For this let’s step back to the original Nikolai Anichkov’s experiment of putting rabbits on a diet of bacon and eggs. His model assumed that a rabbit’s body could be used to simulate the effect of cholesterol on a human body. So while his interpretation of the results of his tests was absolutely right, his model wasn’t a good predictor of the effect of cholesterol on human body. He plugged the right variables in the wrong equation! You see, rabbits are genetically adapted to a zero cholesterol diet. Their diet is devoid of any cholesterol so their body doesn’t adapt well to cholesterol in diet and will naturally lead to inflammation and plaque formation in the arteries. Humans on the other hand are omnivores that are genetically adapted (for over 140,000 years) to dietary cholesterol, via meat and dairy, and so the effect of cholesterol on a human body would be different from that of a Rabbit’s. Let me give you an example. I propose a test where I put a lion on a diet of carrots and broccoli and see the effect on its health. I am pretty certain that the lion will soon fall sick and perhaps not last 1 year on that diet. So, am I to conclude that carrots and broccoli are fatal to health? I am sure you get the point!

2)    Experimenter bias always has the potential to skew the results

Ancel keys proposed the lipid hypothesis, but a hypothesis needs to be proved through careful experimentation. But when an experimenter is so attached to his hypothesis that he/she starts ‘expecting’ the results in alignment with his hypothesis he only sees data that fits his hypothesis. This is exactly what happened with Ancel Keys. He was so attached to his hypothesis that he only cherry picked data from the 7 countries that fit his model. In fact during his time there was epidemiological data available from 25 countries but he only chose the data points that fit his hypothesis. Again to illustrate my point, let me propose my own lipid hypothesis (call it the Yogi Lipid HypothesisJ) which states that the more dietary fat one consumes the lower ones risk for CHD. I’ll go even better than Ancel Keys and pick data from 8 countries (bigger sample size) and show you the correlation. Figure 3 shows the results of my study:

Figure 3. Correlation between CHD Vs. total fat intake (%) in Yogesh Verma’ 8 countries study. Green line is a linear (least-squares) fit to the data.

From this data it’s quite obvious that the more fat one consumed the less the likelihood of developing CHD! Makes sense, right? My point being, that any scrupulous researcher should consider all data before making any conclusions. Figure 4 shows the data from all 25 countries that were available to Ancel Keys.

Figure 4. Correlation between CHD Vs. total fat intake (%) for all 25 countries. Blue line is a linear (least-squares) fit to the data (left). Table showing the 25 country data (right)

Here, if you look at the (blue) least squares fit it shows a flat line, i.e., no correlation! The variation in the data is of a random variable! That’s exactly right there is absolutely no correlation between the amount of dietary fat consumed and the incidence of CHD! Hard to believe, eh? That’s because we have been fed this utter non-sense about dietary fat and how it increases our chances of developing CHD for so long that we have forgotten that the lipid hypothesis was a hypothesis to begin with and it still is to this day. This brings us to the third important failing point.

3)    Epidemiological studies cannot be used to determine causal relationships.

Epidemiological studies (ES) are the observations of certain behaviors of populations in terms of diet and other health habits. It is the cornerstone of public health research and is used to determine the possible causes to a certain disease or physical condition. ES alone cannot be used to determine a cause and effect relationship between a variable and a particular disease (in our case fat/cholesterol and CHD). At best the role of an ES is to link possible variables to a disease and to form a hypothesis about the cause and effect, which then needs to be verified through carefully designed controlled studies. For example in the 7 countries study (if that was the only data available) the data should have only been used to formulate the lipid hypothesis. Ancel Keys should then have conducted carefully controlled studies to determine the cause and effect relationship between fat intake and CHD, not the other way around, i.e., assume a causal relationship and then look for ES data that fits the hypothesis. The data gathered in any ES can have a lot of confounding variables, like smoking habits, drinking habits, socio-economic stresses, etc. that need to be carefully eliminated in a well designed controlled study.

So the whole fat and cholesterol hypothesis was wrong on three counts. It was driven by a few powerful and influential people and organizations that had to prowess to influence public policy decisions. It was built on consensus by the medical community (AHA) “experts” and other NIH personnel because there was no science to back it up. No one could have said it better than Michael Crichton (the author of Jurassic Park):

I regard consensus science as an extremely pernicious development that ought to be stopped cold in its tracks. Historically, the claim of consensus has been the first refuge of scoundrels; it is a way to avoid debate by claiming that the matter is already settled. Whenever you hear the consensus of scientists agrees on something or other, reach for your wallet, because you're being had.

"Let's be clear: The work of science has nothing whatever to do with consensus. Consensus is the business of politics. Science, on the contrary, requires only one investigator who happens to be right, which means that he or she has results that are verifiable by reference to the real world. In science consensus is irrelevant. What is relevant is reproducible results. The greatest scientists in history are great precisely because they broke with the consensus.

"There is no such thing as consensus science. If it's consensus, it isn't science. If it's science, it isn't consensus. Period. . . .

"I would remind you to notice where the claim of consensus is invoked. Consensus is invoked only in situations where the science is not solid enough. Nobody says the consensus of scientists agrees that E=mc2. Nobody says the consensus is that the sun is 93 million miles away. It would never occur to anyone to speak that way. ."

I am not saying that the research community isn’t realizing the flaw with the lipid hypothesis. A recent (Jan, 2010) meta-analysis of 21 cohort studies showed no correlation between saturated fat intake and the incidence of CHD[5]. The recent revision of the food pyramid to a “my plate” is evidence of this change. The emphasis on whole grains and vegetables instead of just any carb is evidence of this change. But fat still remains the pariah of the macro-nutrient group and that needs to change. We have been lied to for the last half century and that needs to stop! The medical and research community needs to come clean because thats the least they owe us.

References:

1)    Anichkov NN. Experimental arteriosclerosis in animals. In: Cowdry EV, editor, Arteriosclerosis: A survey of the problem. New York: MacMillan Publishing; 1933. p. 271-322.

2)    Gofman JW, Lindgren F. The role of lipids and lipoproteins in atherosclerosis. Science 1950;111:166-71.

3)    Norum KR (1978). "Some present concepts concerning diet and prevention of coronary heart disease.". Nutr Metab 22 (1): 1–7. PMID 619310

4)    from NIH Consensus Development Conference, JAMA 1985, 253:2080

5)    Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease, Patty W Siri-Tarino, Qi Sun, Frank B Hu, and Ronald M Krauss, Am J Clin Nutr doi: 10.3945/ajcn.2009.27725.