How history shaped the American diet Part 2: Big change backed by little science

Part 1 reviewed the swift progress and major events in history that influenced the American diet. Part 2 exposes a big, fat, juicy story that will make you question what you believe about what you eat.


Once upon a time...

It came to light in the 1940s that the leading cause of death was coronary heart disease. Scientists, thus, sought to identify the cause and ways of prevention. Numerous research studies taking place over decades, and some still on-going, began collecting data. Among those was Ancel Key’s seven country study. This was an epidemiological study comparing how many people were dying of heart disease with how much fat they were eating. At the time, 22 countries had this data available and were included in the study.

Keys found that the countries with higher fat consumption had more deaths from heart disease (Fig.1). Japan, for instance, consumed less that 10% of total calories from fat and had a much lower death rate from heart disease than the US who consumed 40% of calories from fat. The conclusion drawn from this study was that the more fat in the diet, the more likely to die of heart disease.

Taking a deeper look, there were a number of things wrong with this study. First and foremost, the methods of data collection lacked rigor and would never fly in research today. There were two major issues with data collection. One, percent fat from total calories was determined by food availability per capita in each country. Basically, the amount of food available was divided by total number of people in each country. This is a poor estimate of what people were truly eating. Two, death by heart disease was determined by death certificate, however autopsies were not performed at this time and cause of death is many times confounded by multiple comorbidities. In the 1940s, cause of death was many times a guess. With so many flaws, this study design would not have a leg to stand on in research today. In fact, the original data has been destroyed, so forget reanalysis.

The second major error in this study has to do with interpretation of the findings. In statistics, there is a little saying, “correlation does not equal causation.” Correlation analysis is simply a mathematical way to describe the relationship between two things. For example, height is correlated to weight because as we grow taller, we also weigh more. Another example is that there is a strong correlation between homicide rates and ice cream sales.  Does this mean that buying ice cream causes violent crime? No, that sounds foolish. However, there is plenty of evidence showing that warm temperatures cause an increase in crime, maybe because we are more likely to lose our temper in hot weather. In addition, ice cream is more appealing in warm weather, therefore, sales are higher in the summer vs winter months. This means that temperature and temper confound, or that there is merely more to the story in the relationship between homicide and ice cream sales.

Back to Key’s conclusions. Yes, the data says that people in the US ate more fat than people in Japan and more people in the US were dying of heart disease. However, is fat the cause of heart disease or just a confounder? Was there more to the story?

Nevertheless, in the 1950s, Ancel Keys went to the World Health Organization and American Health Association (AHA) promoting a low-fat diet claiming that a “fatty diet raised serum cholesterol, atherosclerosis, and myocardial infarction.” In 1961, Keys was featured on the cover of Time Magazine making the recommendation that the American diet should consist of 70% of calories from carbohydrates and 15% from fat – a drastic change from the 40% fat calories Americans were consuming. The AHA got on board and released a report that “the best scientific evidence of the time” suggests that Americans would reduce their risk of heart disease by reducing the fat in their diets.

True, though flawed, the best evidence at the time suggested that dietary fat consumption is related to heart disease. In the least, this observation called for further research.

When the fat hit the fire

Big money was put into more research on how and why fat consumption appeared to be linked to heart disease. One, the Western Electric Study by Paul Oglesby, was published in 1963. This study was conducted to identify factors associated with the risk of heart disease. Over 4,000 men age 40-55 years were selected to be in the study. Over a 4.5-year period, mortality rate was compared between those consuming the fattiest and those consuming the leanest diets. The findings were clear, 14 deaths due to heart disease in the high fat group and 16 in the low-fat group. No significant difference. These findings were not only ignored but later blatantly misreported. The AHA and National Heart, Lung, and Blood Institute (NHLBI) published “The Cholesterol Facts” which included the Western Electric Study as one of seven studies showing a correlation between saturated fat and heart disease. Exactly what it did not do.

What could have put an end to the low-fat ideology came in 1973 with The Minnesota Coronary Study. This was a highly controlled diet intervention study conducted on 9000 men and women in 6 mental institutions across Minnesota. While unethical to study this vulnerable population today, investigators were able to control what the participants ate – something that had not been done before on such a large scale. The researchers investigated the effect of saturated fat (thought to be “bad” fat) on heart disease over 4.5 years. The participants in the study were randomized to either a 9% saturated fat diet or control diet (18% saturated fat) for 4.5 years. Total fat was the same for both groups.

Contradicting Ancel Key’s findings, the investigators found no difference in heart disease or overall mortality between groups. In fact, there were 269 deaths due to heart disease in the treatment group while there were 206 deaths in the control group.

Why did this study not put an end to the low-fat diet? The lead investigator was not pleased with the results and the study sat unpublished until 1989!

Fat Chance

Evidence piled up that the amount of fat has no relationship to heart disease. However, the type of fat was found to be important.

There are different types of fat depending on chemical structure. These include saturated, unsaturated, and trans fats. There is some evidence, though controversial, that saturated fats (found in beef, pork, skin on poultry, whole-milk, cheese, butter, eggs, and palm and coconut oil) are related to heart disease. A number of studies have shown that consumption of omega-3 fatty acids actually reduces the risk of heart disease. Today, the AHA recommends no more than 5-6% of fat from saturated fat while unsaturated fats are healthy. The message that was heard, though, is that fat is bad, which was not just misleading but dangerous.

What about cholesterol? Cholesterol is essential for life and present in all tissues of the body. Eating cholesterol (such as egg yolks and shellfish) does not actually contribute much to blood cholesterol levels. In other words, you cannot get high cholesterol from eating foods high in cholesterol. Only about 25% of the cholesterol comes from diet because cholesterol in foods can’t be absorbed by our bodies. The other 75% is produced in the body by the liver. Cholesterol levels are tightly regulated in the body so that when cholesterol level in the diet goes up, the body makes less. It is wrong to say that high cholesterol causes heart disease.

In the 1970s, the NIH funded studies to investigate which specific fractions of cholesterol contributed to heart disease. It was found that total cholesterol does not predict heart disease. However, it was also found that different proteins carry cholesterol throughout the body dropping it off where it is used for cell metabolism. The amounts of these proteins does matter, though. These proteins include high density (HDL), low density (LDL), and very low density (VLDL). Cholesterol carried by LDL that is not used, is left to circulate in the bloodstream where it can then accumulate in the walls of the arteries, leading to plaque formations and atherosclerosis. LDL is therefore thought to be “bad” cholesterol. HDL acts more as a scavenger, collecting cholesterol and taking it back to the liver for reprocessing or excretion. HDL is the “good” cholesterol.

When it comes to heart disease, high LDL is only a “marginal risk factor,” however, low HDL was found to be a 4-fold better predictor of heart disease compared to LDL, and it was the only predictor of heart disease in men and women over age 50. Therefore, pay attention to your HDL and things that raise HDL may reduce the risk of heart disease.

Unfortunately, the findings for these NIH studies too were ignored for 20 years. Another tragedy in the bigger story.  

The Tipping Point

The tipping point came in 1977, though, when the government took hold of the low-fat hypothesis. That year, the US Senate’s Select Committee on Nutrition and Human Needs (aka the McGovern Report) published the “Dietary Goals in the United States.” For the first time in history, a branch of the government urged Americans to eat more fruits, vegetables, whole grains, poultry, and fish and to eat fewer high-fat foods, substitute nonfat for whole milk. In other words, eat more fruits and vegetables instead of fat.

low fat foods.jpg

Tell people not to eat fat and you think they are going to eat more fruits and vegetables? What happened was the invention of a whole new food culture. The “Snackwells phenomenon.” ‘Low-fat,’ ‘fat-free,’ and ‘low-calorie’ foods which largely replaced fat with sugar were mass produced. However, many of these foods still have the same number of total calories as their full-fat counterparts.

Further fueling the low-fat movement, in the 1980s an article based on the Framingham Heart Study was published highlighting obesity as an independent risk factor of heart disease. Consequentially, it was believed that a low-fat diet might reduce the risk of heart disease and promote weight loss by reducing caloric intake (because of all macronutrients, fat has the most calories per gram). In other words, if it wasn’t fat, it wouldn’t make you fat and you won’t die of heart disease.


Then, as if proven, the federal government incorporated the low-fat ideology into federal policy. In 1984, the National Institute of Health consensus stated “there is no doubt that a low-fat diet will afford significant protection against coronary heart disease to every American over the age of 2.” In 1988, the Surgeon General’s report emphasized the dangers of dietary fat. In 2000, the federal “Report on Nutrition and Health” labeled fat as the unhealthiest part of the American diet – not saturated or trans-fat, just fat. Shockingly, in 1992, the American Heart Association (AHA) began labeling low-fat products with their seal of approval, such as Kellog’s Frosted Flakes, Fruity Rice Krispies, and Low-Fat Pop-Tarts.

With this kind of support of course consumers (you and me) believed, and still believe, that a low-fat diet is healthy. Thanks to my mother trying to stay trim and raise healthy kids, I grew up a low-fat kid sustained on cereal, low-fat Wheat Thins, fat-free yogurt, and Reduced Fat JIF Peanut Butter!

The advice being fed to Americans was not only false, but actually dangerous as people were eliminating healthy fats that are important in reducing risk of heart disease. While fat consumption was reduced, carbohydrate consumption increased and so did the prevalence of obesity, heart disease, diabetes, and other metabolic diseases.

More on the consequences of the low-fat dogma in Part 3. Stay tuned!