Wednesday, March 16, 2011

Historical Context, Part 2 - The Diet-Heart Hypothesis

In the 1950's in America, the diet-heart hypothesis was born, theorizing that the fat in our diets caused heart disease.  Proponents of this hypothesis had two very compelling reasons to believe in it.  First, was the increase in heart disease rates, which more than doubled since the 1920's.  The other was the "changing American diet" story; the idea that at the turn of the century, Americans were consuming significantly more grains and less meat and were healthier for it.  These two ideas together formed the basis for the diet-heart hypothesis.  The fat-laden diet of the 1950's must have been the reason for the skyrocketing heart disease rates, right?  I would say no.  Both of these ideas, the foundation for the diet-heart hypothesis, are easily explained by other phenomena.  Much of the information in this and the rest of this historical series will come from Gary Taubes' fantastic book Good Calories, Bad Calories.  I'd recommend it to everyone, but it's so information dense that it reads sort of like a textbook.  Luckily, you have me to summarize for you.

First on the list is the belief that heart disease was rare before the 1920's and grew into America's #1 killer by the 1950's.  Census data showed in 1910 that only 250 Americans out of every thousand would die from heart disease, but in 1950, that number had risen to 560, more than double.  So the real question is, then, did heart disease rates really increase, or was there simply an increase in the awareness of the disease, or perhaps better technology to diagnose it?  As it turns out, the use of the newly invented electrocardiogram in 1918 made heart disease much more easily diagnosable.  People were living longer by the 1950's too due to antibiotics that could control infectious disease; life expectancy had increased from 48 years in 1900 to 67 years in 1950.  As we know today, very few heart attacks are seen in 48-year-olds, and obviously the longer one lives the more likely one is to develop a chronic disease like heart disease or cancer, which, incidentally, also increased in this time span.  An increase in heart disease diagnoses was also due to newly discovered variations of heart disease, like the new cause-of-death category added in 1949 for arteriosclerotic heart disease.  From 1948 to 1949 alone, total heart disease rates increased by 20% for white males and 35% for white females.  A similar pattern was then seen when a category for ischemic heart disease was added in 1965.  Based on all of this information, it appears that this "great epidemic" may not have roots in diet at all.

The other half of this argument, the "changing American diet" story can be challenged as well.  Ancel Keys, a University of Minnesota researcher who will be the subject of the next piece in this series, wrote in 1953, "The present high level of fat in the American diet did not always prevail, and this fact may not be unrelated to the indication that coronary disease is increasing in this country."  Keys, and other supporters of the diet-heart hypothesis, envisioned the turn of the century as an era free of chronic disease due to a high-carbohydrate, low-fat diet. The food disappearance data, though, which this assertion is based on, were not reliable.  The statistics date back to 1909, but the USDA only began compiling the data in the early 1920's.  The resulting numbers for per-capita consumption are acknowledged to be, at best, rough estimates.  Here's an example of the kind of data I'm talking about:  this one shows estimated flour and grain consumption.

The data before 1942 were particularly sketchy, especially when it came to any foods that were grown in a garden or eaten straight off the farm, such as animals slaughtered for local consumption.  In fact, David Call, a former dean of the Cornell University College of Agriculture and Life Sciences, when asked about the early food disappearance data, stated that "Until World War II, the data are lousy, and you can prove anything you want to prove."  Historians of American dietary habits can provide some insight into the diet before the turn of the century, siting several sources indicating that in the 1800's, Americans were a nation of meat-eaters, typically eating meat 3 or 4 times per day.  Also of historical note is the fact that at the turn of the century, pasta was considered by the general public to be "a typical and peculiarly Italian food", according to The Grocer's Encyclopedia of 1911, and rice was still an exotic item imported from the Far East, so Americans may not have been eating much of these foods.  But, if it is true that grain consumption was high and meat consumption was low by 1909, it was probably a brief departure from our meat-based diets of the past.  At the time, the cattle industry was reportedly having trouble producing enough meat to feed the growing United States population, so there would have been less meat available.  Americans would have had to cut back.

What is most interesting to me, is that if these diet-heart hypothesis supporters wanted to use the food disappearance data as evidence to support their claim, why then did they choose to ignore the data on fruit and vegetable consumption?  In these years between 1909 and the 1950's when heart disease rates doubled, vegetable consumption increased dramatically.  Americans nearly doubled their consumption of leafy green and yellow vegetables, tomatoes, and citrus fruit.  Why was this not taken into account?  This is one of many examples of proponents of the diet-heart hypothesis choosing to ignore the evidence that doesn't support their ideals.  This is exactly the type of thing a good scientist tries to avoid, but this type of bias occurred frequently in the development of the low-fat theory.

So in the end, the diet-heart hypothesis was created in order to provide an explanation for the "heart disease epidemic".  But if you look deep enough, there was already an explanation for it; one that made a whole lot more sense.  And if you look at American dietary history before 1909, it looks like we ate a lot of meat, making the low-meat, high-grain diet of the early 19th century a deviation from the norm.  Taken one step further using the lens of evolutionary biology, which they did not have the luxury of in the 1950's, prehistoric humans ate more meat than Americans ever did, they lived into their 70's, and heart disease was virtually nonexistent.  Nonetheless, many researchers, Ancel Keys in particular, became enamored with the idea that a low-fat diet was the key to keeping heart disease at bay, and he was determined to prove he was right.  Stay tuned for part 3!


  1. Nice post man! Keep it up.

    I like the concept of increased prevalence of heart disease vs. increased awareness of heart disease. Very interesting

  2. thanks man, you'll love the next one on Ancel Keys haha