top of page
 FOLLOW Nutrition Wonk: 
  • Facebook B&W
  • Twitter B&W
 A Nutrition wonk: 


A person who loves nutrition science as much as they love food.


No tags yet.

Ancel Keys and The Seven Countries Study: New Paper!

Check out this WHITE PAPER from the True Health Initiative on the Seven Countries Study. What is it, what are the controversies, and what is true?

Ancel Keys, Time Magazine, 1961

In 2016, I wrote a piece as a response to the viral article "The Sugar Conspiracy," reexamining some of the claims of the author, particularly with regards to Ancel Keys and the Seven Countries Study. Aspects of the story struck me as slightly fishy, so I figured I'd dig into the original research and see what I found. As it turns out, most of the claims were not quite right and some were flatly untrue. And yet, it seemed like there was a growing bandwagon of people who were interested in writing off the Seven Countries Study as more of a scandal than a study.

It seemed odd for the trendy nutrition internet to direct its ire at a single study, especially one so old, so seminal, and so, so long. Especially because the truth was sort of mundane: 1. It was a groundbreaking study because it was the FIRST of its kind, warts and all. 2. It was extremely meticulously planned. 3. It featured serious interpretive limitations due to its narrow scope and size BUT 4. It showed that lifestyle factors did seem to be strongly correlated with heart disease, which was not a given at the time. In fact, one of the major strengths of the study was that authors developed a method to categorize heart disease deaths and incidence that was standardized across all the cohorts studied.

And yet the mythology surrounding this study has grown to a fever pitch!

A few months ago, I was contacted by Dr. David Katz who asked if I would be interested in helping him, Dr. Joel Kahn, and Dr. Walter Willett create a paper that focused on the history of Seven Countries and explained the methodology, results, limitations, and controversies surrounding it.

With the co-authors, as well as original authors of the Seven Countries Study (Ancel Keys didn't do this by himself!), we worked on a deep-dive of Seven Countries that attempted to include:

1. The design of The Seven Countries Study (SCS) and how it was carried out

2. A list of the primary arguments "against" the science of the SCS frequently seen on blogs and in books. Each of these was then compared against the literature of the time and addressed. The topics addressed are similar (but much more thorough) to those in my previous Sugar Conspiracy post and include:

- Were countries picked based on desired outcomes?

- Was France purposefully excluded?

- Was the dietary data in Greece collected incorrectly?

- Was sugar ignored in the SCS?

3. Strengths and limitations of the original SCS

4. Some conclusions about how the SCS has held up over time

5. An epilogue by Dr. David Katz

Why are these topics important now?

When we talk about science from a historical perspective, the methodology and findings can almost seem like foregone conclusions. The way elegant and seminal experiments are described make it sound like they came off perfectly the first time, like a soufflé on a cooking show. On the other hand, studies with methodological flaws can be seen with "hindsight 20-20 vision."

Modern portrayals of SCS somehow manage to do both. The scale, timing, coordination, and planning that went into the study is downplayed and matter of fact. But shortcomings are addressed with scoffs, "Why didn't they KNOW that?"

Of course, as the paper delves into, much of the modern critique of the study methodology is mistaken: France was not excluded, sugar was not ignored, and dietary data in Greece was, yes, purposefully collected during Lent. This isn't to say there weren't shortcomings of the dietary aspect of SCS, but most of these, in my opinion, are related to interpretability rather than methodological flaws.

So why do bloggers (including myself) focus on the ideas of a sugar "conspiracy"? Frankly, it's more interesting. As it turns out, saying, "well the applicability of the findings to a wider population is limited by the non-random selection of cohorts and small n when each cohort is used as a single data point," is way less compelling than saying IT WAS ALL A CONSPIRACY!

But objective evaluation of historical and scientific evidence is inherently important.

SCS is largely published in three HUGE monographs (each around or over 300 pages) and one 15-year follow up paper published in 1986. Later books and monographs reanalyzed data from individual cohorts or among individuals. There is even one entertaining memoir about the experience. You can imagine that it's really difficult to distill the essence of this study into a 60 page paper.

What does the paper find?

First, I'd suggest that if you want to have an opinion about a half-a-century old, thousands-of-pages nutrition study, it's important to consider putting in some serious reading. Within HOURS of the paper posting, I had notifications that it had been dismissed as "fake news" that was paid for (it was uncompensated) AND that the paper "should be good" simply because of the identity of one of the authors. NEITHER of these are strong reasons for agreeing with or dismissing a paper. If you have a strong opinion about Seven Countries, you owe it to yourself and the original researchers to read an entire paper on it.

Then, feel free to send me any questions you have or, if you want to write or blog about it yourself, email the corresponding author, Dr. David Katz, for details! I am 100% happy to chat about this.

That being said: We found that the most salacious accusations about the paper were entirely untrue. However, we do discuss in great detail both the papers strengths and limitations. This probably won't surprise too many scientists, but the original Seven Countries authors who reviewed the paper were extremely wary of over-interpretation.

Anyway, please check it out and let me know what you think!

Join our mailing list

& Get Weekly Updates


See Nutrition Wonk disclosures

bottom of page