Ancel Keys ~ The OTHER Minnesota Study

I preface this post with words of wisdom from a science journalist.

Long before Nina Teicholz burst onto the scene as the latest Atkins-mouthpiece (Chapter 10 of The Big Fat Surprise, I rest my case), from almost the very beginning of my exposure to the low carb internet community, I've been hearing how Ancel Keys was single handedly responsible for the low fat plague that has made us all fat and sick.  While I'm sure I'd heard the name Ancel Keys somewhere along the way during my formal studies, it never really rang a bell ... and I'd venture to guess that most outside this Incestral Health Community had not either.  This is the enduring legacy that appears to trace back to Gary Taubes' demonization of the man in Good Calories, Bad Calories.  It also appears to have origins in the Weston Price Foundation movement with the writings of Sally Fallon and Mary Enig, for example, The Oiling of America (2000).

This past May 1, I was sent a link to the sneak peek of a presentation by Dr. Cate Shanahan on a diabetes online summit.  Her description of Ancel Keys and low fat diet in diabetes really hit a nerve.  (You can read my initial response here, and more about her mangling of Keys here.)   

I tweeted out off-the-cuff, and said tweet gets auto posted to FB, and a discussion ensued.  I was basically taken to task for not acknowledging Keys' central role in all of this.  While I don't deny that the man and his work were important in forming government policies, etc., I just don't see that it all leads back to this one man and that all he had to stand on for his part in all of this was one graph with six countries on it.  The claim that decades later we've all just been duped and are still being lied to about how it "all went down" is more than a bit absurd at this point.   The only thing even more absurd is this notion that we were somehow eating LCHF before and we can "fix" this by going back to eating LCHF again.  

It doesn't help matters much that there are really two parallel hypotheses that are easily "confused".  I put that in quotes because I believe this is often deliberate.  There is lipid-heart based on early findings that elevated blood cholesterol correlated with heart attacks.  This has been refined and revisited in the decades since in search of the best lipid profile biomarker:  TC, LDL-C, HDL-C, non-HDL-C, trigs, this ratio or that ratio, particle size, particle number, apoB, etc.etc.etc.  But going back to the initial findings, the next step was the diet-heart hypothesis, which is probably better designated as diet-lipid.   If elevated blood cholesterol could be the cause of heart attacks, what then might be the cause of elevated blood cholesterol?   Is it dietary cholesterol, all dietary fat, just saturated fats, or animal fats?

All of this has been played out on the background of a nation obsessed with dieting  and an obesity epidemic where weight loss has become the primary goal of dietary intervention.  This all seems to get intertwined into one big low-fat-diets-made-us-fat-and-sick mantra that gets dragged out and recirculated every now and then.  It just so happens that a few days after Shanahan's interview, the prelude to The Big Fat Surprise, an editorial in the WSJ by Teicholz, triggered a tsunami of misinformation and the start of the Nu Atkins Revolution.  

As it turns out, what has been repeated often enough to be accepted as fact is not how things really happened after all.  No.  It has been the modus operandi of Big Butter & Bacon to drag out Keys' Six-Countries Plot and accuse him of cherry picking and all that implies.  This, more than anything, seems to unite and rally the troops.

Off to the Nutrition Texts I Went ...

So, back to the FB discussion.  If Keys were so instrumental in getting us to adopt a low fat diet, then there should be evidence all over the nutrition texts.    In the same way that you learn about Newton when gravity is discussed in a physical science class (no matter how general the class), so, too, should the low fat law of the land be attributed to Ancel Keys.  Given that low fat is even more every-day than gravity and such, Keys' middle name should be regular question in trivia games!   Over the past several years, I've amassed quite a collection of nutrition textbooks to complement the biochemistry and physiology texts in my collection.  Most are ebooks which have the added advantage of being searchable in short order.   My admittedly unscientific sampling of nine books (I got bored after that) turned up one off hand mention of the Seven Countries Study as the study that linked diet, cholesterol and CVD (cue ominous music).  In one of the other books, Keys was mentioned for his Minnesota Starvation Experiment, in, you guessed it, the section on starvation.   Heck ... just to make absolutely sure Keys wasn't really capable of time travel, I even checked the 1917 The Elements of the Science of Nutrition by Lusk.  No sign.

Then I remembered getting a hold of the book pictured at right a few months ago ... took me a bit to find it on the old hard drive, as I'd just downloaded a non-descriptly named file to save for future reference.  

Cardiovascular Diseases:  Nutritional and Therapeutic Interventions.  Edited by Nilanjana Maulik.  CRC Press.  2013.

I finally located the book and promptly initiated my search ... Ding ding ding!!!  Right there in the introduction.  OK ... Here we go I thought ... and I read on ...

The who what wut??  So first things first.  Here is the study ... and it's a rarity in that an early 60's full text is free to all, on the journal publisher's site.  IOW -- no excuses.   

Coronary Heart Disease among Minnesota Business and Professional Men Followed Fifteen Years  (Circulation. 1963; 28: 381-395 doi: 10.1161/​01.CIR.28.3.381)   Ancel Keys, Ph.D.; Henry Longstreet Taylor, Ph.D.; Henry Blackburn, M.D.; Josef Brozek, Ph.D.; Joseph T. Anderson, Ph.D.; Ernst Simonson, M.D.

Relative weight, body fatness (skinfold thickness), blood pressure, and serum cholesterol are reported from 281 Minnesota business and professional men, initially clinically healthy and aged 45 to 55, who were followed by annual examinations since the winter of 1947-48. In 15 elapsed years, coronary heart disease developed definitely in 32 men and possibly in 16 other men.

The incidence of coronary heart disease tended to be higher among men above the median at first examination in relative weight, body fatness, systolic and diastolic blood pressure, and serum cholesterol concentration but these segregations were not statistically significant except with serum cholesterol, which was associated with p < 0.001. The data for the last pre-disease year on the men who developed coronary heart disease showed a significant relationship between incidence and systolic blood pressure as well as for cholesterol. Average data over all pre-disease years showed significantly reduced risk among the men in the bottom quartile for diastolic as well as for systolic blood pressure, while the cholesterol level was significantly prognostic over the entire range of that variable. The few men who developed coronary heart disease with low cholesterol values tended to be in the top 20 per cent of the distribution of blood pressure or relative weight or both....

A True Pioneer ...

This folks, is the first prospective epidemiological study ever conducted on heart disease.   This was huge.  Hah-yooooge!!   Nevermind that snide science journalists and PubMed couch jockeys routinely write the entire science off as pseudo or junk science.  It is not.  As with clinical trials, there are the good, the bad and the ugly.  We don't write off well designed and implemented RCTs because there are poorly designed and implemented uncontrolled trials to be found.  The same standards should be applied to epidemiologic studies.  Many conducted in the realm of nutritional science are pretty bad, especially those designed to evoke provocative "medical news".  Self reporting of intake for a day or few over a period of years really doesn't give you high-integrity data to start with when trying to study diet and lifestyle.  So the red meat or dairy studies are usually to be taken with a grain of salt.  In a perfect world these same people would give the sugar studies of similar quality an equally ready dismissal, but consistency is never a strong trait of the dogmatic.

But there's the other type of epidemiological study that we rely on a lot these days.  Just about every person in the IHC, be they low carb, WAPF, paleo or the various veggie and vegan factions (I've decided to include everyone in the IHC ... it really is that incestral after all), uses this sort of epidemiology in one way or the other.  Whether you are crowing over your low LDL, high HDL, low HbA1c, CRP, apoB, FBG, LMNOP, whatever ... These biomarkers are the product of epidemiologic studies.  Observational studies where various measures are tracked over time and outcomes observed.  In the hopes of what?

Rather than treating the disease after it develops (or burying the victim),  preventing it from happening in the first place.   That's what.

This, dear readers, is Ancel Keys' MOST important contribution to science, in my opinion.  And it is one he should rightly be celebrated for instead of being the target of so much small-minded derision.   

But he was not done there.  You see, Ancel Keys was a critical thinker who believed the key to this prevention lay in diet and/or lifestyle:  that heart disease was not necessarily a foregone byproduct of aging or genetic preprogramming.   This is what sparked his curiosity to look at what dietary factors might be responsible for high rates of CVD in some countries while they remained an oddity in others, and indeed had been rare in the US at the turn of the century.    The way I see it, whether or not you agree with which factor(s) Keys focused on, you cannot dismiss the fact that he was looking for any such factors to begin with.

Hippocrates or Hypocrites?

Ring any bells?  Nobody could sit through a summit, an AHS, or PaleoFX, or frequent many a WAPF, Functional Medicine, Paleo, Low Carb, Vegan blog without coming across this one.   This is part and parcel of the whole "it's 80% nutrition" rule, etc.  

a random Google Images search on "let food be thy medicine"

So either you really believe that, or you don't.  And if you really do, then Ancel Keys is your modern day medical science hero.  No ifs, ands or buts.   Otherwise, you are a hypocrite.  There is no wiggle room here.   Folks are free to have their own ideas and opinions as to which foods are to be thine appropriate medicine, and conversely which they may consider to be poison.  But if you believe the premise of the above quote, then Keys is your guy.  I know this will make some heads assplode with cognitive dissonance of the highest order, but this really is Keys' legacy.  

So Why Haven't We Heard of This?

That's a good question, really.   When I first found the Minnesota Business Men study, that was the first question that came to mind.  It was a big deal.  Since then, in several biographies, it is prominently mentioned.  Indeed it is that Six-Countries Plot that is rather obscure.  Surely the ever inquisitive science journalist Gary Taubes, what with his penchant for pre-war through 60's research and citations, had come across the Minnesota Business Men study?!    So off to Kindle I went ... opened Good Calories, Bad Calories ... initiated several searches.  Nothing.  Not surprisingly, there was equally nothing to be found in mini-me-GCBC aka Nina Teicholz's The Big Fat Surprise.    These two books are both held up as meticulously researched ... taking us through the *real* history of the diet-heart or lipid-heart hypothesis/es.   And yet these two along with WAPF-inspired authors like Cate Shanahan, weave a tale where this seminal work seemingly does not exist.

By their telling, Keys just woke up one day and plucked his ideas from thin air.  (Which is an interesting phrase where he's concerned as he did some high altitude research between earning his PhDs.)    In Shanahan's case, it is even more ridiculous as she cites that 1958 CBS The Search program in her book Deep Nutrition (see this post for lengthy excerpts).   Both Charles and Karin get a hat tip from me for finding the video on YouTube.  Thanks!!

Right after Keys knocks over the random wooden pin-men -- as he relays the ominous reality that in that day and age, we didn't know who might suddenly drop dead of a heart attack or why -- do you know what they show footage of?  Hint:  NOT Ancel Keys pointing to a graph.    They show various men arriving, by car or bicycle, at his Minnesota Laboratory of Physiological Hygiene.  These were participants in this, the first ever prospective epidemiological study on heart disease.   The narration leaves zero doubt.    It's quite humorous to watch the men taking stress tests on treadmills with their black dress socks and garters!    

Narrator:  It was the start of Keys' search for what later would be called risk factors for heart disease.

Nah ... Cate would rather you believe the ego-driven eel expert just decided saturated fat caused heart disease and ran with it because the makers of Mazola were greasing his pockets.   And Taubes and Teicholz would rather you didn't know about this study at all, because the results of it were confirmation that Keys was on the right path.  In 1958 he already had a notion of how things were going in his study, and was looking concurrently at what dietary factors might be responsible for the elevated cholesterol that was associated with heart attacks.  No ... he did not presume a causal relationship.  Yes, he looked at other factors, including hydrogenated oils.  

Through the Eyes of Keys and Blackburn:

There's a 2007 book,  The Development of Modern Epidemiology: Personal reports from those who were there, in which Henry Blackburn recants his account of Keys' role in pioneering cardiovascular epidemiology.  Chapter 7, aptly entitled Cardiovascular disease epidemiology, was written by Blackburn and is available here.   Most probably recognize Blackburn's name as being associated with his characterizations of Keys' blunt nature.  When you read this chapter, as I hope you will, it's not so much that I care about the personal side, but that is shows how unfair these isolated (some might say cherry-picked) phrases about Keys' persona are.   That these seem to be all that make it into books that ostensibly seek to recant not just the science, but the "back story history" of it all, becomes rather offensive after a while.   

Since she's fresh on my mind, Shanahan's telling of the Keys story is especially bothersome.  She makes him out to be some egotistical empty scientific suit of a man who accrued a cult following based on his physical stature and appearance?  In this day and age of celebrity everythings (doctors, lawyers, housewives ... heck, why not scientists!) one might see how this could happen on a relatively small scale.   But we're talking about a man who was almost sixty when the Minnesota Business Men Study was published in the Sixties (he was born in 1904), and of retirement age when the first of the Seven Countries publications hit the journals.  Unlike Teicholz, there was no multimedia book tour to be had.  There was no Ancel Keys Facebook page to "like" or Twitter feed or blog or whatever.  There was no Keys movement cult with "boxes" strewn across the land at which his minions regurgitated his nutritional dogma.  

Heck, even peer review is a shadow of its former self.  Surely some innovative thought was squelched in those days where there was a pecking order and necessary chops to be earned before one reached a stage of some stature and respect.  And yet, even reports by venerable scientists -- like Keys' presentation of the Six-Countries Plot -- were often met with criticism by peers published in the same or competing journals (the Lancet review of Dayton/LA Veterans comes to mind as an example of the latter).    In those days criticism seemed to mean something.  There seems to be a lack of this nowadays.  It is quite rare to see serious back and forth letters to the editor.    

But nowadays we all take for granted that we can have our doctors run some blood tests to see if we are at risk.   Oh ... and all you non-diabetics in the audience pricking your fingers to keep an eye on your blood sugars and diabetes risk?  A. Ben Keys!   He did not invent the test and such, but he began the search for *risk factors*, those measures of something in or about you that might predict disease before a frank diagnosis or sudden death, in the hopes that it might be prevented entirely.

Two excerpts from the chapter:

1.  In discussing The Minnesota Business and Professional Men’s Study, Blackburn cites it as a first but sets it up as the predecessor of larger studies from which more definitive risk factors would be derived, and as a stimulus for Seven Countries.  Then:

Of greater historical interest is how the Minnesota study was conceived in the first place.  In 1948, Keys, who was a physiologist and nutritional scientist, had just completed a classic study of experimental starvation.  In it, he had been struck by the rapid modifiability of heretoforeconsidered ‘constitutional’ characteristics such as body type, blood pressure, blood cholesterol, and vascular responses.  Meanwhile, he had become curious about the phenomenon of epidemic heart attack, apparently from Minnesota newspaper reports of the day (Keys 1999).  We know from an action photograph taken in 1946 that Keys had plotted a hypothetical projection toward intersection of the rising slope of heart deaths and the falling slope of other deaths in the city of Minneapolis.  This antedated by two decades published discussions of ‘the epidemiological transition’.   Keys recounts in his memoirs:

As news of the war and its aftermath receded from the newspapers, I was struck by the frequency of reports of executive men dropping dead, victims of what seemed to be a new plague, an increase in the frequency of coronary heart disease. ... The cause ... was unknown.   I thought that we could record the characteristics of a large number of executives when they were apparently healthy, and then follow their status over the years, during which time some of them would develop coronary heart disease.   By comparing the records of the afflicted men with those of the group that stayed healthy, we could discover which characteristics were related to the tendency to develop the [arterial] deposits.  Keys (1999)

2.  On the Seven Countries Study:

In the 1950s, a different epidemiological approach to understanding the causes of CVD focused on comparisons among whole populations across a wide spectrum of diet or lifestyle in efforts to define the mass, as well as the individual, characteristics related to high or low population risk.  The Seven Countries Study tested specific hypotheses among largely rural, traditional populations contrasting greatly in habitual diet. It was carried out by Ancel Keys and international colleagues in a collaboration starting in the late 1950s and ongoing today (Keys et al. 1970, 1980; Kromhout et al. 2002).   Major differences in CVD risk were found among samples of working middle-aged men followed for 30 years in 14 areas of seven countries, with the lowest rates in Japan and the Greek islands and the highest in the United States and Finland.

The Seven Countries Study provided the first credible data about major population differences in CHD and stroke rates. It also confirmed large variations in death rates from all causes.  It found that a habitual diet of less than 10% calories from saturated fatty acids, with its associated lower blood cholesterol values, was apparently a necessary factor in lower population rates of CVD, whereas other risk factors failed to ‘explain’ the geographical differences.Within Seven Countries populations having different mortality, however, the ‘traditional’ CVD risk factors of diet, serum cholesterol, blood pressure, and cigarette smoking were universally predictive of an individual’s risk, albeit with culturally different ‘slopes’ or force of the relationship (Keys et al. 1972; Kromhout et al. 2002).

The Seven Countries Study provided new and strong evidence for a policy of risk reduction at the population level. And it infected with the epidemiological method and mystique the medical and public health ‘virgin forest’ in each of its seven countries. 

Concluding ... For now ...

There is actually quite a bit more to be said about the research of Ancel Keys and what he is and is not responsible for.  The "other" Minnesota study contains some interesting tidbits as well, that I will discuss some time in the hopefully near future.  

I will leave you with this.  The well-funded anti-establishment forces are quick to point out such things as funding sources, conflicts of interest, so-called errors or flaws in methodology in the science, lack of controlled trials and evidence upon which to base recommendations, etc.   These same forces are blinded to and even cover up their own more egregious issues with all of the preceding.  There are no RCT's showing that eating more fat reduces CVD mortality or morbidity or that for other diseases like diabetes.  There aren't even epidemiological studies demonstrating better health with higher fat consumption.  Absent this there are no cultures they can point to as relevant pinnacles of health and longevity.  No the Masai and Inuit are not such cultures, enough with that nonsense already, OK?   

Exactly what diet does Richard Feinman want people to march on Washington and have the USDA replace current recs with?    Are these people so blinded by ketophoria they don't see that even their long-term adherents are constantly tweaking and twerking to fix new-found health issues they blame on their diets of years and even decades past?   It would be amusing if it weren't for the health of so many that lies in the balance.  And no, I'm not a lipophobe either, there's a whole wide spectrum between the extremes.   

But what diet should we replace it with?  Atkins?  Seriously?  That makes a ton of sense.  Not!   The truth is, as we see from the full 23 ... err ... 22 countries in Yerushalmy & Hilleboe (and let's not forget that those are just a sampling of all of the countries on Earth at the time!!), the "low fat" diet was hardly some reckless human experiment.  Even if all the scientific evidential ducks were not neatly lined up, there were enough actual living human testaments to its benefits, not to mention verifiable historical records and tradition.   Low carb, high fat?  Zilch, zero, nada.  Remind me, just who is ignoring the science?