PDA

View Full Version : June Challenge - Ancel Keys and the History of Low Fat


mcsblues
06-08-2006, 07:18 AM
Gabe sent me a gentle reminder (dare I say challenge!) that my contribution to this month's task is ... well ... notable by it's absence! - I do have an excuse ... believe me it is a really, really good one!:) ... but here is something I prepared earlier (ok, who said that is cheating??:p)

Ancel Keys

You don’t have to venture too far into the science behind low carb and conversely ‘our’ ill advised excursion from the already perilous journey since the adoption of agriculture that low fat represents, (ok time for a breath now!) to run into the name Ancel Keys.

PG directed my attention to paper a while back – which contains this quote

“In the classic late-1950s Seven Countries study, a dietitian went into people’s homes, weighed and recorded foods for 7 days and took a replica of each meal for chemical analysis.4”
http://www.mja.com.au/public/issues/184_02_160106/sta10572_fm.html

Yes this piece of drivel comes from no less an authority on diet than Rosemary Stanton :rolleyes: (as mentioned elsewhere, for those outside of Australia, imagine the love child of Susan Powter and Dean Ornish ... oh sorry, were you eating??:p)

It got me thinking that I bet she hasn’t even read the Seven Countries study – which is one of the cornerstones of the low fat movement’s lipid hypothesis, but since I hadn’t either (it’s a bit hard to track down – at least online) I thought I should. Anthony Colpo helped me out with a copy of his along with a few trees worth of other homework (he initially couldn’t find his copy of Keys) – but I have read it now along with the rebuttal by Yerushalmy, which followed shortly after but sadly didn’t achieve the same prominence.

Before I get to Keys – my idea that Stanton may not have read it stretches to near certainty now, as the ‘Seven Countries study’ has absolutely nothing to do with dieticians going into people’s homes, no food samples were taken, the data on which it is based comes from the 40s (not late 50s) – and oh yes, the reference she supplied refers to an obscure comparison of the diets of two Greek villages from 1966. Apart from that, she is spot on.*

OK, way back in January 1953 Ancel Keys then working at the University of Minnesota presented a paper at a symposium in New York. His presentation centred around the idea that Americans were succumbing to heart disease at a greater rate than people from many other countries and included his suggestions as to why this might be so. He felt that as the nature of the problem was apparent from comparative statistical analysis of death rates and the causes of those deaths in various countries – the answer might also be found in what information was available about differences in national dietary intake.

But before we get into what he ‘found’, he discussed why it was he was specifically looking at fat intake. He surprisingly (to me) admits that dietary cholesterol has no impact of human serum levels. I was under the impression that he only tumbled to this fact later in life – but no, he conducted tests and reports similar research where human subjects were given massive doses of cholesterol with minimal (and very short term) impact. He notably dismisses animal research in this context because the impact of cholesterol intake on animals who don’t usually consume any (such as rabbits and chickens) is hardly relevant in a human context. He also acknowledges that cholesterol either as a total or in its various sub fractions is a very poor indicator of an individual’s risk of CHD … but he finds as a (very dubious) fact that there is a general association between those that have had angina or a myocardial infarction and higher levels of cholesterol and a similar association with diabetic patients. He then makes the causation link by referring to … animal studies! :rolleyes: and his final ‘fact’ is that the analysis of atherosclerotic plaques reveals they contain cholesterol, so put down your glasses folks … cholesterol obviously causes any and all diseases of the cardiovascular system, so any dietary means of reducing cholesterol will save lives! :lol:

So how do you lower cholesterol? No statins back then, but at the time it was well accepted that cholesterol could be lowered by means of the “rice fruit diet” (hmmm really sounds low carb doesn’t it?) Now this diet had virtually no cholesterol, but as he had established this wasn’t relevant, our attention is directed to the other distinguishing feature - it was extremely low fat … “about 1 to 1.5% of total calories”!! :eek: Just as an aside it seems these extreme diets were trailed on institutionalised schizophrenic men – which I think speaks volumes about the nature of their consent to be used as human guinea pigs, just as it does for the likelihood of such a diet being followed through in the real world. Anyway in Keys’ mind, as it was established that cholesterol was the problem and low fat was the answer, fat intake was the area of enquiry for the epidemiological data.

It doesn’t take a giant intellect to point out the huge holes in his logic, but once these facts were established and accepted by the scientific community of the time, the rest of his thesis probably made so much sense that little attention was directed at the statistics, which supposedly supported it.

This brings us to the data and I won’t bore you too much with the details. Suffice to say the information he relied upon was by its very nature inaccurate

1. The fat content was not calculated on known food consumption but upon how much was available for consumption (think about that for a bit!)
2. The classification and recording of causes of death was markedly (and obviously) different in many of the countries (notably Italy).
3. The time frame in which the data was gathered (late 40s) was hardly a time to obtain a realistic picture of long term normal dietary intake which may have influenced death rates – shortages and rationing persisted well after the war in many of the countries studied leaving a complete misconception of what the so called “Mediterranean Diet” amounts to - which persists to this day.

But the number one thing Keys is renown for his transparently obvious selective representation of the data available to him to get the result he wanted. Quite simply he picked 6 countries of the 22 he had data for to show an astonishingly simple relationship between fat intake and CHD. Thus the Six Counties study was born (Sweden was subsequently added … because it also fitted quite nicely on Keys graph – hence the confusion between some people also referring to Keys piece of artwork as the “Seven Countries Study”).

Here is Keys graph;
http://home.iprimus.com.au/scott_mc/Keys%20diagram.jpg

But as Yerushalmy pointed out, if you plot all the data points there simply is no meaningful association between fat intake and death rates from heart disease.

Sadly Keys was an influential part of an advisory committee to the American Heart Association (AHA) and his fraudulent assessment of CHD dietary risk factors remains an article of faith amongst those who still promote this dangerous low fat high carb nonsense today.

Keys A. Atherosclerosis: a problem in new public health. Journal of Mount Sinai Hospital, 1953; 20:118-139

Yerushalmey J, Hilleboe HE. Fat in the diet and mortality from heart disease. A methodological note. The New York State Journal of Medicine, 1957; 57: 2343-2354

If anyone wants the pdf versions I have made of these 2 papers, just email me.

* ( I may have been ever so slightly unfair to the Stanton woman - there were certainly a lot of other Keys papers - including a Greek villages one - but the "classic" "Six Countries study" (or Seven) is certainly the first one from which all this nonsense was born.)

mcsblues
06-08-2006, 07:25 AM
The Lutz Version

W. J Lutz (yes, the Life Without Bread guy) wrote a paper in 1995 - "The Colonisation of Western Europe and Our Western Diseases" - Medical Hypotheses (1995) 45. 115-120 - where he offers another suggestion for the differences in heart disease mortality rates in different countries (yes all of the ones Keys had data for, not just 6!!).

His idea is that the differing times various populations adopted agriculture would have an effect on their ability to handle the greater carbohydrate load. I think this suggestion has quite a deal of merit as it is known that early adopters of animal husbandry/herding have a substantially greater % of the population being lactose tolerant even after a few thousand years. So while in evolutionary terms timescales like this are pretty short - it seems some changes can occur.

Although 10,000 years is the widely quoted figure for the introduction of agriculture there is evidence that some groups started earlier, some later and some (like the Aborigines and Inuit) never did. On top of that, there would have been groups who harvested wild grains on a seasonal basis before planting their own crops - all of which brings us back to Lutz who makes the distinction between countries whose ancestry stems from southern Europe and the Middle East, where the earliest evidence of organised agriculture is found, compared to that of the Northern Europeans who have had significantly less time to adapt.

Like his book, I'm not that convinced by his use of this idea to massage Keys data - but I think the idea is sound, and with today's science of genetic halotyping. you would imagine it would be a relatively simple thing to test the relative insulin resistance of people with different ethnic backgrounds.

Certainly the evidence of groups like Aborigines and Inuit being dramatically affected by the sudden impact of a high carb Western diet would give credence to Lutz's proposition.

This idea ties in to a certain extent with what Loren Cordain was referring to in the papers recently refernced by Mike Eades.

Again, if anyone wants a pdf of the paper let me know.

Billie
06-08-2006, 08:17 AM
Malcom, gentle reminder or not, you always bring alot to the discussion groups. We just got home from a 17 hour drive about 1 AM and as soon as I get a few hours to get collected, get to the office and do payroll, pick up the pooch and generally get in reality mode, I want to catch up. Thanks!

Shadow
06-08-2006, 10:27 AM
Wow, Malcom, what an awesome - and amusing - synopsis! I made a note to get back to it today to read it closer but in the meantime wanted to say "thanks" for a job well done :D!

Billie
06-08-2006, 08:17 PM
Malcom beside the Lutz version, are there any more papers that you are aware of on this matter? I am wondering, because of the dates of the paper obviously, but it seems so worthy of a study? And if you would send me a PDF of it when you get a chance I would like to read it printed out.

Thanks
Billie,

mcsblues
06-08-2006, 08:34 PM
Malcom beside the Lutz version, are there any more papers that you are aware of on this matter? I am wondering, because of the dates of the paper obviously, but it seems so worthy of a study? And if you would send me a PDF of it when you get a chance I would like to read it printed out.

What, homework as well??:p

Gabe has uploaded the Lutz paper here;

http://www.proteinpower.com/forum/showthread.php?t=68

- and I believe he is going to do the same with the Keys and Yerushalmey pdfs I sent him ... in the mean time he might share them with you? ;)

The 'adaptation theory' - as Lutz calls it I think sounds eminently logical to 'us' and of course people like Lutz and Cordain ... but I'm not sure how many other scientists are even looking in this area ... but I will try a search (and perhaps others can do the same). Part of the problem with people like Lutz is he published a lot of papers in some foreign language ;) (how dare he do that!!:)) I can (mostly) translate American into English ... but there are limits to my mastery of foreign tongues!

Billie
06-09-2006, 06:47 AM
Ahhhh the Gabster is the villan here, he will be chastized! He is reading papers I do know that, at least he is carrying around a yellow marking pen and papers :D . I did the reading on vacation he played with all my sons technology gadgets--they both can be dangerous in that area! Thanks Malcom I will get him moving on those uploads.

Gabriel Guzman
06-09-2006, 08:24 AM
"While you were sleeping..." I was trying to upload two of the papers but it didn't work. I think they're just too big (more than 1 MB) in size so I'm afraid we can't upload those. If somebody is interesting in reading them, I'll be happy to e-mail the papers.

mcsblues
06-09-2006, 08:55 AM
See what happens Gabe? You stir me into action and end up getting chastized yourself! (although you may enjoy that ;))

Likewise, if anyone wants the pdfs I made, they can email me for them.