Talk:Seven Countries Study/Archive 1

Criticism of research

The major flaw with this study is that Ancel Keys had access to data from 22 countries but discarded the 15 countries that did not fit the hypotheses he tried to prove. The Danish doctor Uffe Ravnskov has some very enlightening information on the issue here: http://www.ravnskov.nu/myth4.htm

This is exactly true, and the Criticism section needs to be greatly enlarged, to include some of the very trenchant criticisms about the study from Gary Taubes, Michael Eades, Tom Naughton, and other low-carb scientists, researches, and commentators. The dropping of non-conforming data from countries like Holland and Chile needs to be prominently mentioned, so that impartial readers understand that this study, the very basis of the Lipid Hypothesis, has structural flaws invalidating its conclusions. — Preceding unsigned comment added by 155.33.208.227 (talk) 17:13, 27 May 2011 (UTC)

Right now the criticism section is fairly week. It does not really come across how bad this study actually is. "The data from 15 of these countries was not included in the overall study, a fact which some have claimed amounts to 'cherry picking. This sound a little too held back for my taste. Unless there is a counter point why only the countries which "fit best" are included and all others are excluded, this IS "cherry picking". It should come across a little better that the criticism is not just a few people moaning about the study (which it kinda feels right now), but that there are major scientific flaws and that the inclusion of the data of all countries does invalidate the hypothesis. I know that I am biased and that this is a hard topic, because most people probably stand on either side of the fence, but I think it does not come across well enough how "flawed" (to not use the word "manipulated", which in my oppinion is the correct term) this study really is. Maybe a graph of the data of all 22 countries and a graph of those countries and the sugar and/or carbohydrate consumption could be included to demonstrate the difference. I know I have seen someone showing them in a lecture/talk. --92.224.55.1 (talk) 16:25, 6 July 2012 (UTC)

I removed some of the critisism since much of that have no relevance for the article. Ancel Keys referred to six countries in his landmark speech and the paper that followed in 1953. That was based on the data compiled by the FAO. The 7CS study was initiated at the end of the 1950s and came out in 1970s. The 7CS was prospective cohort study and had nothing to do with the paper Yerushalmy and Hilleboe was addressing, a data from the 22-countries that Keys also used in 1953. A point which is made in numerous textbook. The authors Y & H simply pointed that Keys had overlooked the role animal protein which had whopping 0,756 correlation to deaths from atherosclerosis (see Yerushalmy and Hillboe paper 1958, figure 15 with correlation co-efficients from various dietary components to death rates) compared to 0,684 for animal fats to deaths from atherosclerosis. The critisism towards Ancel Keys is mainly done in low-carb layman books which often have their Ancel Keys story 100% factually flawed. These are not peer-reviwed works and should be dealt with high level of skepticism.Podomi (talk) 10:48, 14 July 2012 (UTC)

I removed the condemnatory statement about the study in the introduction. Until the issues above are resolved, it is inappropriate to place statements like that outside of the Criticism section. Thr33phas3 (talk) 00:19, 24 March 2013 (UTC)

The issue seems unanswered and seems very staight-forward. Is there any documentation for Keys selection of his particular subset of available countries? The "data compiled by the FAO" came from how many countries? Can it be shown that Keys, doing reasonable diligence, would not have reasonably been able to collect data from more than these seven countries? If, as is claimed, data from many more countries was available to Keys, and especially if the addition of thesel countries data significantly alter the plots, as has been charged by many authors, then either Keys needs to have explained his selection process or accept the view that he "cherry picked". The above paragraph seems to divert from this core issue. — Preceding unsigned comment added by 50.47.4.186 (talk) 22:40, 30 July 2012 (UTC)

The statement in the criticism section suggesting that the countries selected was "primarily a matter of convenience" either need to be meaningfully referenced or deleted. What is meant by convenience? Did Keys state that this was his reason? Why seven; wouldn't three be even "more convenient"? Is convenience, whatever it is, a legitimate argument? — Preceding unsigned comment added by 50.47.4.186 (talk) 22:54, 30 July 2012 (UTC)

The abstract of Ancel Key's "Seven Countries: A Multivariate Analysis of Death and Coronary Heart Disease" reads, "The highest risk factors were found to be age, systolic blood pressure, and serum cholesterol concentration (related to saturated fatty acids in the diet). Differences in incidence rates could not be shown to be related to characteristics of the cohorts in relative body weight, smoking habits or physical activity." How does this square with the Wikipedia article's claim that "The Seven Countries Study confirmed that cigarette smoking is a highly significant predictor of the development of coronary heart disease, leading to excess rates of angina pectoris, myocardial infarction (MI), and coronary death[...]"? 24.218.111.172 (talk) 11:55, 31 March 2013 (UTC)

Everyone here seems to be talking about Keys' paper from 1953, Atherosclerosis: a problem in newer public health, where he plotted food and mortality data from six countries, not the Seven Countries Study. Although he only plotted six countries when data was available for 22, it was found that the data in the other 16 countries was unreliable. Some of the countries didn't even have proper death certificate procedures in place. He wrote in that paper, "The list in Table 22 includes the countries with good vital statistics which are reasonably comparable in race, climate, culture, medical services and vital statistics. The omissions are Western Germany and Finland, because of major population shifts and other effects from the war, and Iceland, Luxembourg and some colonies and semi-independent states with very small populations." The data was not cherry picked, as claimed. If he did cherry pick, then he wasn't very good at cherry picking as he could have chosen a different 6 countries (or even more) to better suit his argument. Keys' six country graph also only plotted total fat, not saturated fat, against heart disease. This is because, at the time, Keys thought that all fat was to blame. He later changed his mind on this and focused on saturated fat. When Yerushalmy and Hilleboe, in their 1957 paper, Fat in the diet and mortality from heart disease; a methodologic note., examined all 22 countries they didn't find a statistically significant link to fat, however they did find a statistically significant link to animal protein. The reason for this being animal fat's direct relation to saturated fat. The authors wrote "...dietary fat is not more strongly associated with heart disease mortality than are the other components of the diet… In all categories of heart disease the association is strongest for animal protein expressed in total calories." This fact is never pointed out by low-carb bloggers and book peddlers like Taubes and Teicholz, nor cholesterol deniers such as Ravnskov. Swampf0etus (talk) 15:03, 21 November 2016 (UTC)

References

Source of criticism

Hello. Can anyone here tell me what is the source for the following? This is the majority of the criticism.

"They also pointed out that Keys was studying a "tenuous association" rather than any possible proof of causation. Moreover, there were other factors that could have been considered. For example sugar consumption was not studied, yet might have shown a stronger correlation, and been a better candidate for dietary intervention than fat. It should be noted that the most healthy people, in Crete and Corfu, ate less than 15 pounds of sugar per person per year, and in Japan, less than 40. This can be contrasted with 60 pounds of sugar per person in England and Wales during sugar rationing in World War II."

The only thing we have is a chapter from a book about diabetes that doesn't mention either Keys or Seven Countries. I'll be removing most of the paragraph in about a week if there are no objections. Thank you. -SusanLesch (talk) 00:58, 15 February 2015 (UTC)

 Done. I'm trying to replace this with cited criticism. -SusanLesch (talk) 20:32, 20 February 2015 (UTC)

SusanLesch Thanks for reviewing this and taking action. Blue Rasberry (talk) 03:32, 21 February 2015 (UTC)

Thank you, Bluerasberry. Sorry to say I don't know how to do citations without templates. Because templates probably weren't used by other editors I'll learn how. -SusanLesch (talk) 04:37, 21 February 2015 (UTC)

References

Help with Taubes

Hello. Could anyone here write a summary of Good Calories, Bad Calories by Gary Taubes? It belongs in the Criticism section. At the moment I'm looking up a number of articles by Taubes. Thanks in advance for any help. -SusanLesch (talk) 17:31, 11 March 2015 (UTC)

Misquote in lead

Hello. I have rewritten this sentence which was in the lead:

"Certainly in the view of Ian Leslie, this debate first became a shouting match owing to the conduct of Mr. Keys, as Keys was "critical to the point of skewering”."

Really bad news. May I suggest that the person who wrote that please read the source. -SusanLesch (talk) 01:50, 10 June 2016 (UTC)

French paradox

Maybe the French paradox article should be referenced somewhere.MaoGo (talk) 14:23, 23 November 2017 (UTC)