As one reader recently commented on a different blog, I lit a fire to the China Study five years ago but Denise Minger just burned the whole thing down.
Lighting the Fire
In the spring of 2005, I wrote a review of Dr. T. Colin Campbell’s The China Study for the Weston A. Price Foundation’s journal Wise Traditions. The mostly unchanged version of this review on Cholesterol-And-Health.Com, “What Dr. Campbell Won’t Tell You About the China Study.” currently ranks on the top page of Google results for a “china study” search. We offered Dr. Campbell the opportunity to respond within the pages of Wise Traditions, but after intially accepting the offer and continuing a lengthy private dialogue with me over email, he later rejected the offer and published a mostly ad hominem attack on me the following year on the VegSource.Com website. My original review and my much broader and more in-depth response to Campbell have served as essential reading among those looking for critical reviews of The China Study for several years now.
Burning it Down
Well, there’s a new contender in town and while I may have hit the beast with a few strong punches and made a run for it, Denise Minger went for the jugular and then put a nail through the coffin.
Not the gentlest metaphor, but this “beast” is not The China Study itself and certainly is not its well respected author, Dr. Campbell, but is rather the misuse and abuse of statistics and logical inference that has falsely elevated The China Study into proof that, as Dr. Campbell wrote, “Eating foods that contain any cholesterol above 0 mg is unhealthy.” Ms. Minger’s lengthy but powerfully scathing review of The China Study can be found here:
The China Study: Fact or Fallacy?
Minger makes a number of great points. She provides an excellent preliminary analysis showing that schistosomiasis, a parasitic infection, confounds many of Dr. Campbell’s conclusions. Minger, herself a lover of green vegetables, makes an excellent argument showing that frequency of green vegetable consumption is often inversely correlated with a disease when total green vegetable consumption is not — a phenomenon she calls “The Green Veggie Paradox” — because the “frequency” variable is just a marker for a warmer climate. She also devoted an entire section to the Tuoli people, the “outliers” among the rural Chinese who consumed almost two pounds of dairy every day. These “mysterious milk drinkers” defy the central theme of The China Study by possessing excellent health.
None of these points, however, great as they may be, constitute nails in the coffin. None of them cast the fatal blow, slaying the beast of statistical misuse and logical abuse that has inflated the findings of what would otherwise be an interesting observational study into supposed proof of the superiority of a “plant-based” (vegan) diet. None of these blades, sharp as they may be, bring their incision even close to the jugular vein.
Of what substance, then, is this fatal blow made?
Back to the Experimental Evidence
Denise Minger went not just beyond the bestselling book and into the abyss of thousands upon thousands of data points known loosely as the original monograph but traveled even further through the lands of the roughly one thousand references Campbell provides among his supporting evidence to find a critical study that Campbell himself had published showing, in fact, that the central thesis of his book — that animal foods, and specifically animal protein, are uniquely harmful to consume — is false.
That’s right. Minger’s detailed analysis of the China Study’s raw data is superb, but it’s a handful of sentences she devotes to Campbell’s animal experiments using protein to promote alflatoxin-induced pre-cancerous lesions in rats that lay the issue clearly to rest.
Why are these so important? Because these experiments are, while conducted in rats, the most solid experimental evidence Campbell presents in his entire book that animal products might have a unique ability to promote disease.
Let’s consider for a moment the evidence he presents. The China Study itself is an observational study. It can be used to generate ideas, but not to test them. Moreover, the data from the China Study do not support Dr. Campbell’s position anyway. Even if they did, there were over 8,000 statistically significant correlations. With p values <0.05, this means five percent or one in twenty of them are false correlations that arose by chance. Thus, chances are that there are about 400 correlations in the China Study that are statistically significant but nevertheless false and arose completely by chance. Campbell uses far fewer than 400 correlations to make his argument, so all but the most rigorous correlations significant at the p<0.001 value could well be false. Campbell did indeed cite some experimental evidence in humans, such as the experiments of Esselstyn and Ornish, but these experiments were either poorly conducted or failed to isolate animal foods as a variable.
Of Lab Rats and Men — Men Who Feed Them Protein Isolates
The only rigorously controlled experimental science that Campbell cites in favor of his hypothesis that animal foods, and specifically animal protein, are uniquely harmful to our health, is his own experiments in rats showing that casein, but not wheat or soy proteins, promoted cancer in lab animals.
After finding that feeding rats increasing amounts of casein (one of the several proteins in cow’s milk) promoted tumor growth in rats induced by aflatoxin (the carcinogen found in peanut butter), Campbell went on to investigate whether plant protein also promotes cancer. He wrote on page 59 of The China Study:
So the next logical question was whether plant protein, tested in the same way, has the same effect on cancer promotion as casein. The answer is an astonishing “NO.” In these experiments, plant protein did not promote cancer growth, even at the higher levels of intake. An undergraduate pre-medical student doing an honors degree with me, David Schulsinger, did the study. Gluten, the protein of wheat, did not produce the same result as casein, even when fed at the same 20% level. We also examined whether soy protein had the same effect as casein on foci development. Rats fed 20% soy protein diets did not form any early foci, just like the 20% wheat protein diets.
After discussing similar effects in other animal models of cancer and a few studies showing the ability of antioxidants to prevent cancer growth, Campbell concluded that “a pattern was beginning to emerge: nutrients from animal foods increased tumor development while nutrients from plant-based foods decreased tumor development.”
Really? Was such a pattern truly emerging?
The Fatal Blow
When I wrote my review in spring of 2005 I pointed out that Campbell was jumping the gun by making conclusions about other animal proteins and even all “nutrients from animal foods” when he only studied powdered casein, but I went no further than that. Denise Minger, however, dug up the original study and used it to blow a death knell to Dr. Campbell’s argument. You can find the study here. When the amino acid lysine was provided in the diet, wheat protein had the same effect as casein! The research showed definitively that the only reason wheat protein didn’t promote the cancer was because it is not a complete protein!
Grains like wheat tend to be deficient in lysine while legumes like soy tend to be deficient in methionine. When we say “deficient,” we mean deficient to provide optimal growth. For example, a study from 1979 found that infants fed soy formula in the first six weeks of life grew less and failed to thrive as well as infants who were breast-fed or fed animal milk unless the soy protein was supplemented with methionine.
For a time it was thought that vegetarians must meticulously combined grains and legumes at every meal in order to make a complete protein. But, as Dr. Campbell informs us on pages 30 and 31 of The China Study,
While the “lower quality” plant proteins may be lacking in one or more of the essential amino acids, as a group they do contain all of them. . . . We now know that through enormously complex metabolic systems, the human body can derive all the essential amino acids from the natural variety of plant proteins that we encounter every day. It doesn’t require eating higher quantities of plant protein or meticulously planning every meal.
Yet, amazingly, Dr. Campbell never informs us that while soy protein or wheat protein alone does not promote cancer, as a group plant proteins do contain all the amino acids to promote cancer, and that we now know that “the natural variety of plant proteins that we encounter every day” can, in fact, promote cancer just as powerfully as animal protein.
It appears, then, that what Dr. Campbell found was that under certain experimental conditions where high doses of a carcinogen have been used to initiate pre-cancerous lesions, complete proteins divorced from all the natural protective factors they are associated with in whole foods will, through their growth-promoting properties, promote the growth of those cancers.
Dr. Campbell and his colleagues, including one of his graduate students, conducted the study, and should have informed us of the result.
But it took the skepticism and skillful detective work of Denise Minger for us to find out.
Read more about the author, Chris Masterjohn, PhD, here.
🖨️ Print post
Although I am sympathetic ethically to vegetarianism, having been a practitioner for decades, removing gluten and casein products from my diet, while adding liver, raw eggs, heavy cream, coconut oil, and organic dark chocolate has renewed my health and vitality by shedding 70 pounds of adipose tissue, and improving brain function. I have not felt this good, nor looked this good since high school! Judicious use of animal products can improve the health of many individuals. Even the 30 year follow-up of the Framingham Massachusetts heart study published in the Journal of the American Medical Association states clearly that after age 50 high cholesterol is correlated with longevity, and falling cholesterol is correlated with mortality. Yet, it is this senior population that is bamboozled into taking statin drugs to damage their health outcomes.
Since cholesterol is the substrate of steroid hormones that decline with age, and is a key compound in membrane integrity of every cell in the body, no wonder seniors benefit from cholesterol, especially when their is no contraindication of a vasculature that is inflamed by oxidative stress. Joseph Campisi
Anderson KM, et al. Cholesterol and mortality. 30 years of follow-up from the Framingham study. Journal of the American Medical Association, 1987; 257: 2176-2180.
Jacobs D, et al. Report of the Conference on Low Blood Cholesterol: Mortality Associations. Circulation, 1992; 86: 1046-1060.
Strandberg TE, et al. Low cholesterol, mortality, and quality of life in old age during a 39-year follow-up. Journal of the American College of Cardiology, Sept 1, 2004; 44 (5): 1002-1008.
Although I am sympathetic ethically to vegetarianism, having been a practitioner for decades, removing gluten and casein products from my diet, while adding liver, raw eggs, heavy cream, coconut oil, and organic dark chocolate has renewed my health and vitality by shedding 70 pounds of adipose tissue, and improving brain function. I have not felt this good, nor looked this good since high school! Judicious use of animal products can improve the health of many individuals. Even the 30 year follow-up of the Framingham Massachusetts heart study published in the Journal of the American Medical Association states clearly that after age 50 high cholesterol is correlated with longevity, and falling cholesterol is correlated with mortality. Yet, it is this senior population that is bamboozled into taking statin drugs to damage their health outcomes.
Since cholesterol is the substrate of steroid hormones that decline with age, and is a key compound in membrane integrity of every cell in the body, no wonder seniors benefit from cholesterol, especially when their is no contraindication of a vasculature that is inflamed by oxidative stress. Joseph Campisi
Anderson KM, et al. Cholesterol and mortality. 30 years of follow-up from the Framingham study. Journal of the American Medical Association, 1987; 257: 2176-2180.
Jacobs D, et al. Report of the Conference on Low Blood Cholesterol: Mortality Associations. Circulation, 1992; 86: 1046-1060.
Strandberg TE, et al. Low cholesterol, mortality, and quality of life in old age during a 39-year follow-up. Journal of the American College of Cardiology, Sept 1, 2004; 44 (5): 1002-1008.
Look forward to your revised analysis from Denise. As established by an expert who interprets similar statistics for a living (a cancer epidemiologist) Denise’s initial analysis was heavily flawed and therefore invalid. Of course this doesn’t mean she can’t redeem herself!
A few links of importance…
Rebuttal to the Rebuttals
http://www.30bananasaday.com/group/debunkingthechinastudycritics/forum/topics/a-cancer-epidemiologist?page=1&commentId=2684079:Comment:633526&x=1#2684079Comment633526.
Joseph,
Thank you for writing. I’m glad you were able to improve your health!
Chris
Dear Freelee,
Thank you for writing.
This cancer epidemiologist clearly has an excellent grasp of statistics. On the other hand, this writer appears to have skimmed Ms. Minger’s analysis without actually reading her conclusions and clearly did not even read The China Study.
They write the following:
“She uses correlations and ecologic comparisons to draw conclusions about relationships between diet and outcome (cancer, cardiovascular disease, etc.). WRONG WRONG WRONG!!”
On the contrary, the “China Study” provides no other data except “correlations and ecologic comparisons.” Minger did not draw unjustified conclusions from this data, but simply showed how Dr. Campbell ignored much of the data, and, in the few cases where the data seemed to support his conclusions, he ignored confounding variables. She did not use the data to justify dietary conclusions.
Moreover, it appears you did not read my review either. I clearly concluded that the most significant contribution of Minger’s work was uncovering the rat experiments that Campbell formed showing that the only reason plant proteins did not promote cancer in his experimental model was that they were deficient in certain amino acids, which implies that were multiple plant proteins combined to form a “complete protein,” as they are in any reasonably natural and healthy vegetarian diet, they would have the same effect in promoting cancer.
This epidemiologists criticisms have already been posted on Ms. Minger’s blog and elsewhere, so if you would like to make a criticism here, please make it pertinent to the aspects of her work I focused on in my blog.
Nevertheless, thank you for your interest in this important subject and for sharing your opinion.
Chris
Thanks Chris for your wonderful review and insight on Denise’s critique of the China Study. Omission of these interesting facts on plant proteins by Dr.Campbell sounds like blatant biasness to me. Both your work and hers have surely put the nail through the coffin on this issue. Rest in peace, China Study.
I found it really amusing that all the criticisms aimed at Denise’s methodologies and findings even though we do not know Campbell’s exact method in analyzing the study. They insist too for her work to be peer reviewed by a reputable journal. What unfairness!
Thanks for writing, Daniel. I agree that most of the criticisms of Denise’s work are pretty off-target and often use extreme double standards. I do think, however, that she has raised some issues that would be excellent to analyze with more in-depth statistical methods and submit to a peer-reviewed journal.
The other thing that seems to be missing from the casein-cancer connection is any elucidation of underlying mechanism. Maybe I just haven’t looked hard enough, but even though that variable is theoretically “isolated” in the rat experiments, it’s very difficult to extrapolate to free-living organisms with any confidence if you don’t have at least some idea of the underlying metabolic cause.
And aren’t proteins generally reduced to amino acids in the GI tract? Following Campbell’s logic, that would mean that it is the specific proportion of amino acids that drives tumor growth, regardless of whether those proteins are of plant or animal origin. Amino acids is amino acids, whether they come from wheat or cow. I believe your point on completeness hits it on the head, and at least provides the basis for design of experiments that can provide deeper insight.
Chris – I think I’ve mentioned it to you before, but I am again going to recommend Jaynes book “Probability Theory: The Logic of Science”. I am confident you would “get it”, in spades. It’s pricey, so get a preview by reading the first three chapters free online:
[url]http://bayes.wustl.edu/etj/prob/book.pdf[/url]
The preface alone is well worth reading for concise insight into what is right and wrong with how statistics are commonly used (and misused) for scientific inference.
Dear Dave,
I agree with you that we need to understand the underlying mechanism better to interpret the findings. However, I believe it makes a lot of a sense that anything required for growth would also be required for tumor growth. The cells cannot multiply without a sufficient supply of the amino acids necessary to make new proteins. This is the same reason protein is required for growth during childhood, growth during bodybuilding, tissue repair, or any other case of growth.
You are right that it is about the amino acid composition, and I do think Dr. Campbell’s ommission of this lysine factor is pretty incriminating.
Thank you for the recommendation. God willing I will read it — probably not till next year though!
Thanks again for writing,
Chris
I followed the McDougall Program which is a high carbohydrate, low fat diet for nearly four years. I believed whole heartedly in this book and even bought copies for all of my relatives when it first came out. Although my health improved initially as a vegan I eventually developed fatigue and dental problems which only went away after I began eating animal fats and saturated plant fats again.
I saw this critique and before even reading it, recognized the name as I knew Denise here in Portland several years ago when she taught at my sons after school program. I knew she was a wonderful artist but did not realize she also was such a skilled researcher and writer. I am amazed by the time and effort this project must have taken and blow away by her intelligence. I had to ask my husband for help understanding some of the more technical points but it’s obvious she has a strong grasp on this.
I am also amazed anyone could respond so civilly after receiving a response like the one Campbell wrote accusing her of working for an industry or lying about her identity. I can vouche she is every bit as gracious and kind hearted in real life as she comes across in person. Campbell could benefit from following her example.
Chris and others, in case you have not seen her response she posted it today on her blog.
http://rawfoodsos.com/2010/07/16/the-china-study-my-response-to-campbell/
Dear Corinne,
Thank you for your comments, and for posting the link to Denise’s response. I did see it today but thank you for making it available to others who may have missed it. The reply is excellent and Denise is brilliant.
I’m glad you’ve been able to recover your health. I also had a bad experience with vegetarianism. I hope you continue to maintain wonderful health and continue to improve it.
Chris
In a new blog I respond to a few comments directed my way by Dr. Campbell as well as a paleontologist-in-training named Robert from the “30 bananas a day” site in response to a query given him by Freelee, who posted a comment on this blog:
http://www.westonaprice.org/blogs/denise-mingers-refutation-of-campbells-china-study-generates-continued-debate.html
I don’t know if it is only complete vs incomplete protein when it comes to cancer. Denise cited several studies showing that whey protein had anti-cancer effects and whey is a complete protein.
But Campbell is definitely wrong when he says that his studies showed that casein caused cancer while wheat protein did not. In the case of wheat, it was the incompleteness that made it not cause cancer. It is a very good for you and others to point it out.
But there has to be something more to it, considering whey. I am not familiar with research regarding other proteins effects on cancer, maybe the cancer effects are unique to wheat protein (gluten?) and casein, or maybe there is some other pattern when one looks at other research regarding other proteins?
Mike,
The effect of whey protein is probably due to its promotion of glutathione synthesis. I’ll be posting on the effect of undenatured whey protein, raw milk, and other raw foods on glutathione levels.
Chris
This blog may not be monitored anymore but I will throw my not so formal or officially educated opinion into the ring.
My mother was diagnosed with a cystic adenoid carcinoma last year. Since then I have taken an extremely keen interest in the initiation and development of cancer, specifically the nutritional aspect of it. It is well known that the treatment of chronic disease on this continent (i am from Canada) is horrendous. We essentially provide people the opportunity to live (comfortably? ) with their ailments by supressing symptoms.
As mentioned above I have no formal education or training in the medical field or nutrition, for that matter. However, an opinion or viewpoint from someone outside the box can perhaps be informative (maybe!).
I have just recently finished reading Proteinaholic by Garth Davis. A nice follow up to both the China Study and Whole. His book sums up when and where our obsession with protein came from, how it has been perpetuated and how it affects our bodies (not to mention the environmental impacts of the entire animal agricultural industry).
I am not a self-proclaimed authority by any means on this topic but I can say from the personal experience of drastically altering my diet from very animal based to plant based I have seen incredible results. At one point in my life I would consume a one pound rib eye steak for supper about 3 times a week. Nothing else, just steak. I couldn’t imagine a better tasting meal! Even though my taste buds were very happy there was still something in the back of my head that said I was probably doing the wrong thing by consuming that much meat. However, if I consumed a pound worth of greens or apple or any plant based food that thought in the back of my head would never have come to life. I think the idea of animal consumption is more cultural than anything, especially a requirement for optimal health.
I have not completely removed animal based foods from my diet but I have reduced them from roughly 30-40% of my food intake to less than 10%. My cognitive ability and clarity are much improved and my energy levels are where they should be for a 32 year old even with having two children under seven.
I admit I haven’t read Denise’s review of the china study but I would imagine it follows a typical review guideline of pointing out flaws and arguing semantics. Not to say the china study is flawless by any means but it gives a perspective that is out of the normal paradigm of thinking , something I have always been inclined to do on many levels. Without risking sounding like a conspirist, the doctors and professionals performing these studies have either been privy to or are well versed on how much financial influence industry has over what people think is good for them and what eventually ends up on their fork.
I imagine most people writing and reading these posts are familiar with the concept of mimicry. From an evolutionary stand point gorillas are closely related to humans; a species that are essentially plant based eaters. They only consume animal based foods in time of famine. We always seem to relate animal based protein with the ability to thrive, become stronger and build muscle. Well, in the words of Garth Davis, “I challenge you to pick a fight with a gorilla.” Obviously a facetious statement but I think you Gt the point.
Don’t be too hard on my uneducated opinion but I figured it might be worth throwing in my two cents. 🙂
I like meat.
I think I should study multivariate multiple nonlinear regression and the DOE like I’m Fisher’s man, and then pop it into python. I think the power of liver and marrow will give me the ability to grasp the intricacies of causation and correlation for this Caucasian. Data (is so delicious)!!!!!!
Hey, I like fruit and veggies too.