Wheat and Obesity: More from the China Study

Obesity seems to have the same cause the world over, in both humans and lab animals: It results from food toxicity compounded by malnourishment.

The most important food toxins are cereal grains (especially wheat), omega-6-rich vegetable oils, and fructose from sugar and corn syrup.

The United States government in its wisdom chose to give agricultural subsidies to wheat, corn, and soybeans, thus reducing the price of the most toxic food crops. The natural result was to make Americans the fattest people in the world, as this chart comparing body-mass index (BMI) in 30 countries attests (from http://diabetescure101.com/BMI.htm, via Matthew Yglesias):

Look at the two low-obesity countries at the left of the chart: Japan and Korea are the only two Asian countries in the sample. What are they doing right?

Well, everything:

  • Their staple starch is rice, not wheat. Rice is the only non-toxic cereal grain and one of our “safe starches.”
  • They consume far less vegetable oils than Americans.
  • They consume far less sugar than Americans.

With a minimally toxic diet, it’s almost impossible to become obese.

Of the three main toxins, which is the most important single factor in causing obesity?

An interesting place to look for an answer is the China Study. Since some regions of China traditionally eat wheat and others rice, the China Study is especially effective at distinguishing the obesity-promoting effects of wheat toxins. The China Study compiled massive epidemiological data by region, including correlations between wheat and rice consumption and body weight and body-mass index (BMI).

Denise Minger, who is fast making herself the world’s leading interpreter of the China Study, has continued her analysis of the raw data. Denise points out that in China, wheat eaters are “fatter with fewer calories”:

In both China Study I and II, wheat is the strongest positive predictor of body weight (r = 0.65, p<0.001) out of any diet variable. And it’s not just because wheat eaters are taller, either, because wheat consumption also strongly correlates with body mass index (r = 0.58, p<0.001). 

Here is the data in pictures:

Denise goes on to point out that the correlation of BMI with wheat intake is 56%, with calorie intake only 13%.

It seems the evidence is consistent.  It’s not eating more calories that makes a person fat. It’s overdosing on toxins – especially wheat toxins.

Leave a comment ?


  1. I’m curious what you think of quinoa as a starch as compared to rice? How toxic is it considered to be?

    Thanks and looking forward to your book.


    • Hi Mark,

      Well, quinoa is not a grain or legume and so does not fall in our excluded foods, but we haven’t been including it in our “safe starches” list either, mainly due to caution and unfamiliarity on our part.

      Quinoa does contain saponins (http://www.ncbi.nlm.nih.gov/pubmed/18452959) even after being debittered (http://www.ncbi.nlm.nih.gov/pubmed/11829658), and needs proper handling including washing before cooking. With proper handling it seems to be relatively safe. A Pubmed search on “quinoa toxicity” yields nothing; “quinoa saponins” yields 22 hits.

      Quinoa does have a lot of protein (12-18% per Wikipedia), which makes me suspicious. Plant proteins are behind nearly all the toxicity effects of grains and legumes, and I prefer to avoid plant protein. It’s possible there are undiscovered toxicity effects.

      There was a debate in the Journal of the American Dietetic Association in Nov 1999 about this. (See http://www.ncbi.nlm.nih.gov/pubmed/10570670 and http://www.ncbi.nlm.nih.gov/pubmed/10570669.) One writer notes that the Celiac Sprue Association classed quinoa as “unsafe” because some celiac patients have had bad reactions to it. This does not mean that quinoa would cause problems to someone with a healthy bowel, but there is a lack of evidence either way.

      Off the cuff, tentative view? Quinoa seems to be quite a bit safer than grains, and since “the dose makes the poison” it can probably be eaten in moderation as a way to diversify the diet. However, its safety is not as well proven as, say, rice or sweet potatoes or taro, and personally I wouldn’t make it my primary starch source.

  2. I have been trying to construct a Perfect Health Diet, perfect day’s meal plan to use as a starting point. I know taste varies and variety is key, but I would like to know what a 1800 or 2000 calorie per day would look like, based on percents, weight and calories. I’m using a spreadsheet to track Glucose, Fructose, SaFA, PUFA, MUFA, Protein etc. I’m starting with yogurt, ribeye, salmon, coconut oil, olive oil, sweet potatoes, bananas, strawberries, onions, mushrooms, brocolli, tomatoes, cucumbers and lettuce. Is any meal planning available somewhere?


  3. Hi Richard,

    It sounds like you’re reproducing a lot of the work in our book … makes me feel guilty for not having it out yet … your food list looks great though!

    The book gives tables of food properties, advice for constructing meal plans, and pictures of food and food preparation methods that might be unfamiliar to many Americans. We are going to do a separate free e-book with meal pictures, to give more ideas for meal planning. That should be available soon after the book is out.

    A lot of food pictures will go up on the blog too.

    The basic prescription we give in the book is to combine carbs or protein with fat in a 1:2 calorie ratio. Examples:
    – rice with egg yolks and coconut milk (e.g. in Bibimbap)
    – sweet potatoes mashed with butter and coconut oil
    – taro with cream cheese topping
    – fruit or berries with cream or ice cream as a dessert

    Fatty/oily meats like ribeye and salmon naturally come with the 1:2 calorie ratio. Lean meats like chicken breast should be combined with oil or fatty foods like avocado.

    Carb-protein ratio is optimized with ~3 times more plant mass than meat mass. We eat about 1.5 lb plants, 0.5 lb meat per day, plus oils.

    Include vegetables to taste, don’t count them as calorie sources.

    Mix a starch-fat combination with a meat-and-vegetable combination and you have a meal. A fruit/berry – dairy or coconut fat combination is a good dessert.

    Most standard recipes can be made Perfect Health Diet recipes by ingredient substitution. E.g., use rice noodles for spaghetti or lasagna.

    Paleo cookbooks are good resources, although you need to add starch sources, oils, dairy fats, and fatty cuts of meat, as they tend to follow Cordain’s prejudices!

    Best, Paul

  4. I should add, what Richard is asking for is mostly content that was lacking in the early draft that is now for sale, and that we have been adding into the final revision. Since much of what we would have liked to add won’t fit, we’re going to put the excess in a free e-book.

  5. Paul,

    If you are taking requests for blog posts, expanding on the potential dangers of gluconeogenesis that you describe in your e-book could be a good idea. Just how could the byproduct ammonia be bad? How do you know if you have low mucus production and therefore dry eyes or dry mouth? Just what types and amounts of vitamins are used in gluconeogenesis and how should people on a low glucose diet respond by increasing those vitamins?

    Thanks, Jeremy

  6. Hi Jeremy,

    Yes, I do take requests for blog posts, and that’s a good one.

    I’m actually planning one along that line shortly, on why the Optimal Diet (a very low-carb diet) seems to have a higher risk of gastrointestinal cancers; I think it’s due to insufficient mucus production (the phenomenon of dry eyes / dry mouth, but in the intestine).

    That’s been held up because it’s a fairly research-intensive post and I have a couple of consulting projects to finish and the book to get out this week, plus two kids going to college. I may be able to get the post out week after next.

    After that I can discuss some of my own experiences when I went too low-carb.

    Biotin and B6 are both needed for gluconeogenesis, so the more glucose you need to manufacture, the more you’re hurt by a deficiency of these. Both are fairly common deficiencies, especially biotin since multivitamins don’t supply much.

    Best, Paul

  7. Thanks for the reply and I will anticipate the future post on the Optimal Diet and on the dangers of going too low carb.

    Above, you say you eat 1.5 lbs of plants and 0.5 lbs of meat a day. In your e-book, on page 2 you say 2 lbs of plants and 1 lb of meat. Maybe the difference between 1.5 and 2.0 lbs is not huge, but the difference between 0.5 and 1.0 lbs of meat is large.

  8. Yes, we were a little imprecise in the early version. In the final version we’re more clear.

    Our recommended protein intake is 200 to 600 calories per day. One pound of meat is about 500 calories of protein, so 0.5 to 1.0 lb is a good range for meat intake. We personally eat toward the low-protein side, so 0.5 lb meat. This is good for longevity and immunity against chronic bacterial infections, less good for bodybuilding.

    The recommended carb calorie intake of 400 +- 200 is roughly provided by 0.5 lb starches (if more calorie dense starches like rice & taro are included, 1.0 lb if solely from potatoes/sweet potatoes/yams) and 0.5 lb fruits and berries. That leaves vegetable intake, which we leave to your discretion. It’s pretty easy to eat a pound of vegetables per day if you like salads but we often eat about 0.5 lb, mostly cooked with our meat.

    Ideally, one might eat more sweet potatoes and less rice than we do, to get more fiber and nutrients, and more vegetables, and so eat 2+ lb of plant foods. But life is so busy, we often take the easy way out and eat rice and minimal vegetables. We pretty much average out to 1.5 lb plant foods, 0.5 lb meat, plus oils.

  9. By the way, our recommended protein and carb ranges are both 200 to 600 calories, but it’s important to eat at least 600 calories of the two combined. So if you eat at the low end of one range, you have to eat toward the upper end of the other.

    For a long time I ate a high-protein low-carb combination. But now I’m finding I favor a higher-carb lower-protein mix. My wife has always favored that.

    Thus, the 2 lb plants 1 lb meat is a pretty fair approximation to how I used to eat, but now I’m eating (a) less protein and (b) less total.

    The shift to less protein occurred when I realized that eating fewer carbs wasn’t reducing my average blood glucose level, whereas reducing protein would probably help my immunity and longevity.

    The shift to fewer total calories came about when I started intermittent fasting (I now eat only between 2 pm and 9 pm most days) and drinking coffee, which has an appetite suppressing effect. This pattern of eating seemed to drive me toward a lower caloric intake.

  10. Wow Paul, thank you very much for the informative response on quinoa vs. rice. I really don’t have a strong preference for it (but have read in other places why it’s “healthier”). I think your reasoning behind prefering rice and other starches is spot on. I’ll finish the quinoa I have and then make my preference for white rice.


  11. I’m new to this site, so I apoligize if this has been asked before, but do the people in Asia eat white rice?

  12. I’m new to the site, and it looks very interesting. This is the first time I’ve heard of wheat being regarded as toxic!

    (I know people in the States may have gluten allergies, specifically gluten wheat allergies, but does that make it toxic?)

    I try to maintain a low calorie / low carb diet, but I don’t count fiber from whole wheat (insoluble) or from oats (soluble), against my carb quota (150g/day).

    I like the idea of eating more saturated fats whenever consuming carbs, hadn’t heard that, and it’s easy enough to incorporate!

  13. Hi snoop,

    The book has an extensive discussion of wheat toxicity. Gluten is dangerous for everyone, not just people in the States!

    Since fiber provides carb calories to gut bacteria only, you’re right not to count the fiber against your carb quota.

    Saturated fats are helpful when combined with fructose or alcohol, but they don’t do anything to detoxify wheat.

    Best, Paul

  14. Hi Paul,
    I will definitely be among the first few to grab a copy of this book.It just sounds so sane amidst all the madness around !
    Ofcourse health always come first, but of a lot of concern to me is also weight loss.I was a fat child who lost all my weight following the john yudkin lower carb diet.It would be of interest to you to know that I didnt touch wheat for 12 years and remained slim.Then got married,and started eating like my husband and ballooned and have stayed that way for many years.
    For the last month I have cut out 2 ingredients completely from my diet….wheat and sugar.I feel immensely better, but need to do more to shed the weight and balance my blood work….will your optimal eating plan help me lose as well or do I still need to go extremely low carb OR count calories

    • Hi jo, The book has a section on weight loss. Yes, the Perfect Health Diet is good for weight loss: there are other diets for losing weight faster, but I think ours is optimal for keeping it off over the long run. We explain what we think are seven keys to weight loss in the book. As far as low carb, most overweight people should aim for 400 carb calories from starches, but some with metabolic damage may benefit from a ketogenic diet with around 200 starch calories and lots of coconut oil. As far as counting calories, well, you should do that for 3-4 days until you have a good sense of portions. We recommend eating as a normal slender healthy person does. Since obese people burn extra calories carrying around their extra weight, this is a calorie deficit for the obese. This would be around 1800 calories for most woman, 2400 for most men. However, on the ketogenic diet, take more calories than that, you need a lot of coconut oil. Once your metabolism has healed, your weight should go low again as long as you stay off food toxins.

      Best, Paul

  15. Thanks. Will wait for the book and also waiting for the post on going too low carb.

  16. I also wonder if Asians are just not as prone to get fat physiologically. Not to say that their diet is not better for you, but just that their physiology may also contribute to their low fat levels.

    I also noticed that the Scandanavian countries are doing very well! Im sure they eat a lot of unprocessed foods.

  17. This goes along with the first commenter’s question. What’s the deal with millet? I’ve taken an interest in harvesting the foxtail that grows wild in my backyard, since it’s just a variety of millet. I can’t find much good data on it, though.

    • Hi Becky,

      Millet is a grain and so it is guilty until proven innocent. Grains in general are very rich in toxins, so odds are millet is no exception.

      That said, it is an ancestral variety which may not have undergone a lot of modification by breeding. Dr. Davis has presented some evidence that the biggest problem with wheat may be that it has twice been hybridized, expanding its genome from 14 chromosomes in its ancestral form (einkorn wheat) to 28 chromosomes (emmer wheat) through hybridization with goat grass, later to 42 chromosomes through another hybridization with Triticum grass to create modern wheat. This means wheat has toxins from three separate species, which may explain why it is the most toxic grain.

      Dr Davis has reported that some people who cannot tolerate modern wheat can tolerate einkorn: http://heartscanblog.blogspot.com/2010/06/in-search-of-wheat-einkorn-and-blood.html, http://heartscanblog.blogspot.com/2010/06/in-search-of-wheat-emmer.html.

      I recommend avoiding millet and other grains except rice, whose safety is well-attested. Since we only need about 400 carb calories a day, it’s not too difficult to confine one’s self to known-to-be-safe carbs.

      But, daredevils, drunkards, and other risk-lovers may find millet worth the risk!

  18. Hi Paul, I know you favor white rice over whole grain or brown, but I was wondering how you felt about Haiga rice. It is, basically, a Japanese rice that has the bran removed but the germ intact. We like it a lot.

    Here’s a link to a Science Daily article:


  19. Hi Maggy,

    In general the outer layers have more protein toxins than the inner layers, so Haiga rice would be intermediate in toxicity between white rice and brown rice. We do favor white rice, but are tolerant of brown rice for those who like the test, so eat Haiga rice if you like it, but we will stick with white rice.

    Re the Science Daily article, it doesn’t fill me with confidence that it contains a protein which behaves like a pharmaceutical drug. My bias is toward food which doesn’t act like a drug (even if it’s a “good” drug, i.e. effective against a disease and relatively free of side effects).

    Best, Paul

  20. Hi Paul, that makes a lot of sense. Our other favorite rice is Basmati, the white variety from Tilda. Since I’m the cook here at home, I will be leaning more towards Basmati than Haiga, unless a more neutral-tasting rice is called for in a particular dish.

    Thanks again for the response.


  21. Perfect Health Diet » Why We Get Fat: Food Toxins - pingback on January 20, 2011 at 7:45 pm
  22. November 9, 2011 « CrossFit Motiv8 - pingback on November 8, 2011 at 11:17 pm
  23. Lose a Belly, Gain a Life | Wheat Belly Reviewed - pingback on November 21, 2011 at 4:09 pm
  24. We come from an outer Sydney, Australia, suburb – we’re very interested in the studies on wheat and obsity, etc. Please notify us when your book is published so we can buy a copy and spread the word!

    Ver best wishes, Mark & Denzil Fallon

  25. Lars (in Norway)

    I just discovered wheat toxicity thanks to my wife to went to a week of healthy cleansing in Denmark.
    I have mild asthma that my doctor says is natural in older folks (I am over 65). Could this be related to wheat toxicity? I love the bread in Norway because it is so much healthier than what is in the U.S. Recently I bought a loaf of 85% spelt bread. How about that?
    What about sprouted grain bread like Ezekial 6:9? Does that fit the toxic profile. My wife says I should just get of anything with wheat, even non-gluten. Is she right?

  26. Hi Lars,

    Yes, I think your wife is right.

    Different breeds of wheat have different levels of toxins, sprouted grains have lower toxin levels, and fermentation does reduce toxin levels somewhat, so sourdough sprouted grain breads are less toxic than the general run of commercial breads. That said, even the best types of bread still have some toxins. If it is indeed bread that is producing your asthma, then it would be best to eliminate all types of wheat bread. Instead, eat gluten-free breads and noodles made from rice flour, potato starch, and tapioca starch.

    Best, Paul

  27. Nalliah Thayabharan

    Wheat raises blood sugar higher than most of the other foods. 4 slices of whole wheat bread raise blood sugar higher than 12 teaspoons of sugar. That’s a simple fact as per the table of glycemic index.

    Almost all wheat in USA is from a dwarf strain, which produces a far greater yield but has contributed to the current obesity epidemic.

  28. Rice is NOT “the only safe grain”….it contains loads of arsenic

    • Correction. Brown rice contains lots of arsenic — particularly if it was grown in the South, where former cotton fields were heavily treated with anti-pest formulas that contained arsenic. White rice contains significantly less arsenic as most of the arsenic is found in the bran.

      White rice from California contains relatively little arsenic.

  29. You don’t show the data for rice intake. Is there any correlation between rice intake and body weight?

    Could it be that larger people tend to eat more, of anything? Correlation is not necessarily causation.

  30. I’m trying to reconcile what Dr. Davis says in his “Wheatbelly” blog with what you’ve said here. While you both suggest staying away from wheat, he seems much more restrictive in terms of carbohydrate input. For example, rice is not on his list of substitutes, and it seems to be on yours. Too much information!

  31. Certainly interesting, but vegetable oil is the only oil used for cooking in Japan.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.