Another China study
I want to give you a few words of advice right up front. Keep this post close at hand so that you can send it out whenever anyone makes one of the following comments to you:
- The Chinese don’t follow low-carb diets and they’re healthy
- The Chinese eat a lot of carbs and they don’t get fat
- The Chinese follow a low-fat diet and they don’t get fat
- Fruits and vegetables don’t make you fat
- If vegetables really made you fat, the press would be all over it
- Researchers never misstate their findings
- Show me the study!
The current issue (June 2008) of the International Journal of Obesity, published by the Nature Publishing Group (Nature is the most prestigious scientific journal on the planet), contains an article titled Vegetable-rich food pattern is related to obesity in China that is most enlightening and interesting on many levels.
Researchers evaluating the obesity status of the Chinese discovered that
China is facing an obesity epidemic. Between 1992 and 2002, the prevalence of overweight and obesity increased in all gender and age groups and in all geographic areas of the country. Using the World Health Organization body mass index cutoff points, the combined prevalence of overweight and obesity increased from 14.6 to 21.8% during this period. [An increase of about 50%, which pretty much mirrors what has happened in the US, although from a lower starting point.]
What is even more telling is that the prevalence of central obesity, a much more representative measure of overall health, was as follows:
Central obesity was 19.5% in men and 38.2% in women.
In an effort to figure out what is going on researchers evaluated the diets of almost 3000 Chinese men and women (average age 47 yo) from all walks of life and all educational levels living in rural and urban areas. They found the typical Chinese diet to be
characterized by a high intake of vegetables and other plant foods, and thus the intake of carbohydrates and fiber is high.
When these researchers started evaluating the specifics of the Chinese diet, they found themselves ensnared in a thorny thicket of cognitive dissonance. How they hacked their way out shows just how nutritionally biased researchers are today and how they will twist and turn and torture the data (and outright misrepresent it) to make it confirm their biases.
Let’s take a look.
They report early on in the paper that despite the ongoing nutritional transition that the Chinese are going through (brought about by Western influences) that
The average intake of vegetables is still higher than those in the western countries. At a national level, the mean intake of vegetables was 276 g per day in 2002, which is 40 g per day less than 20 years ago.
They then state:
The beneficial association between intake of vegetables and fruits and obesity has been well documented in different populations. In Western populations, high intake of fruits and vegetables is associated with lower intake of energy and a healthy lifestyle and thus a lower risk of obesity. No research has examined the association between obesity and a diet rich in fruits and vegetables in China.
Which they set out to do. It is obvious from the first two sentences of the quote right above what the bias of these researchers is: fruits and vegetables are good for you. And more of them is even better. Problem is that there really isn’t any definitive research showing this, although it is widely believed.
I’m pretty certain that when these people decided to do this study their game plan was to divide the study population into quartiles or quintiles of fruit and vegetable consumption. And I’m sure they figured that those subjects in the quartile or quintile of the highest fruit and vegetable consumption would show the least obesity while those in the lowest fraction would show the most. But it didn’t turn out that way for them.
In fact, it was just the opposite. The researchers did divide the subjects into quartiles (fourths) as a function of fruit and vegetable consumption, and they found that the subjects eating the most fruits and vegetables were the fattest. As they put it:
Increased intake of the vegetable-rich pattern was found to be positively associated with risk of obesity in the study. the prevalence of obesity among women with the highest intake of this pattern was more than double than that of the lowest intake group.
The vegetable-rich food pattern was associated with higher risk of obesity/central obesity in Chinese adults in both genders.
So what do good scientists do when they discover that the data doesn’t conform to their preconceived notions? They change their thinking to fit the data. What do sorry scientists do when the data doesn’t conform to their preconceived notions? They interpret the data so that it does fit with their thinking. Which is just what these researchers did.
The data is pretty clear that those subjects who ate the most vegetables were the most obese. It’s also clear from the data that those subjects who ate the most vegetables also ate the most calories to the tune of about 200 kcal more per day. Now it would be easy and within what the data show to state that those subjects consuming the most calories were the most obese. No one would argue with that. But it flies in the face of the popular idea that you can eat all the calories you want as long as they are fruits and vegetables. Dean Ornish wrote a bestselling book titled Eat More Weigh Less on this very subject.
Our researchers are in a quandary. Their data that they expected to show one thing is not cooperating. And try though they might to legitimately present it, they can’t without somehow contradicting the idea that vegetables (in whatever quantity) are healthful. Where to go? What to do?
What do you think they did?
What all good mainstream researchers do. They blamed it on fat. I’m not kidding.
Here we go right from the paper. As Dave Barry would say: I’M NOT MAKING THIS UP.
The average intake of vegetable oil in China was 33 g per day in 2002, while this figure was only 22 g per day in 1992 and 18 g per day in 1982.
High intake of vegetable-rich food was associated with high energy intake, probably due to the high proportion of energy coming from fat which contributes to a high energy density, which is known to affect obesity prevalence.
In the present study, a positive association between intake of vegetables and vegetable oil was observed.
Stir-fry cooking method in the Chinese population may thus partly explain the positive association between a seemingly vegetable-rich food pattern and obesity…
Thus, when making recommendations on intake of vegetables, the importance of limiting the use of cooking oil as well as cooking method should be emphasized...
You can certainly see where they’re headed with this. And they don’t disappoint.
In conclusion, we found a positive association between intake of vegetable-rich food pattern and obesity. This association can be linked to the high intake of energy due to liberal use of vegetable oil for cooking vegetables.
There you have it. Fat is the culprit. And although vegetable oils have been the darling fats of the mainstream folks for ages, they don’t hesitate to throw them under the bus when their beloved vegetables are challenged.
But hold on. There is a fly in the ointment here.
There is absolutely no difference in the amount of fat consumed in the group that ate the most vegetables as compared to the group who ate the least. In men the grams of fat were exactly the same in both groups; in women their was a slight difference, but it didn’t reach statistical significance. And as a percent of energy intake (the measurement the mainstream loves), fat intake dropped from 32 percent to 31 percent in both males and females as vegetable intake went up.
You think I’m pulling your leg on this? Researchers publishing in a major journal (and the peers who reviewed them) wouldn’t say all this if it weren’t supported by the data. Right?
Well, here is the chart from the article showing all the data.
Click on it to blow it up to readable size and look down to where it lists the fat intake and come to your own conclusion. Then go back up to the quotes from the article that I have put in bold print. Unbelievable. This is why you never want to panic when you read the press report of a new piece of research. Oh, and BTW, have you noticed how the press has ignored this study?
I saw no mainstream press articles when I Googled; the one article I did see from an online publication picked up the misleading conclusion from the paper.
A Chinese study came up with the — to Americans — paradoxical finding that the more vegetables people ate, the fatter they were.
Why? Because the Chinese in this population in Jiangsu Province were stir-frying their vegetables in “generous” amounts of oil, and the more vegetables they ate, the more energy-dense oil they were eating.
If you’re going to draw any conclusions at all from this data in terms of macronutrient consumption and obesity, it’s got to be that carbs make you fat since the data show that carbs are the macrontrient that increased the most in both men and women with increasing obesity.
Since when are vegetable oils, wheat flour, and rice considered fruits or vegetables? Am I missing something here?
If I read the chart right, the thinnest quarter of people at twice as much rice.
The charts aren’t laid out by thin verses obese; they are laid out in terms of vegetable consumption. Those in the quartile with the least vegetable consumption did eat more rice.
I began increasing my fat intake. I cook with coconut oil, butter, peanut oil. I add extra butter to my vegetables. I also supplement my diet with a ketogenic cocktail of other oils: flaxseed oil, MCT and EPA, recommended by Dr. McCleary in his Brain Trust Program. I am losing weight and I have an outstanding blood lipid profile. So in my honest opinion, those Asians who gained weight because of the oil added to their food, just are not using enough oil.;-)
Another part of the study should inform us how often did these people eat? Meal frequencies also makes a difference as I have experienced from intermittent fasting. I am sure that there are some modern cultures who eat 1 to 2 meals a day instead of 3 meals plus in-between-snacks. In other words did all participants in this study eat identical meal frequencies or did they vary. I strongly believe that this would also make a difference. I guarantee you that if I were to eat 2,000 calories in a day and split those calories up between 3 meals and 2 snacks, I would gain weight. Whereas dividing these calories into 2 meals or even 1 meal…I would lose weight. Of course, every one of these calories would be low carb.
I saw a documentary on TV of how Sumu wrestlers build girth. This was several years ago, before carbs were a concern for me. All I recall from that documentary was that these wrestlers consumed tons of rice. Not scientific but it is an observation.
Frankly, the more I read about fats, the more I believe that fat is the most important nutrient we can consume worlwide.
I think there’s some misunderstanding of this study. First, they sorted the subjects into 4 “food patterns” and used regression analysis to show that the “vegetable rich” pattern (as a whole, not by quartiles) was associated with increased vegetable oil intake. They say “data not shown.”
The table then sorts the 4 groups into 4 quartiles. For all but the vegetable-rich group, they show only energy intake (this part of the graph doesn’t show on the link given here). So there’s no way to compare fat intake in the other groups with the fat intake in the vegetable-rich group except to take their word for it (“data not shown”). Very confusing.
But they do say at the end, “The fact that we found no differences in energy percentages from fat across quartiles of this pattern [the vegetable-rich group] may suggest that the total energy intake rather than the role of fat in explaining the association is the most important factor.”
I didn’t average the energy across all quartiles of all the 4 groups. I have other things that seem more pressing.
I personally find all studies like this murky when you group foods into categories based on preconceptions, like a “prudent diet” including fish and chicken and fruits and vegetables and an “unhealthy” diet containing red meat and hot dogs and fast food and luncheon meats and then conclude that red meat is unhealthy and chicken is healthy because the latter group (who were undoubtedly eating a lot of french fries with their hot dogs and fast food) didn’t do so well.
Hi Greta–
Good points. I hate the idea of these so-called ‘factor-analysis’ studies myself. I didn’t post on this one in an effort to show that eating more vegetables makes one fat, but to show how the authors misrepresented their data. If you go through the study carefully – as you obviously have – then you find all kinds of weasel phrases such as the one you mentioned where they do kind of represent their data accurately. But, overall they out and out misstate what the data show, and a casual reading or a reading of the abstract only gives one the wrong impression of what the study really says.
Cheers–
MRE
But I don’t think they *do* misstate what the data show. The vegetable-rich group apparently ate more fat than the other groups. There was no difference in animal fat intake between groups, so it must have been the vegetable oil that was increase.
It’s only *within* the vegetable-rich group that the fat intake is identical.
The problem is that they don’t show the data that show that the vegetable-rich group ate more fat than the other groups. We have to take their word for it. They say the regression coefficient was 1.40 (95% CI = 0.41, 2.40). I’m not a statistician. I know that for relative risks, if the CI overlaps 1.0, it means the results aren’t significant, as they could be 1.0. I don’t know if this is also true for regression coefficients.
Hey Greta–
Sorry it took me awhile to get this comment up, but I wanted to take the time to review a hard copy of the study to see what you were talking about. As I understand it, the researchers were looking at subjects in what they determined was the vegetable-rich food pattern and not comparing between the vegetable-rich food pattern and other patterns. Within the vegetable-rich food pattern the men showed no difference in vegetable oil consumption among the quartiles whereas the women did show an increase with increasing energy and vegetable (and fruit) intake. But you are absolutely correct in stating that the relative risk isn’t significant because the CI overlaps one. Which is the point I was trying to make. The authors are trying to pin the blame on vegetable fat when statistically there is no basis for it. Maybe you read it differently, but that’s my impression.
Cheers–
MRE
Have you seen this WHO map for cardiovascular disease? Looks like China is equal with US in that department. What do you make of this? Canada looks like they are in better shape than the US. Is their diet or other risk factors that much different?
http://www.who.int/cardiovascular_diseases/en/cvd_atlas_14_deathHD.pdf
Maybe Canada has less heart disease because Protein Power was a huge bestseller up there.
I spent a month in Asia last Christmas. The majority of Asians are very slim. Even those with a bit of weight on them are at most chubby by our standards. During my entire trip I only saw three people I would classify as obese.
They seem to eat a lot of vegetables and rice so I really don’t know how they stay so slim. The only possibility I can come up with is that their portions are small. However portion size is driven by appetite which is driven by insulin. So their insulin levels must be pretty well controlled IMHO. Hence they stay thin.
“They seem to eat a lot of vegetables and rice so I really don’t know how they stay so slim. The only possibility I can come up with is that their portions are small.”
I can think of another possibility. Perhaps a carb heavy diet based on whole grains, fruits, and veggies isn’t as fattening as some people believe.
There’s those “whole grains” again. Does anybody actually eat “whole grains”? I’m pretty sure that nearly all whole grain food (and almost certainly those wheat-based foods consumed by the most obese quartile) are made via industrial processing of grains, i.e. they don’t resemble anything like “whole grain” by the time they’re going in people’s mouths. Rice is one of the major exceptions, and rice does contain less of the anti-nutrients found in grains.
I’m still waiting for someone to explain just what precisely is healthy about “healthy whole grains”.
Hey Dave–
If you want to know everything there is to know about whole grains, go to this site and pull down the pdf from #7 on the list. It’s Loren Cordain’s long, comprehensive paper on cereal grains. Well worth reading.
Best–
M
Lol! Let me utter a quick laugh at the idea that Chinese people are by default healthy.
I AM Chinese. Now I know this is just anecdotal, but here’s my family’s medical history: My beloved maternal grandmother had heart disease her whole life and eventually died of it. My paternal grandmother has type II diabetes; her second son, my uncle, died of a blood clot in his forties. My mother’s side has heart problems, and my father’s side has blood problems. On top of all this, at only 21 years of age, I have a nasty apple shape — not the most becoming on a woman. I’ve had this shape all my life, despite being a competitive swimmer and eating a traditional Chinese diet at home.
I’ve visited China several times and it is true that the people look slimmer. However, this is in large part because their frames are much smaller. If you look at the middle age, they’re nontrivially thick around the middle. Also (again anecdotal) it seems to me that heart trouble is a big deal there in the elderly. My hypochondriac grandfather is always taking some preventative or another for his heart.
As for me, there is hope! I started Protein Power two months ago and am feeling well for the first time in a decade. Low-carb is definitely the way to go!
Joyce
P.S. My father’s family is originally from Shandong. From what I understand, it’s the “Italy” of China, consuming a lot of wheat flour — dumplings, noodles, etc. I wonder whether they have more problems there or not.
Thanks for providing us all with the interesting nutritional and disease history. Most people, it seems, appear to think that Asians have no health problems whatsoever. And are all thin. Thanks for setting the record straight, at least as it concerns those whom you’ve observed.
By the way, another anecdote:
When my parents reminisce about the poverty of the Cultural Revolution, they often talk about the scarcity of meat. They had little to eat but rice and vegetables and they said they were never full. My mother enviously remembers a wealthy family who fed their two children a whole chicken every morning. She remembers that the children grew very robust and healthy.
From what I know of my mother’s diet while pregnant, I’m sure I was born with a ridiculous amount of insulin resistance. Moral I’m taking away from this: Eat meat!
Joyce
I think you may want to be careful about the implication that it’s “vegetables” that are the problem when the study category is “vegetables and fruit”. The leafy green vegetables Chinese eat have very few calories, and little dietetic effect except on blood pH. The fruit, on the other hand, has plenty of carbohydrates, causing the associated insulin spikes and hunger.
Joyce: yes, Shandong is in the north; very wheat intensive. My mother is from Canton where they eat rice instead; no heart attacks that I know of on that side of the family, though I think I had a degree of insulin resistance until I went on a low carb diet.
From what I know of my mother’s diet while pregnant, I’m sure I was born with a ridiculous amount of insulin resistance. Moral I’m taking away from this: Eat meat!
[...] the researchers, who hummed and hawed their way to a half-baked conclusion—check out this post by Michael Eades). And as Stephan Guyenet explained, the study really showed a trend between wheat [...]
Joyce is correct. So what if some Asians in Asia are skinny. What makes you think they are healthy?
I have Chinese friends who are thin but have health problems. One of my Chinese co-workers is having an operation for a tumor – and she’s one of the thinnest people at work!
My mother is Chinese and she developed heart disease after moving to the city and began eating more Western foods – vegetable oil, wheat, sugar, processed foods. She’s been skinny her whole life, but her heart wasn’t healthy.
However, my great grandmother was a centenarian! She ate loads of lard and meat. Yes traditional Chinese cooking uses lard not vegie oil. She was thin till the day she died (at age 100) and was walking around doing stuff before she passed away.
You have two women in the same family who are both skinny, one who lived on rice, lard, meat, eggs and fish – the other on wheat/rice, vegetable oil, little meat, veggies and some tofu. Who lived to 100, disease-free, with natural death? My great grandmother who ate a mostly paleo diet.
My grandmother is still alive =) she’s pretty healthy and she also eats similar to my great grandmother. Although it pains her to remember the Cultural Revolution where there was little meat to be had – but she made up for it in the 1980s.
My mother is returning to her roots and eating like my great grandmother did. It’s funny because I’m teaching HER how to eat traditionally when it’s usually the other way around. It’s hard, because she still thinks saturated fat is bad and she’s addicted to wheat products (bread, pastry, cookies, biscuits). She’s starting to refrain from buying that stuff.
Excellent post. Thanks Dr. Mike.
Also interesting in the data table was that the group most prone to obesity got far more of their fat from veggie sources and veggie oil (PUFA’s & n-6 galore), and much less of their fat from animal sources.
[...] 4. http://www.proteinpower.com/drmike/obesity/another-china-study/ [...]
[...] http://www.proteinpower.com/drmike/obesity/another-china-study/ This entry was posted in Health and tagged china study by Rick. Bookmark the permalink. [...]