Monday, July 7, 2008

Dissonance Redux

Michael Eades' most recent blog gives an outstanding example of cognitive dissonance in action. In short, a recent study found that obesity in China was correlated with higher intake of vegetable foods. But "everybody knows" that vegetables are healthy, right? So the authors conclude it can't be the vegetables causing obesity, it must be the vegetable oil the Chinese cook them in. The funny (but simultaneously sad) thing is that their own published data not only refutes this hypothesis, but also clearly supports the alternative hypothesis that refined carbohydrate consumption leads to hormonal imbalances which in turn lead to obesity. It's right there on the page. Talk about cognitive dissonance.

Dr. Eades did an excellent job hitting the big points, so I just want to add a few thoughts. I can't get access to the original paper (I have no interest in handing Nature any of my money for this paper, especially given that someone was clearly asleep at the wheel to let it past peer-review), but I would infer that it is an epidemiological study. So we can't give it a lot of weight, any more than we should give much weight to the mostly epidemiological evidence that support current mainstream dietary recommendations. That said, the authors' conclusion is clearly goofy, and ignores considerable information. The differences in fat consumption between less and more obese groups was fairly minimal: essentially zero in men, and about 6g in women. The statistical measure of the "trend" (indicating correlation between quantities) indicates lack of signficance in correlating fat intake with obesity for both men and women. But here's a list of items whose trend was indicated at greater than 99.9% confidence (the "+" and "-" indicate positive or negative correlation):
  • Fresh vegetables (+)
  • Fruits (+)
  • Rice (-)
  • Wheat flour (+)
  • Whole grain (+)
  • Root vegetables (+)
  • Pickled vegetables (+)
  • Fish (+)
  • Milk (+)
  • Eggs (+)
  • Calories (+, women only)
  • Protein (+)
  • Carbohydrate (+)
  • Plant food fats (+)
  • Animal food fats (-)
  • Vegetable oil (+, women only)
  • Physical activity (+)
Thus, people who ate more vegetables were more obese, people who ate more fat from animal sources were less obese, etc. Indeed, if you look at the actual amounts of food consumed, what we find between the least and most obese quartiles are major increases in wheat flour (5x men, 20x women), whole grain (18x men, 22x women), root vegetables (4x men, 4.7x women), and milk (75x men, 92x women). Physical activity in the most obese quartile was somewhat less than two times greater that in the least obese. In other words, those people following a diet and lifestyle (at least given the presented data) similar to that recommended by the USDA were the most obese, on average. The focus on vegetable consumption is actually a little weird, being only about 50% higher in the most obese quartile, compared to the least obese, considerably less variation than than that for wheat etc. I guess it's more dissonance: "everyone knows" that wheat, whole grains, etc. are healthy, so why even go there?

The ultimate scientific test of a theory is its predictive power. The theory underlying USDA recommendations basically says if you eat like the food pyramid and exercise more, you should be at lower risk of obesity. When confronted with data which contradicts the theoretical predictions, you have two choices: question the data, or question the theory. These guys did neither, instead waving their hands and adding an additional hypothesis which still failed the predictive test for one study group (men), but which I guess let them sleep at night.

Let's now consider an alternative hypothesis: that most obesity is a symptom of an underlying hormonal imbalance caused by overconsumption of refined carbohydrates. This theory predicts precisely the results seen, without any bogus ad hoc additions. Carbohydrates drive insulin drive fat storage. For that matter, milk proteins may also have a larger effect on insulin than other proteins, and of course milk does add to the carbohydrate load. There's even evidence supporting Gary Taubes' hypothesis on the connection between caloric intake and physical activity: those with higher caloric intake were, on average, more physically active. There's no evidence of causality: it may be that eating more calories increases activity, or that those with increased activity get hungrier, or both. While the weight supplied by this study is rather thin, it does at least provide further confirmation of what we would expect given current knowledge of metabolic regulation, which is considerably better than the nonsensical conclusions put forth by the authors.


Anonymous said...

Stephan, the author of the blog Whole Health Source has an interesting view of this study.

From my admittedly cursory reading, it appears that carbs in general are not the cause of metabolic syndrome. Rather, the cause is due to various proteins and lectins, combined with a high intake of carbs. The proteins (such as wheat agglutin) and lectins cause problems with nutrient absorption, appetite control, etc., which when combined with high insulin levels due to carb consumption seem to cause metabolic syndrome.

There are some cultures, such as the Kitavans, who consume a high carb diet yet don't acquire metabolic syndrome.

Another potential complication is the role of omega-3/omega-6 ratios and serum vitamin D levels.

It appears that many facets of the modern Western diet contribute to metabolic derangement.


Dave said...

Hi Ptolemais. Excellent points. Metabolic syndrome presumably reflects a hormonal imbalance. But the disease doesn't care how that imbalance is caused. So it may very well be that wheat brings other aspects to the party which tend to drive this imbalance more than, say, rice. Similarly, the Hyperlipid discussion of the Kitavans notes that they are likely calorie restricted, so though the percentage of carbs they eat is high, the actual amount may be somewhat lower. Those carbs further may be both rather lower glycemic and nutrient dense. Yams, for example, definitely bring fair quantities of micronutrients to the party, and wild yams have not been cultivated to increase both the quantity and availability of carbohydrate.

I think the idea that "too many carbs are bad" is another oversimplification, similar to "too many calories are bad". It's just that in Western culture, most of our carbs come in forms that are nutrient deficient (or effectively so from a bioavailability standpoint) and have extreme effects on metabolic hormones: refined wheat flour, sugar, high-fructose corn syrup, etc. So a "low carb diet" effectively means trading these items out for more nutrient-dense lower glycemic items, because that's what we have available. Similarly, once the "bad" foods have induced a disease state, the body is probably far more sensitive to certain inputs. I wonder how a Type II diabetic would do on the Kitavan diet? It would be interesting to find out.

Anonymous said...

Ptolemaios noted, " appears that carbs in general are not the cause of metabolic syndrome."

To me, it appears that researchers are finally beginning to home in on the major nutrient responsible for metabolic syndrome, fructose. It's been a long time in coming. It may also take a long time to replace saturated fat as the most demonized nutrient in the history of nutritional investigation.

Have you read this Nutrition & Metabolism article yet?
"Fructose, insulin resistance, and metabolic dyslipidemia"

Dr. Peter Havel seems to be at the forefront of fructose research. He's mentioned frequently in articles about fructose. A few examples:

I just HAVE to include a quote from a July 3 article published in our local paper. The headline read, "At age 106, Bigfork angler enjoys day of lake fishing." The quote at the end of the article read, "As she fished from the dock slowly retrieving a lure and hoping for a strike she passed along her secret for a long life. 'I eat bacon every morning,' she said. 'Crisp bacon.'"

Dave said...

Hi David. Thanks for the links, I'll check them out.

My personal semi-informed guess is that we're seeing the convergence of several dietary factors resulting in metabolic syndrome (and maybe a lot of other things like increased cancer). Human health pretty clearly started going downhill once grains were domesticated and began forming the bulk of the diet. While technology has helped offset some of the problems caused by this dietary shift, it also has made the problem worse by allowing us to concentrate things like fructose. The consequences to health of this combined assault appear nothing short of disastrous.

Speaking of fructose: I imagine many of you have already see this, but I absolutely loved this excellent post on fructose and glycation of LDL at Hyperlipid.

Personally, I like my bacon "floppy", along with 5 or 6 fried eggs.

Dave said...

The fructose article mentioned by David Brown is outstanding, and recommended reading. It is fairly technical and detailed. Here is a direct link:

Whenever I read some of the stuff referenced in the comments above, such as the potentially harmful substances in grains and detrimental metabolic effects of fructose, it amazes me that humans don't wind up in the hospital after a burger, fries, and regular soda. The body is wondrously robust, and can handle a lot of punishment before health starts to noticeably decline.

The fructose article contains a few amusing instances of cognitive dissonance. Here's my favorite:

"Certainly, diets high in saturated fats have been shown to induce weight gain, insulin resistance, and hyperlipidemia in humans and animals [19-22,31], but the emphasis on fat reductions has had no significant benefits relative to the obesity epidemic."

The references given relative to humans are, of course, the Seven Countries Study, and one obscure review for which I am unable to find anything but an abstract online. As you read this article, you'll find detailed biochemical evidence that fructose induces precisely those effects which, for over 50 years, have been confidently assigned to saturated fat. Yet even these authors, deeply immersed in this information, continue to hang on to the saturated fat hypothesis.

Anonymous said...

I would suggest that the main reason excess fructose is a problem is DNL to palmitic acid -- a saturated fat.


Dave said...

Hi Paul. Why is palmitic acid detrimental to health?

Stephan Guyenet said...

Hi Dave,

Just discovered your blog, looks good! I wanted to make a couple of points about the carbohydrate thing. As you know from previous comments referring to my blog, I'm of the opinion that all carbs are not created equal.

In the Chinese epi study, carb intake only increased slightly with obesity, and it may simply have been balancing a slightly elevated energy expenditure (the fat ones exercised a little more). The largest change was the replacement of one refined carb with another, i.e. rice with wheat. As you go through the quartiles, rice goes from practically the exclusive carb to less than 50% of carb intake. Why they didn't call it the "wheat diet" rather than the "veg diet" blows my mind. I discussed the study a while back on my blog.

It is just an epi study, so I'm not going to bet the farm on it. However, I believe that studies of societies like this one that are in the middle of a nutrition transition are the best place to do epi studies because there are really large differences between groups. The same type of study in the US can only look at whether you get your wheat from crackers or rolls. So I do think it's more telling than studies like Framingham.

About the Kitavans, one of the things Dr. Lindeberg mentions repeatedly is that they have abundant food, and scarcity is almost unknown. Their caloric intake is very similar to modern Americans, and Kitavans are only moderately active on average. They eat no grains whatsoever (starch is from tubers), although they do get fructose from the 920 kJ of fruit per day that they eat. Furthermore, their close genetic relatives on nearby islands are extremely susceptible to obesity, diabetes and the rest of it when they switch to processed food including wheat and refined sugar. It's really hard for me to see any explanation other than that it's the type of carb that matters, rather than the quantity. I'm going to post on the Kitavans in the next few days.

So basically, that's part of the reason why I think wheat is a big culprit. I have some mechanisms in mind; I'd be happy to tell you about them if you're interested.

Now there's another layer of complexity to my hypothesis that explains some of the American data. I believe that once you've blown your pancreas (and hormones) to smithereens, you may have made yourself sensitive to all carb, not just wheat. That's why low-carb is so effective for unhealthy Americans. Also because low-carb is low-grains.

Sorry for the super-long comment.

Dave said...

Hi Stephan. Please feel free to make your comments as long as you'd like! It's good stuff.

As I learn more about the details of metabolic regulation, I'm increasingly inclined to agree with some of your major points. "All carbs are bad" is probably a serious oversimiplification. Fructose, for example appears significantly worse than glucose in terms of how it bypasses metabolism. On the other hand, we almost certainly evolved eating a certain level of fructose in fruit. It would seem that evolution did design us to consume some amount of fructose, delivered in a particular package, which may, for example, limit the rate of absorption compared to a soda, or even fruit juice.

Anecdotally, I'm also inclined to agree with your assessment about pan-carbohydrate sensitivity. It's certainly been my experience. Since going low-carb, I really struggle with even a moderate amount of a high-glycemic food. I have a hard time believing that I had such a bad reaction when I was 18.

I'd be very interested in hearing your thoughts on wheat and metabolic syndrome.

Stephan Guyenet said...

Hi Dave,

I'm going to do a little brain-spew here, with the intention of putting some ideas on the table for discussion. One of the major logical steps in the idea that "carbs drive insulin drives fat" is that eating carbs causes elevated insulin. It's true on an acute basis, but what is not clear is that carb consumption per se causes chronically elevated insulin, which is the real problem. For example, the Kitavans have normal fasting insulin levels (considerably lower than Swedes, who eat less carb).

I think insulin spikes are probably not harmful as long as they successfully keep glucose in the right range. As I'm sure you know, there can be a big difference in the responses of biological systems to pulsed signals vs chronic ones. If your insulin is only elevated for 6 out of 24 hours (3 meals x 2 hrs each), that leaves plenty of time for fat to escape your adipose if basal insulin levels are normal.

For me, the most relevant question is, what causes chronically elevated insulin? I believe the answer is more complex than just carbs, and perhaps more complex than refined carbs even. My best hypothesis is that it's a long-term effect of some combination of wheat and sugar. I don't think sugar alone is capable of it, based on my reading about the Kuna. I don't know of any population that eats wheat flour and not sugar so I don't know if wheat alone is enough.

I have a couple of mechanisms in mind. All are nothing more than informed speculation. The first is that wheat in particular interferes with the absorption of fat-soluble vitamins. Peter has posted a number of good pieces on the damaging effects of gluten to the small intestine. Celiac patients suffer from malabsorption of fat-soluble vitamins, including K2. There are hints that most people may be gluten-sensitive to a certain degree. Vitamin K2 (Price's "activator X") is most concentrated in the salivary glands, then the pancreas. It seems to be important for insulin release from the beta cells.

Another mechanism for wheat also involves gluten damage. The small intestine secretes dozens of peptide hormones and communicates with the brain via the vagus nerve, all of which affects satiety and nutrient disposal (including insulin release and sensitivity). Wheat could simply be throwing a wrench in the gears by inducing immune activation in the small intestine or by way of insulin-like mitogen activity (mediated by wheat germ agglutinin (WGA)) on the intestinal lining.

WGA also gets into the bloodstream, at least in some people. It has a high affinity for the leptin receptor in vitro, and blocks leptin binding to it. Leptin resistance causes met syndrome in mice.

Now for sugar. It could simply super-stimulate appetite control centers beyond their normal bounds. Or it could just be that dumping a truckload of fructose on the liver impairs its functions relating to insulin disposal, glucose disposal, or something else.

Let me know what you think. The hypothesis is always evolving, so I'm open to input.

Dave said...

Hi Stephan. Good stuff. I need to get ahold of Frayn (on vacation at the moment) to check a couple of things. However, I do think that a carbohydrate meal raises insulin for a fairly long time afterward. Frayn has some graphs on this. Anyway, if one eats every 4-6 hours, it may be that insulin is chronically elevated. Carb snacks between meals of course makes this worse.

Also, if I remember correctly, the pancreatic response to glucose is a "sigmoid", maxing out at high glucose levels. Again I need to check the book to see if this occurs at physiologically relevant glucose levels, but if it does, it implies that too much carbohydrate exceeds the pancreas' ability to keep up, so you may wind up with transient hyperglycemia and hyperinsulinemia.

Also, junk food is high in both carbs AND fat, something the Kitavans don't experience (right?). I discussed this a bit in the comments on Hyperlipid, but dietary fat peaks in the blood several hours after a meal, and increased plasma NEFA induces insulin resistance. Dumping carbs in the system at this point may exacerbate elevated blood glucose and insulin.

Anyway, I need to check this stuff, and whether the conditions described above lead to chronic metabolic syndrome is another question. My gut feeling is that the antinutrients in grains do indeed play a role, and add to that the lack of fat-soluble vitamins (including D) in the modern diet and the lack/reduced bioavailability of minerals, and maybe the observed widespread health problems on a Western diet should not be surprising.

Stephan Guyenet said...

Hi Dave,

The 2 hr figure was off the top of my head, let me know what you find in your text. Of course, it will depend on insulin sensitivity and the composition of the meal. Anyway, even if it's 4 hours per meal, that's still 12 hrs a day at basal insulin levels. I think the most important piece of evidence however is simply the empirical data from the Kitavans.

They do eat a low-fat diet, 21% of calories with 17% of it saturated. Very low in PUFA with a very low n-6:n-3 ratio.

Too much carb should only exceed the pancreas' ability to keep up if the person has poor insulin sensitivity. I believe people with good sensitivity and a well-functioning pancreas can eat a lot of carbohydrate without elevating their blood sugar outside the normal range.

Dave said...

Hi Stephan. A few interesting notes from Frayn's Metabolic Regulation.

Figure 6.1 shows plasma insulin over 24 hours, adapted from a Reaven paper. Insulin is above fasting baseline for about 2/3 of the time. Plasma glucose varies between about 5 and 6 mmol/l over the same period. Figure 6.4 shows insulin and glucose after an overnight fast and a meal of 96g carbohydrate and 33g fat. Glucose returns to baseline about 3h after the meal, while insulin takes about 6h.

A dose/response curve for glucose/insulin is shown in Figure 5.5. It varies quickly over the range 2.5 to 7.5 mmol/l glucose, flattening out pretty quickly on either side. So I would interpret this as meaning that if you were able to push your blood glucose that high, it would exceed the pancreas ability to keep up. This makes some sense, as I believe the pancreas pre-manufactures and stores insulin. If you raise blood glucose enough to exhaust the stores, you may wind up with elevated blood sugar for an extended period of time.

I suspect refined carbohydrates have the potential to screw things up. Insulin response probably evolved over a range of naturally occuring carbohydrates, with the highest bioavailability being in fruit. Similarly, digestive processes probably "assume" some range of carbohydrate concentration. I'm guessing refined carbohydrates throw off the regulatory system by making too much glucose available over a short time period. Add to this the insulin resistance induced by fat, and the Western diet seems a recipe for chronically elevated blood glucose and the damage that results from it. The Kitavans both eat whole food sources of carbohydrate and low fat, and may thus avoid this scenario.

That said, I'm not convinced that by itself chronically elevated glucose/insulin result in metabolic syndrome. Chronic insulin resistance seems to arise from signaling problems at the cellular level, and the signaling defect may not be the same for different tissue types. It would not surprise me if some components of grains disrupted insulin signaling. The evolutionary strategy of grains seems to be to limit bioavailability of contained nutrients to discourage predation (e.g. phytates binding minerals), and this scenario is certainly consistent.

Stephan Guyenet said...

Hi Dave,

Thanks for the information. I wonder if insulin would remain elevated for so long in a Kitavan? The "average" American or European already has some degree of insulin resistance. Even young Kitavans have a lower fasting insulin than young Swedes.

The body adjusts to the carbohydrate content of a diet. The amount of first phase insulin released depends in part on the carbohydrate content of the last few meals. So there can be adaptation to a high-carb diet that you might not see in a figure depicting an acute carb feeding.

Kitavans are eating a lot of carbohydrate, both as a percentage of calories and in absolute amounts. They're getting them from starchy tubers, which don't have a particularly low GI or GL. The GI of sweet potatoes is higher than that of pasta and about the same as white rice, for example. And they aren't slowing down digestion with a lot of fat.

The fact that they're invariably thin, free of vascular events, have low fasting insulin and live to an old age suggests to me that they aren't having big glucose spikes. Either that or glucose spikes aren't harmful. I favor the former explanation.

I like your point that grains seem to use strategies that limit nutrient availability. I never really put it together but it does make sense. Protease inhibitors, mineral chelators, lectins etc.

Dave said...

Hi Stephan. Sounds like what we need is a glucose control study in the Kitavan population. What would be cool is to do a comparative study with another population that eats little or no carbs, e.g. the Masai or traditional Inuit.

You make a good point as well about metabolic research, in that it is usually performed in Western populations. The researchers often presume this to represent a "baseline" for human metabolic conditions, which is a ridiculous assumption.

Stephan Guyenet said...

That experiment would be great! I'll e-mail Dr. Lindeberg and see if they ever did a glucose challenge or postprandial glucose. That information would be very useful for this discussion.

There is glucose challenge data for some other non-industrial populations. If I recall correctly, they tend to have good glucose control. I don't know if that would apply to carnivores like the Inuit though; I can't imagine it would.

Stephan Guyenet said...

Hi Dave,

Dr. Lindeberg says he didn't do a glucose challenge, monitor postprandial glucose or look at HbA1c. I guess that leaves us with our speculations.

Dave said...

Hi Stephan. I wonder if the tubers eaten by the Kitavans are comparable in GI to those in the West? I've also seen some claims that wild yams/sweet potatoes may help with insulin control. See e.g.

Unknown said...

people saying wheat is better than rice to reduce weight.why?any fact in that?and why?

Unknown said...

people says wheat is better than rice to reduce weight.why?any fact in that?and why?

Dave said...


I am not familiar with any science that wheat is better than rice for weight loss. I suspect they're both poor choices, but if I had to choose I'd go with rice due to the lower lectin content.