Review of a Critique, Part 2

(Continued)
Mr. BS proposes four predictions he derives from his understanding of the CH.  Naturally, if a prediction follows from a hypothesis, and that prediction can be falsified by known data, the hypothesis may be considered falsified.  For example, if I hypothesize that the sun comes up because I wake up, I could predict that if I did not wake up, the sun would not rise.  I can then falsify my hypothesis by noting that the sun has risen many times when I slept.  However, this kind of science is only as good as the participant’s understanding of the hypothesis in question.  Since Mr. BS appears not to understand the CH, then it should come as no surprise that his suggested test predictions are inadequate tests of the CH (a more objective test of a hypothesis is of course an intervention study, and such is proposed for the CH in some detail in GCBC on page 466 of the paperback version).

Having established these straw man predictions, based on an incomplete or incorrect understanding of the CH, Mr. BS proceeds to knock down all four.  The question is whether his test predictions are valid – do they really test the CH?   

Mr. BS’s first test prediction: if the CH is true, then the following should be observed:  “An inverse relationship between carbohydrate intake and body fat loss during food restriction, independent of energy intake.”

His first test prediction would be more accurately stated as follows: 

“Carbohydrate intake resulting in chronically high blood sugars and insulin levels should result in increased body fat accumulation when compared with an isocaloric diet composed of a macronutrient skew which does not generate chronically high blood sugars and insulin levels.”  Interestingly, Taubes suggests in one article that he thinks 100g/day or less is a ballpark number for a safe amount of daily carbohydrate consumption.  The uncited study (or studies) that Mr. BS says he refers to in his prediction falsification effort supports Taubes:  “I only observed an independent effect of carbohydrate intake at very low (italics mine) carbohydrate levels (<100g/day).”

Nutritionists on the whole seem to think that the minimum carbohydrate intake per day should be a level that I would consider "high" - 200 grams or more in order to ensure the brain gets enough glucose from ingested carbs, and to ensure that adequate calories can be ingested without overconsumption of protein (typical conjecture is that “too much is bad for the kidneys”) or fat (in line with the now conventional wisdom that “too much causes obesity and CVD”).  Those who use carbohydrate restriction with clients, however, would not call 100g/day a “very low intake level.”  Folks like the Drs. Eades or Dr. Atkins target 50-100g/day as normal carbohydrate intake.  In short, Mr. BS’s bias as a traditionally trained nutritionist is on display in his analysis.

Test prediction number 2 is, if the CH is true, then one should observe:

“A positive association between insulin levels and body weight gain.”

Mr. BS’s second test prediction is close to a Taubes hypothesis:  “8. Insulin is the primary regulator of fat storage. When insulin levels are elevated - either chronically or after a meal – we accumulate fat in our fat tissue. When insulin levels fall, we release fat from our fat tissue and use it for fuel.”  Mr. BS’s prediction refutation is invalid because he does not cite studies, and without being able to examine the studies he uses to refute the hypotheses prediction, we cannot know if they say what he interprets them to say.  Testing this hypothesis will take a carefully designed study.  If such a study has been done, I wish Mr. BS had cited it so I could examine it.

Mr. BS’s third prediction is, if the CH is true, then it should be true that “Fructose, a carbohydrate that generates very little change in blood sugar and very little insulin secretion, should result in less fat gain than glucose."  Mr. BS does not clarify what he means by fructose, nor does he cite the studies he claims falsify his (invalid) prediction.  Most likely he means high fructose corn syrup.  More importantly, Mr. BS’s third test prediction (as shown above) isn’t valid.  A valid prediction based on the CH would be:  “Fructose accelerates the fat storage signaling of excessive carbohydrate intake by increasing insulin resistance, which reduces the ability of non-adipose tissue to respond to insulin signals, therefore requiring that adipose tissue take up a greater portion of excess blood sugars.”  This prediction is validated to at least some degree by the non-cited study result Mr. BS refers to.

Tomorrow, we'll look at the fourth test prediction, and a few additional items covered in Mr. BS's post.