Gary Taubes responds to Gina Kolata’s review of Good Calories, Bad Calories. The scientific article by Leibel et al. that Taubes and Kolata refer to.
Thanks to Dave Lull.
Gary Taubes responds to Gina Kolata’s review of Good Calories, Bad Calories. The scientific article by Leibel et al. that Taubes and Kolata refer to.
Thanks to Dave Lull.
To me, Taubes missed another major flaw with citing Hirsch and Leibel — the study’s minimum calorie proportion of carbs was 15%, and all carbs were from celerose, which appears to be a form of glucose. So, if my math is right (~9500kJ/d total * .15 carbs / 16.7kJ/g = 85g carbs/d), even the “low carb” group was getting 3 ounces of straight glucose a day. I don’t know the biology, but that seems like a lot of glucose; perhaps enough that higher doses would have no effect on the subjects.
I also find the Leibel et al. study of limited relevance because patients appear to have not been allowed to regulate their own food intake. Part of the reason low carb diets might work is that people who are sensitive to insulin are less hungry on them. There are several ways in which “a calorie is a calorie” as it is popularly understood might not hold.
I also think Taubes understands and believes in the set point concept — he cites the work of Sims and other related to set points — but he avoids “set point” because it’s become a loaded term. Cabanac is right that there’s widespread confusion between a set point and a reference, and, when it comes to obesity, Kolata and others have popularized the idea that set points are fixed and/or uncontrollable. Since Taubes’ whole thesis is that permanent weight loss is possible, he implicitly agrees that set points are adjustable.
What struck me when reading the book, particularly the description of how insulin drives hunger, is that SLD, at least in its oil form, is something of a stealth low-carb diet. The two-hour window around the oil essentially makes it a meal with a minimal insulin response, so one is guaranteed to be in a low-insulin state (no hunger, no carb cravings, etc.) once or twice a day. I also suspect that nearly all SLDers eat fewer carbs than they did before starting the oil. I have no idea how much of an effect this has, but I think it’s worth considering that carb restriction might play a role in SLD.
James, it turns out that Taubes does discuss the set point concept (pp. 427-428). He doesn’t believe it (”the concept of a set point or a lipostat has little precedent in physiology”).
When you do SLD by drinking sugar water, it is obviously not stealth low-carb. That the sugar water and flavorless oils produce roughly the same effect per calorie implies that carbohydrate reduction has nothing to do with the effect.
Professor Roberts, you’re absolutely right, I’m confusing set-point-like local equalibria (which Taubes buys into) and a centrally controlled set point for adiposity (which I had forgotten was part of the SLD book).
FWIW, my own self-experimentation has shown that sugar water is not as effective for me as oil when it comes to appetite suppression. (I really wanted sugar water to work, too — it would be much easier when travelling.) In fact, my first attempt with SLD was with sugar, and I had little AS in two weeks; a few months later, I tried the same number of oil calories, and started feeling the effects within days. To put this in context, I started out quite heavy (365 lbs.) and I stopped eating refined carbs after starting oil-based SLD (it became very easy to do so), so I may be atypical.
I read Taubes’ book awhile back and remember wondering about its relation to SLD. If I remember correctly, he dismisses the set-point theory, but goes on in the next few pages to discuss research linking an initial surge in insulin to the taste of food. I thought this actually dovetailed nicely with SLD, with low-flavor food diminishing the response/association. So maybe SLD isn’t stealth low-carb, but is stealth low-insulin response.
Admittedly, the sugar water is somewhat problematic for this hypothesis, but perhaps Dr. Roberts is right in thinking that sugar water that doesn’t have other flavor is treated differently than sugar water with added flavor. In this case, the difference would be insulin response.