If you don't remember, Dr. Yoni Freedhoff and Dr. Kevin Hall discussed Hall's poster at the International Congress of Obesity (it's a longish video, click here for a transcript). Among the wonky nutrition set, it became something of a controversy. Essentially, Hall stated that this study would invalidate the "Insulin Hypothesis" of obesity.
"But as Bazinet points out, 'The study ... doesn’t see any [relationship between a decrease in insulin and an increase in fat loss]. Show me a better study that supports this.'
There isn’t any, he added."
What has been the reaction so far? Below is a round up of news & blog coverage of the study. Send me your (science based) thoughts/blogs/or other coverage you see and I can add them.
Read the study and see what you think - send questions and comments OR share them on FB.
One question that's come up a lot, is diet sequence. While all the study participants were on a "standard" diet and a "low-carb" diet, they ALL did the diets in the same sequence. First, they were on a standard diet and then they did LC.
On the Nutrition Wonk FB page, one reader mused: "Not sure why they didn't use a crossover design, so half the group had the KD (ketogenic diet) before the HC (high carb) diet."
I asked Kevin Hall why this was the case and here was his response:
"The study was designed such that the baseline run-in diet was intended to match the typical composition of the subjects' habitual diet. We screened out people whose habitual diet was too different in composition.
Therefore, randomizing the diet order would introduce a significant order effect that could confound the interpretation of the data given that the ketogenic diet would represent an extreme change from their habitual diet upon entry to the study.
It would have been nice, albeit likely boring, to have a group consume the baseline diet the entire 2 months since this is probably a better way to properly control this kind of study. However, we wanted to get the pilot study completed ASAP to estimate whether the primary energy expenditure effect size was large enough to warrant a follow-up on this endpoint in a more diverse group with more realistic diets.
(The reason this was called a pilot study was because it was conducted using an extreme diet in a relatively homogeneous group of subjects.) As you can see, the effect size was small and the investigators determined that it was not worth following up with a larger, more diverse group eating a less extreme diet."
Despite fascinating conclusions, the unanticipated increase in calorie burn (~500 kcals per day) when participants were NOT in the metabolic chamber greatly surpassed any difference that could be seen from macronutrient adjustment.
Over at CaloriesProper, Bill questions what differences we would have seen if half the group had been put on the ketogenic diet first and posits that the continued fat loss at the same rate in the second month (as there was in the first month on the higher carb diet) is, in fact, evidence of an advantage of low-carb. After all, wouldn't you see a profound slowing of fat loss in the second month of dieting as participants' metabolisms slowed in response?
He compares the experiment to the classic Minnesota starvation studies (figure below) where you can see that fat loss slows over time.
Dr. Ludwig disagrees with the interpretation of Hall's paper. In his words, "This final version continues to downplay remarkable findings for a limited pilot: a significant increase in metabolism on a very-low-carbohydrate diet."
In his opinion, the small increase in energy expenditure experienced by the participants (which was statisticaly significant) shows that there surely IS a metabolic advantage to a hugely low carbohydrate diet even though study authors dismiss this as unlikely.
So, are the carbs back in our good graces? Or are they better used as war paint?