Unlike observational studies that can’t strictly control the behaviour of subjects, the new experiment was carried out in a metabolic ward. In other words, researchers could measure every kilojoule consumed and burned by subjects. The research followed 19 obese adults on either a high-carb/low-fat or a low-carb/high-fat diet during six days of living in the ward. After a “wash-out” period, the subjects followed the same procedure with the opposite diet for an additional six days. Both diets reduced daily kilojoule intake by about 3347 kilojoules/day per subject, and both provided subjects with the same number of kilojoules/day. Protein contributed the same amount, roughly 21 percent of total kilojoules, to both diets.
After six days, the high-carb group lost an average of 89 grams of fat a day, compared to 53 grams per day for the low-carb group. The low-carb group lost more body weight – 1.8 kilos versus 1.2 kilos – probably as a result of increased water loss at the beginning of a low-carb diet. But, “Fat loss is a more important goal than weight loss,” Hall wrote.
Importantly, the high-carb group slightly increased its total consumption of simple sugars during the six-day trial. Still, the group lost more weight than the low-carb and low-sugar group. If simple sugars are the trigger for fat gain, that wasn’t apparent in this study.
Rather, the low-carb/high-fat diet group lost less fat than the high-carb group. This remained true despite a large increase in fat burning among the low-carb subjects. It appears that the high-fat burn is insufficient to overcome the high-fat consumption. The authors used the word “remarkable” to describe the very small decline in fat burning among the low-fat eaters.The NIH research team measured a number of other outcomes among subjects. As expected, total cholesterol and HDL cholesterol dropped more among the low-fat dieters, while triglycerides dropped more in low-carb dieters.
Low-carb fans will likely criticise the new paper for the modestly high carbohydrate consumption of the low-carb group. Some critics would like to see subjects consuming fewer than 50 carb grams a day (rather than 140). Hall agreed that such very-low-carb diets would “result in fat losses comparable to low-fat diets.”
However, he said his mathematical model predicts only small differences in results with equal-kilojoule diets that include a wide range of fat and carbohydrate ratios. “The body acts to minimise such differences,” he said, “and the total kilojoules in the diet is the primary driver of body fat loss.”
In other words, a kilojoule is a kilojoule, and the type of kilojoule makes little if any difference. This appears to be particularly true over the long periods of time – years and decades – during which most adults gain weight. Or, less frequently, lose it.
Several years ago, Hall oversaw a multi-center trial, funded by Taubes’ research group, to test the carbohydrate-insulin-fat gain hypothesis. The results of that trial are being prepared for publication. It will be interesting to see how they compare to the new NIH study results.