Low-Carb versus Low-Fat

Ladies and gentlemen of the low-carb community: We have a hat-trick.

1. On March 7, 2007, the Journal of the American Medical Association (JAMA) published an article showing that, at 12 months, women assigned to the Atkins (low-carbohydrate) diet lost more weight and experienced more favorable metabolic effects than did women assigned to follow the LEARN, Ornish or Zone diets.

2. On July 17, 2008, the New England Journal of Medicine published an article describing a two-year study of men and women in Israel. The study showed that, compared with the low-fat diet, the low-carbohyrate diet produced greater weight loss and had more favorable effects on lipids. The authors concluded that low-carbohydrate diets may be an effective alternative to low-fat diets.

3. And finally on August 3, 2010, the Annals of Internal Medicine published an article describing a two-year low-carb vs. low-fat study of men and women in the United States. The authors concluded that, "Successful weight loss can be achieved with either a low-fat or low-carbohydrate diet when coupled with behavioral treatment. A low-carbohydrate diet is associated with favorable changes in cardiovascular disease risk factors at 2 years."

Three refereed articles in three well-respected journals. Although the second study had some funding from the Dr. Robert C. and Veronica Atkins Research Foundation and might be faulted for that reason, the first and third were supported by the National Institutes of Health (NIH). All three studies showed that a low-carbohydrate is effective for weight loss. All three showed that metabolic effects, including an increase in HDL cholesterol, improved with the low-carbohydrate diet. And while the first study lasted a year, the last two studies covered a two-year span, demonstrating that the benefits of a low-carb lifestyle are not limited to a few weeks or months.

Currently the third article is only available for free in abstract form. However, Jimmy Moore purchased the article and did an excellent summary which can be found here. I purchased the article, too, and found that most of my observations agreed with Jimmy's, so I'll refer you over there if you would like a thorough discussion of what the article showed. [Great news! Thanks to LynMarie Daye in the Comments, we now have a link to a free PDF of the entire article: Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet.]

I'll just emphasize a few points.

First of all, the people in the third study lost an average of 11 kilograms at six months, while the average six-month loss for low-carbers in the first two studies was about 6 kilograms. That's probably because the average BMI in the third study was about 36, vs. about 31 in the other two studies. As a general rule, the more a person weighs, the easier it is to lose a given amount of weight.

Second, in the Annals of Internal Medicine study, the low-carb dieters lost (and regained) almost exactly the same amounts of weight as did the low-fat dieters. This may be because both groups received regular instructional sessions lasting from 75 to 90 minuntes throughout the two years of the study. Or it may be because the low-carb group was treated differently from the low-fat group. The low-carb group began the study at 20 grams of carbs per day, but at three months they were raised 5 grams of carbs per week until they reached a level of carbs at which they could maintain their weight. The low-fat group began the study eating 1200 to 1800 calories per day with less than 30% of their calories from fat, but they were never transitioned to a maintenance level of calories per day. In contrast with the low-carb group, at the end of the study the prescribed regimen for the low-fat group had not changed. It is hard to know how much additional weight the low-carb participants would have lost if they had been allowed to transition to their Critical Carbohydrate Level for Losing (i.e. the number of carbs that would allow them to continue losing 1-2 pounds per week) rather than being moved directly to a maintenance program.

Third, the weight loss in the low-carb group was not a loss of water weight. Both groups experienced similar reductions in lean mass (about 5%) and in fat mass (11% to 20%).

Fourth, in all three studies, the LDL cholesterol increased for the low-carb groups at three to six months, but was at or below baseline by the end of the study. Why this happens is not clear, but it seems to be a common finding when people begin a low-carb diet. Unfortunately none of the three studies measured LDL particle size, an important factor because small, dense LDL particles are more atherogenic than large, fluffy LDL particles. And people with higher HDL, as was seen in the low-carb group, tend to have the large, fluffy form of LDL cholesterol.

Fifth, an interesting aspect of the Annals of Internal Medicine study was the fact that it addressed the issue of dieting and bone loss. Both the low-carb and low-fat groups lost 1.5% or less of their bone mineral density during the course of the study. A small loss is unsurprising because the bones of both groups had less weight to carry as the study went on. However, there was no between-group difference in loss of bone mineral density in either the hip or the lumbar spine. There are blogs all over the internet suggesting that the relatively high protein intake of a low-carb diet causes calcium to be leached from bones and results in osteoporosis. The theoretical basis of this idea is shaky at best, and in a practical sense the Annals of Internal Medicine study indicates that this type of fear mongering is unfounded.

To sum it up, low-carbers now have solid scientific evidence that low-carb works for weight loss and that it improves metabolic health markers as well. If your doctor objects to your practice of the low-carb lifestyle, you might want to print out these three articles, read them, and take them along to your next office visit. For those who are skeptical about the benefits of low-carb, the positive scientific evidence is only getting stronger.