In the study, 21 overweight volunteers were divided into three categories: Two groups were randomly assigned to either lowfat or low-carb diets with 1,500 calories for women and 1,800 for men; a third group was also low-carb but got an extra 300 calories a day.
The study was unique because all the food was prepared at an upscale Italian restaurant in Cambridge, Massachusetts, so researchers knew exactly what they ate. Most earlier studies simply sent people home with diet plans to follow as best they could.
Each afternoon, the volunteers picked up that evening's dinner, a bedtime snack and the next day's breakfast and lunch. Instead of lots of red meat and saturated fat, which many find disturbing about low-carb diets, these people ate mostly fish, chicken, salads, vegetables and unsaturated oils.
"This is not what people think of when they think about an Atkins diet," Greene said. Nevertheless, the Atkins organization agreed to pay for the research, though it had no input into the study's design, conduct or analysis.
Everyone's food looked similar but was cooked to different recipes. The low-carb meals were 5 percent carbohydrate, 15 percent protein and 65 percent fat. The rest got 55 percent carbohydrate, 15 percent protein and 30 percent fat.
In the end, everyone lost weight. Those on the lower-cal, low-carb regimen took off 23 pounds, while people who got the same calories on the lowfat approach lost 17 pounds. The big surprise, though, was that volunteers getting the extra 300 calories a day of low-carb food lost 20 pounds.
"It's very intriguing, but it raises more questions than it answers," said Gary Foster of the University of Pennsylvania. "There is lots of data to suggest this shouldn't be true."
Greene said she can only guess why the people getting the extra calories did so well. Maybe they burned up more calories digesting their food.
Dr. Samuel Klein of Washington University, the obesity organization's president, called the results "hard to believe" and said perhaps the people eating more calories also got more exercise or they were less apt to cheat because they were less hungry.
Any questions? Lots. Does the Atkin's diet work? Based on this really, really good study, yes it does. Wow.