Low-carbohydrate poached salmon with green beans sesame was just one of the freshly prepared dishes on the menu for subjects of School of Public Health Visiting Scholar Penelope J. Greene's comparison of low-carbohydrate and low-fat diets. (Photo by John Doiron)

A Harvard School of Public Health study may stand dieting wisdom on its head, after low-carbohydrate dieters lost more weight than low-fat dieters despite eating 25,000 extra calories over a 12-week study period.

The findings generated national attention after Penelope Greene, a visiting scholar in the School of Public Health's Nutrition Department, presented her research last week (Oct. 13) at the annual meeting of the North American Association for the Study of Obesity in Fort Lauderdale, Fla.

The study, conducted with Nutrition Department Chair and Frederick Stare Professor of Epidemiology and Nutrition Walter Willett, put three groups of dieters on different regimens. They included a low-fat group, a low-carbohydrate group that ate the same number of calories, and a third group on a similar low-carbohydrate plan that included 300 extra calories a day.

Participants in all three groups lost weight, Greene said, with the low-fat group losing an average of 17 pounds and the low-carbohydrate group that ate the same number of calories losing 23 pounds. The biggest surprise, however, was that the low-carbohydrate dieters eating extra calories lost more than those on the low-fat diet. Participants in that low-carbohydrate group lost an average of 20 pounds.

While participants in that third group were apparently able to eat more and lose more weight, Greene cautioned that the study was just a pilot for the larger study she had planned from the start and which she is preparing now. As a pilot, the groups were relatively small, containing just seven participants each. That means that despite the raw numbers, statistically the three groups lost about the same amount.

But even that finding was notable because the higher-calorie, low-carbohydrate group was not expected to be the same as the other two. Conventional dietary wisdom indicates that the extra daily calories eaten by those in the third group should equal about seven pounds, a difference that, had it held true, would have been statistically significant even in such small groups.

"That third group convinces me there's something there worthy of further study," Greene said.

Project Director Penelope J. Greene, a visiting scholar at the School of Public Health's Nutrition Department, shows a trick for mixing a specially-designed recipe to Chef René Portillo of Ristorante Marino in Cambridge, where fresh foods for her study of low-carbohydrate diets were prepared. (Photo by John Doiron)

Investigator in a chef's hat

The study was carefully controlled for what participants ate over the 12 weeks. Rather than giving participants a list of approved foods and setting them free, Greene had the food prepared fresh daily at a Cambridge restaurant, Ristorante Marino.

First Greene herself, then the restaurant's chefs, prepared meals from a meticulously crafted menu, and bagged them so participants could pick them up daily.

Each bag, color coded and picked up in the late afternoon, contained that night's dinner, a snack, and the next day's breakfast and lunch.

Greene said she spent considerable time before the study began creating the menu, trying out different dishes and different preparations of the same dishes. Her aim was to make the meals tasty and as similar as possible even though people were eating very different diets.

The result was that participants had similar meals throughout the study period, though portion sizes and preparation did vary.

Greene selected aging baby-boomers as participants because of concern about health threats such as obesity and diabetes as the boomers age.

"These people are over 50, overweight, and over-scared," Greene said. "They're hearing from their physicians that if they don't lose weight they're going to have a heart attack or get diabetes."

The result, she said, was a motivated group of 21 study subjects and an unusually high retention rate. Only one participant did not finish the study.

Subjects were randomly divided into three groups. The low-fat group ate diets that consisted of 55 percent carbohydrate, 30 percent fat, and 15 percent protein. The two low-carbohydrate groups ate diets of 5 percent carbohydrate, 30 percent protein, and 65 percent fat.

The low-fat group and one low-carbohydrate group ate diets limited to 1,500 calories for women and 1,800 calories for men. The second low-carbohydrate group ate diets of 1,800 calories for women and 2,100 calories for men.

Greene herself was extensively involved in the food preparation. Not only did she design the menu, she also worked long days in the restaurant's kitchen in the study's first weeks preparing meals and training staff there.

As the chefs became familiar with the menu and aware of the need for meticulous accuracy in measuring ingredients and portions, Greene turned more and more of the preparation over to them. Even at this point, which she said came as a bit of a relief, she still conducted spot checks, tasting and weighing random portions to ensure the protocols were adhered to.

These lengths were needed, Greene said, to address potential failings, such as faulty memories and cheating, in studies that rely on subjects self-reporting their diets.

A measure of her study's success, Greene said, is that all participants lost weight over the study period. The average loss amounted to two to three inches on their hip measurements and four to five inches on their waist measurements. Several subjects asked for the study to continue even after the 12 weeks concluded, a request Greene had to decline.

Weighing in on "low-carb"

Greene's interest in low-carbohydrate diets goes back to 1999. Intrigued by the diet's supporters' seemingly counterintuitive claims, she spent a summer researching previous studies and then visited the New York City clinic of Robert Atkins, one of the diet's main champions.

Staff there presented evidence culled from patient records over 18 months. While Greene felt the evidence of weight loss presented by the clinic was compelling, it didn't satisfy her, mainly because the records would naturally be skewed to include more patients who experienced success on the diet and stuck with it than those who didn't and left.

Greene decided the best way to fill in the blanks between the Atkins data and the older studies she had reviewed was to conduct her own study and to provide the subjects with fresh food so she would know exactly what they ate.

"It seemed counterintuitive," Greene said. "How can you eat so much fat? The claim was that you can eat as much as you want as long as you keep the carbohydrates down."

Though Greene's study didn't allow low-carbohydrate dieters to eat all they wanted, it still provided extra calories that, somehow, didn't translate into extra weight.

While some have said her results defy laws of thermodynamics - since more calories should equal more weight, whatever the calories' source - Greene said the explanation is obviously more mundane.

It has already been shown that it takes more energy to metabolize protein than carbohydrate, Greene said, and that may provide a small part of the answer. Other possibilities are that the body may absorb less of what's eaten in a low-carbohydrate diet, which is high in protein and in fat. Whatever the answer, she said, she's sure the laws of physics still apply.

"I don't think for a minute that anything is violating the laws of thermodynamics," Greene said. "[But] there's no smoke and mirrors here."