"This is the first real assessment of popular diet books to see what results they will produce," said Thomas Wadden, director of the Weight and Eating Disorders Program at the University of Pennsylvania School of Medicine. "The results are modest. The study shows that no single approach has a monopoly on weight loss."

But the study also found that even modest weight loss pays off: All the diets cut the risk of heart disease by 7 to 15 percent. "It shows yet again that modest weight loss can improve risk factors," said Kelly Brownell, director of the Yale Center for Eating and Weight Disorders.

About two out of every three American adults are overweight or obese, according to the latest government figures, placing them at increased risk of a host of health problems, from arthritis and cancer to diabetes and heart disease. About 300,000 deaths each year are linked to obesity.

Estimates are that about half of men and two-thirds of women are trying to lose weight, which has helped fuel the $37 billion annual weight-loss industry in the United States. Most dieters hope to shed about 30 pounds, a goal that the new findings suggest is unrealistic.

Researchers said the results, announced at a news conference at the American Heart Association's annual meeting yesterday in Orlando, might help settle the growing debate about the best way to shed unwanted pounds for good.

"This supports our data that it doesn't make any difference if you lose weight with high-carbohydrate or low-carbohydrate diets," said Gary Foster, clinical director of the Weight and Eating Disorders Program at the University of Pennsylvania School of Medicine, who did a recent study comparing the Atkins diet to traditional low-fat diets.

"Diets work if you use them. They all work probably by the same mechanism, which is that they get people to eat fewer calories."

Led by Tufts University researchers and funded by the Tufts-New England Medical Center with federal support, the study randomly assigned 160 overweight and obese men and women to follow one of the four diets for a year.

Participants ranged in age from 22 to 72. All had tried to lose weight before and all had at least one major risk factor for heart disease — high blood pressure, elevated blood cholesterol, abnormal blood-sugar level or diabetes.

All groups received instruction in their programs and four counseling sessions during the first two months. They were assessed for their ability to follow the regimens and then left to their own devices for the next 10 months. Researchers measured body weight, took blood and urine samples and collected food-intake records through the year.

About half of those in both the Atkins group (very low carbohydrate, high fat) and the Ornish group (very low fat, high carbohydrate vegetarian) dropped out before the study was completed. About a third dropped out of the Weight Watchers group (low fat, moderate calorie, similar to the diet advocated by the U.S. Dietary Guidelines) and the Zone (which relies on a system to track how much food raises blood-sugar levels).

"The more extreme diets like Atkins and Ornish were tougher to follow than the Zone and Weight Watchers," said Michael Dansinger, director of obesity research at the Tufts atherosclerosis-research lab and the lead author of the study.

All four groups showed about 3 percent weight loss overall over the year — equivalent to a 200-pound person trimming about six pounds. But among those who stuck with their diets for the entire year, results were slightly better, ranging from a 4 percent loss for Atkins to 6 percent for the Ornish group. Both the Weight Watchers and the Zone dieters lost about 5 percent.

Although this modest weight loss may not be what dieters are looking for, it brought significant health benefits. Risk of heart disease dropped 7 percent in the Ornish group, 11 percent in the Zone group, 12 percent on Atkins and 15 percent for Weight Watchers. None of the groups experienced adverse effects.

"A lot of people have been concerned that the Atkins diet will cause cholesterol levels to go through the roof," Dansinger said. "But we found that total cholesterol dropped by about 3 percent, and low-density lipoprotein went down about 8 percent."

Although neither of those changes is statistically significant, Dansinger said those numbers are more than offset by a 15 percent rise in "good" cholesterol — high-density lipoprotein.

Another concern addressed by the study is whether very high carbohydrate diets, such as Ornish, increase the risk of diabetes by overtaxing insulin production. Although there was an initial rise in insulin levels in the Ornish group, at the one-year mark insulin had dropped 27 percent.