Diets that distinguish between "good carbs" and "bad carbs" are not an effective way of controlling blood sugar levels, a new study suggests.

Although made popular by the South Beach Diet and others, the glycemic index has never been fully embraced by most dietitians and remains a point of debate among scientists.

Now, diabetes researcher Elizabeth Mayer-Davis of the University of South Carolina says the use of the index should be ended altogether in favor of more traditional methods of losing weight and reducing the risk of diabetes — eating less and exercising more.

"The glycemic index is sufficiently flawed as an index that it is not helpful for scientists or people trying to create a healthy diet," Mayer-Davis said.

The glycemic index is a 100-point scale, with white bread at 100 points, that measures how quickly carbohydrates enter the bloodstream as sugar.

According to index supporters, people should avoid high-glycemic foods such as white bread and potatoes because they will quickly raise a person's blood-sugar level. Meanwhile, low-glycemic foods such as carrots and apples are absorbed more slowly, making a person feel full longer and reducing cravings, which helps with weight loss.

Promoters of the diet also say that eating low-glycemic foods will result in less fluctuation in their blood sugar levels.

Both the Atkins and South Beach diets have raised interest in the theory, and an entire series, "The Glucose Revolution," guides consumers through a diet based on it.

Beth Kunkel, a professor of food science and human nutrition at Clemson University and president of the South Carolina Dietetic Association, said that while there is debate among dietitians about its validity, it would be a mistake to reject the concept altogether. Kunkel was not involved in the University of South Carolina study.

"To just reject it out of hand and quit working on it would be a mistake," Kunkel said. "I just think we're five to 10 years away from really understanding it from a research viewpoint."

Previous studies have shown conflicting results. One small study showed that people on a low-glycemic diet were less hungry later in the day than a group fed a high-glycemic diet. Another study, involving 39 overweight people, showed that those on a low-glycemic diet lowered their risk of heart disease. Both studies were conducted by Dr. David Ludwig of Boston's Children's Hospital.

American Heart Association officials have disputed the significance of those findings.

The new study, published in the February issue of the British Journal of Nutrition, relied on food questionnaires from more than 1,000 people over five years and assessed their consumption of high- and low-glycemic foods. Researchers tested their blood sugar levels twice during the study period and found no significant correlation between the glycemic index of foods and the blood-sugar levels of participants.

Mayer-Davis said that researchers should develop a new measure of how different carbohydrates can affect health. She said a better index would be based on the physical characteristics of foods, such as fat content and calories, because numerous factors influence a food's effect on blood-sugar levels.