COLUMBIA, S.C. — The glycemic index (G.I.), which ranks carbohydrates according to their ability to affect blood glucose, is promoted as a diet that will help people lose weight and reduce their risks for heart disease and diabetes. But a study by a researcher at the University of South Carolina’s Arnold School of Public Health has found G.I. may not help people determine the foods they should eat — or avoid — to improve their health.
The findings, published in the February issue of the British Journal of Nutrition, show people should exercise caution with the G.I. diet, said Elizabeth Mayer-Davis, a diabetes researcher and the study’s lead author.
"There are valid reasons to question the G.I. scientifically," Ms. Mayer-Davis said. "This is an area in the field of nutrition that is controversial. It turns out despite all of the interest in the G.I., the scientific literature is very mixed."
Some studies show beneficial effects of low G.I. diets on diabetes or other conditions, and other studies show no effect, she said.
The limitation of the G.I., Ms. Mayer-Davis said, is the numbers in the index are based on blood-sugar levels recorded two hours after the ingestion of test foods, in a controlled experimental setting and after a person has fasted overnight.
"However, many factors can affect the impact of food on glucose levels in a ‘real life’ setting, including the length of time food is cooked, your body’s hormones and other foods that are eaten at the same time," she said. "In scientific literature, the G.I. of foods is based on fasting. This is unrealistic because we eat throughout the day."
The study, funded by the National Institutes of Health and conducted over five years, followed more than 1,000 people at four clinical sites.
The researchers wanted to determine whether study participants with a relatively low G.I. diet had lower overall blood-glucose levels compared to participants with a relatively high G.I. diet. Using several different measures of blood-glucose levels, the researchers found the G.I. of the diet was not related to any of the measures of blood glucose.
This outcome means the G.I. is probably not picking up the specific effects of food on blood glucose, Ms. Mayer-Davis said.