The glycemic index (GI) of a food indicates the speed with which blood sugar can be expected to rise after a person eats it. Each food gets a score out of 100 on the index, for example, 40 for baked beans.

Glycemic load is a measure that applies the GI to a portion of food. The glycemic load for a 150-gram serving of baked beans would be 6.

GI is used to help people with diabetes to control their blood sugar. Some food labels carry GI measurements. More recently, a number of popular diets have been based on GI. Lists are available for people to check the GI and glycemic load of different foods.

However, its usefulness is controversial, because individual responses to a particular food can vary.

Scientists from the Jean Mayer United States Department of Agriculture (USDA) Human Nutrition Research Center on Aging (HRNCA) at Tufts University have questioned whether GI is a valid measure, after randomized, controlled, repeated tests on 63 healthy adults suggested otherwise.

Results for white bread vary widely

The volunteers participated in six testing sessions over 12 weeks. They fasted and abstained from exercise and alcohol before each session.

During the session, they ate either white bread or a glucose drink, in random order. The bread was the test food, and the drink was a reference control.

Each item contained 50 grams of available carbohydrate.

Blood glucose levels were measured several times over the next 5 hours, and GI was calculated using standard methods.

Findings put the average GI value of white bread at 62, making it a "medium" GI food, but scores varied by 15 points in either direction.

In 22 participants, the blood sugar response was low, in 23 it was medium, and in 18, it was high. This range effectively put white bread in all three GI categories. Moreover, individual responses varied by up to 60 points between tests.

This could be partly due to insulin index and baseline HbA1c levels. These readings reflect long-term glucose control.

In this study, these factors showed the highest range in measures, around 15 percent to 16 percent difference. This suggests that an individual's metabolic response to food impacts their GI values.

As a result, the scientists have questioned how useful the GI is for predicting a food's impact on blood sugar levels.

Can GI measures tell us which foods are healthy?

Lead author Nirupa Matthan points out that a person consuming the same amount of the same food three times should have a similar blood glucose response every time, but this did not happen.

As a result, she suggests that GI values are "unlikely to be useful in guiding food choices."

"A food that is low glycemic index for you one time you eat it could be high the next time, and it may have no impact on blood sugar for me,"

Instead, she suggests following a diet "primarily composed of vegetables, fruits, whole grains, nonfat, and low-fat dairy products, fish, legumes (beans), lean meats with preference to preparing food with liquid vegetable oils, and equally as important, to choose healthy foods and beverages you really enjoy."

Questioning the role of GI in clinical and public health

The team used a higher number of participants, more tests and a longer measuring window than most GI tests do.

They also looked at how the biological factors such as sex, body-mass index (BMI), blood pressure, and physical activity impact GI variability. In most cases, the impact was minor.

The authors point out that the results do not mean that food with a high GI is necessarily healthy, nor that one with a low GI is unhealthy, but it does imply that the usefulness of these measures for managing clinical and public health may be limited.

The authors call for experts to rethink the link between GI estimates and chronic disease risk.