Glycemic Index: Unreliable?

As we’ve described in several different posts and articles here at, the glycemic index is a rating system that aims to predict how your blood glucose will respond to a certain food. A food with a high glycemic index (like white bread) will raise your blood glucose level quickly and dramatically, while a food with a low glycemic index (like kidney beans) will cause a less pronounced rise in blood glucose. Of course, how a food affects your blood glucose will also depend on how much of it you eat.

But a recent study has found that even when people eat the same amount of a given food, the glycemic index isn’t a reliable guide to how their blood glucose responds. Published earlier this month in The American Journal of Clinical Nutrition, the study looked at the effect of eating white bread in 63 participants — men and women — who ranged in age from 18 to 85, and in body-mass index (BMI) from 20 to 35 (low-normal to obese). None of the participants had diabetes, since the researchers wanted to know whether the glycemic index was reliable in people with a normal biological insulin response. It stands to reason that in people with diabetes, the biological response to a given food could only be less predictable.


As described in a HealthDay article on the study, the researchers found that after eating an identical portion of white bread three different times over 12 weeks, participants showed blood glucose responses that varied, on average, 20% within the same person and 25% between different participants.

The researchers took various biological measurements from participants, and found that differences in participants’ insulin index (how much a food raises insulin levels) accounted for 15% of the variation in blood glucose responses, and differences in HbA1c (a measure of long-term blood glucose control) within the normal range accounted for 16% of the blood glucose variation. This means that most of the variation in blood glucose response — both within each participant, and across all participants — couldn’t be explained by these easily measured differences in insulin response and long-term blood glucose control.

The researchers concluded that the observed variation in blood glucose responses to white bread means that the glycemic index “is unlikely to be a good approach to guiding food choices.” However, there are certain limitations to this study — notably that only white bread was used as a test food. It remains to be seen whether a lower-glycemic-index food (like chickpeas) could lead to a more predictable blood glucose response, both in a single person and between different people.

What’s your view on the reliability of the glycemic index — do you use this system to guide your food choices? Does your blood glucose respond in a predictable way to eating the same food or meal on separate occasions, or does this response vary? Do you adjust your doses of insulin or other medications depending on the glycemic index of foods that you eat? Leave a comment below!