The concept of the glycemic index (GI) to rate the carbohydrates in foods was originally proposed in the early 1980s as a way to help people with diabetes manage their blood sugar. Since then, the GI has been promoted as a key tool for healthy eating in general and serves as a foundation for many popular diets, such as the South Beach Diet and the Zone. Low-GI diets are purported to reduce the risk of obesity, diabetes, and heart disease and even help prevent cancer, arthritis, osteoporosis, and a host of other ills. Some health experts have even called for the GI to be included on food labels and for it to be emphasized in dietary guidelines from the government. But in 2013, Health Canada ruled against such GI labeling because it is potentially misleading, and the recently proposed Dietary Guidelines for Americans all but ignore it.
Should you buy into this GI bill of goods?
A carb catalogue
The GI ranks carbohydrate-containing foods and beverages according to how much they raise blood sugar (glucose) when eaten by themselves on an empty stomach, with scores ranging from 0 to 100—the higher the number, the greater the effect. Sugary foods, refined grains, and potatoes generally rank high. Higher-fiber foods such as beans, whole grains, fruits, and most vegetables rank low to moderate. Various GI databases give different numbers, but here are some general (and sometimes surprising) rankings:
- Foods with low GI (under 55) include beans, barley, pasta, most fruits and juices, carrots, non-starchy vegetables, nuts, and dairy products.
- Foods with moderate GI (56 to 69) include unprocessed grains (such as most whole wheat and brown rice), corn, table sugar, most honey, soda, sweet potatoes, and ripe bananas, mangos, and papaya.
- Foods with high GI (70 and above) include white bread, white rice, pretzels, instant oatmeal, many cold breakfast cereals, boiled potatoes, and most melons.
The carbohydrates in foods with a high GI are quickly broken down into glucose and absorbed, leading to a rapid rise in blood sugar and a greater release of insulin, the hormone that moves glucose from the blood into cells, where it is used for energy. Eating foods with a low GI (assuming they are eaten alone) could thus result in fewer spikes in blood sugar.
The GI simply ranks the effect of carbs in foods, but does not take into account the amount of carbs eaten. A related concept called glycemic load adjusts for the amount of carbohydrates in a serving of food, so that when you eat more, the number rises.
It gets more complicated
The GI is not quite as straightforward as it may sound, however. First, it's not always easy to predict where foods will rank on the index. For instance, many sugary foods do not have a high GI. Table sugar actually has a moderate GI, and premium ice cream and chocolate cake score low.
Different types of rice, bread, and pasta have different GIs, as do different varieties of fruits and vegetables. How a food is processed, stored, and prepared affects its GI.
The more processed a food, usually the higher its GI. Thus, juice has a slightly higher GI than whole fruit; mashed potatoes higher than a baked potato; regular whole-wheat bread a higher GI than stone-ground whole-wheat bread. Added fat lowers GI, so potatoes that are fried in oil or prepared with cheese have a lower GI than plain boiled potatoes. Pasta boiled for shorter times (al dente) has a lower GI than soft-cooked pasta. As fruit ripens, its GI increases.
Further complicating the big picture is the fact that the GI measures foods in isolation under fasting conditions, which is not how we eat. Meals contain a mix of foods, which can make the glycemic response unpredictable. Fat and protein in a meal can blunt the effect of high-GI foods. What’s more, there are variations in glycemic response among people who eat the same food.
In recent years there have been hundreds of studies on low-GI diets, largely with mixed results. Even when studies find benefits from such diets, it's often unclear how much these are due to the GI and how much to the nutrients, phytochemicals, and fiber that such diets tend to be rich in. When researchers control for such factors, the benefits weaken or disappear. Results may also depend on whether people have insulin resistance or not.
Diabetes. Many observational studies, including a large Harvard analysis of nurses and other health professionals in the American Journal of Clinical Nutrition, have linked high-GI diets with an increased risk of developing type 2 diabetes. Now underway in Europe is a large three-year clinical trial on lifestyle interventions in people with prediabetes, which should help clarify the effect of the GI on the risk of diabetes.
What about people who already have diabetes? A review by the Cochrane Collaboration concluded that such diets can help them control their blood sugar and improve insulin sensitivity. However, several studies and reviews since then have called that into question.
The European Association for the Study of Diabetes advises people with diabetes to substitute low-GI foods for high-GI foods when possible. In contrast, the American Diabetes Association states that the total amount of carbohydrates in foods is generally a stronger predictor of blood glucose response than the GI. So for most people with diabetes, it concludes, the primary tool for managing blood glucose is some type of carbohydrate counting. Still, "the GI may be helpful in fine-tuning blood glucose management."
Cardiovascular disease. Some large observational studies have found that high-GI diets—and, even more so, diets with high glycemic load—are associated with increased risk of heart disease and strokes. One theory is that high blood sugar and insulin levels can, over time, increase low-grade inflammation and damage blood vessels, among other adverse effects. But clinical trials of low-GI diets (in which people are actually put on specific diets and then followed) have found inconsistent effects on cholesterol levels, triglycerides, blood pressure, biomarkers for inflammation, and other cardiovascular risk factors, especially when compared to high-GI diets with comparable fiber levels. Thus, some research reviews have concluded that there is insufficient evidence to support the use of low-GI diets for cardiovascular health.
In 2014, one of the most rigorous clinical trials on the effects of the GI on major risk factors appeared in the Journal of the American Medical Association. Researchers from Harvard and Johns Hopkins compared low-GI and high-GI versions of two healthful diets in overweight or obese people without diabetes over five-week periods. One diet was low in carbs, along the lines of the Mediterranean diet; the other was high in carbs and based on the DASH diet (designed to lower blood pressure). Contrary to the researchers’ expectations, the low-GI versions of the two diets did not improve cholesterol levels, blood pressure, or insulin sensitivity any more than the high-GI versions did. In fact, the low-GI version of the high-carb diet was less beneficial than the high-GI version, since it decreased insulin sensitivity, raised fasting glucose, and increased LDL (“bad”) cholesterol.
"The unexpected findings of the study ... suggest that the concept of glycemic index is less important than previously thought, especially in the context of an overall healthy diet, as tested in this study," according to the accompanying editorial.
Cancer. Several studies have suggested that the GI or glycemic load of foods may play a role in cancer development or prognosis. A study in the Annals of Oncology found associations between high-GI diets and prostate, colorectal, and pancreatic cancers. But again, there are inconsistencies. For example, a systematic review in Cancer Causes and Control found no clear link between a high-GI diet and colorectal cancer. And while some studies linked high-GI diets to an increased risk of breast cancer in postmenopausal women, others found no association.
Weight loss. When eaten alone, high-GI foods cause a rapid surge in blood sugar, which triggers a spike in insulin and then a sharp drop in blood sugar. Theoretically, this may lead to weight gain by stimulating appetite, activating brain regions involved in cravings, increasing fat storage, and reducing fat burning.Aresearch review found that while study results have been inconsistent, the evidence overall suggests that low-GI diets increase weight loss at least a little. In an observational study posted online in theAmerican Journal of Clinical Nutritionin 2015, Harvard researchers found that lower overall GI and glycemic load are associated with lower weight, especially when high-GI foods are replaced by high-protein foods, such as dairy products or seafood.
In contrast, a Spanish clinical trial in that journal in 2014 found that both low-GIandhigh-GI diets (calorie-restricted, moderate carbs) were effective in reducing weight over six months.
Bottom line: The GI is an interesting way of thinking about what we eat and may help people with diabetes fine-tune their diets. But the health benefits of using it remain unproven, and from a practical perspective it makes eating unnecessarily complicated. In any case, if you just follow the guidelines of a healthful diet (eating more whole grains, beans, fruits, and vegetables and cutting back on highly processed and sugary foods and beverages), you’ll almost automatically lower your glycemic load. Mediterranean, DASH, and other heart-healthy diets, for example, tend to have relatively low glycemic impact. Keep in mind, too, that the GI is not a measure of a food's overall healthfulness, which is far more important than simply its effect on blood sugar.