Diabetes Forecast

Carbs: Beyond The Basics

The pros and cons of using the glycemic index for carb counting and meal planning

By Tracey Neithercott , , ,

Adam Gault/Getty Images

The glycemic load is the most immediate and direct measure of how a serving of food will affect blood sugar.
David Ludwig, MD, PhD

Carbs count. Eat too many and your blood glucose can spike. It’s the first lesson people with diabetes learn at diagnosis: Watch the number of carbohydrate grams you eat at each meal or snack. That’s all well and good, but what about the type of carbohydrate you choose to eat? Increasingly, researchers are asking that question. Their answer may surprise you.

The GI Revolution

Back in 1980, most people with diabetes were using carbohydrate exchange lists, and “glycemic index” was a term foreign to researchers, doctors, and patients alike. Scientists at the University of Toronto, led by David Jenkins, MD, PhD, DSc, began digging around in the body’s glucose response to different foods, and the term came into use. In a 1981 study published in The American Journal of Clinical Nutrition, the researchers listed what they called the glycemic index for 62 common foods.

The glycemic index (GI) measures the glucose response to a given number of grams of carbohydrate. Imagine a chart with a line representing glucose levels during the first two hours after you eat 50 grams of carbohydrate. The area under that line, when compared with results from a test using 50 grams of pure glucose, indicates a food’s glycemic index (graph, opposite). Foods with a high glycemic index have higher peaks and more area under the line than those with a low GI. The glycemic index, then, is a ranking of foods from zero to 100 based on blood glucose levels after eating. So 50 grams of a plain white baguette has a GI of 95 (and a taller graphed line) while 50 grams of an apple has a GI of 39 (and a shorter line).

“If you’re carb counting and if you have 20 grams of carbohydrate from an apple or a banana or rice, it’s [as if it’s] all the same,” says Thomas Wolever, BM, BCh, DM, PhD, coauthor of The New Glucose Revolution: The Authoritative Guide to the Glycemic Index—the Dietary Solution for Lifelong Health, professor in the Department of Nutritional Sciences at the University of Toronto, and coauthor of the 1981 glycemic index study. “But it’s not.”

You might imagine that the glycemic index would be useful for people with diabetes. Yet more than three decades after its discovery, the glycemic index still isn’t among the most-recommended meal-planning tools.

The Glycemic Load

Graphing Blood Glucose Response
To determine the glycemic index of foods, researchers plot glucose levels for two hours after subjects eat portions containing 50 grams of carbohydrate and compare the area under each line with that of pure glucose.

SOURCE: Adapted from The American Journal of Clinical Nutrition, January 2009

In its 2013 nutrition position statement, the American Diabetes Association says picking low-GI foods over high-GI ones “may modestly improve glycemic control.” Yet despite the ADA’s slight encouragement, the recommendations note that while some studies showed drops in A1C (a measure of average glucose for the past two to three months) from following a low-GI diet, others found no blood glucose improvement at all.

Most GI experts realize that use of the glycemic index alone won’t do much for diabetes control. That’s because eating too many low-GI foods can still cause high blood glucose if total carbohydrate grams aren’t managed, too. Proponents of using the glycemic index as a meal-planning tool for people with diabetes really support a combined meal plan of carb counting and choosing lower-GI foods. It’s what’s known as the glycemic load. While carb counting considers the total amount of carbohydrate and the glycemic index takes into account carbohydrate quality, the glycemic load factors in both.

Take, for instance, a cup of mashed potatoes and a cup of watermelon. Both foods have a high glycemic index. But the potatoes have a higher glycemic load because they have more grams of carbohydrate per 1-cup serving. “The glycemic load is the most immediate and direct measure of how a serving of food will affect blood sugar,” says David Ludwig, MD, PhD, professor of pediatrics at Harvard Medical School, professor of nutrition at the Harvard School of Public Health, and chair of pediatric endocrinology at Boston Children’s Hospital.

The Pros

Picking low-GI foods while managing how many carbohydrate grams you eat at each meal may help keep your blood glucose consistent. “If I eat, say, 25 grams of carbohydrate as potatoes, I will find my blood sugar much higher than if I eat 25 grams of carbohydrate as pasta,” says Jennie Brand-Miller, professor in the Boden Institute of Obesity, Nutrition, Exercise, and Eating Disorders at the University of Sydney in Australia and author of more than a dozen books on the glycemic index. “There is a threefold difference in the glucose response to the same quantity of carbohydrate.”

 Another plus: Low-GI foods are generally healthier. (Generally, though not always. While carrots are lower on the glycemic index than cookies, peanut M&Ms—thanks to their fat content—have a lower GI than vitamin-rich bananas.) A low-GI diet promotes fruits and vegetables, whole grains, high fiber, and dairy, Wolever says:  “When you look at the research, this is exactly in line with the recommendations people are making.” Unprocessed, unrefined foods (think an apple with peel) are harder to digest than their more processed counterparts (applesauce or juice), which means they usually rank lower on the glycemic index.

“When [processed] stuff gets into the upper part of the intestine, it’s as good as digested,” says Johanna Burani, MS, RD, CDE, a registered dietitian and certified diabetes educator in New Jersey and author of Good Carbs, Bad Carbs: Lose Weight and Enjoy Optimum Health and Vitality by Eating the Right Carbs. Take oatmeal, for instance. Steel-cut oats are cut whole oats, while instant oats are steamed and rolled as thin as possible. Both are whole grains and both have the same amount of healthy fiber, but because instant oats are easier to digest, they have a higher glycemic index. Either form of oats, however, is more nutritious than a doughnut.

The Cons

Read enough about the glycemic index, and you’re bound to find the common myth that a food’s GI measures how fast blood glucose rises after a meal. That incorrect definition conjures up images of mountain-steep glucose spikes for high-GI foods and gradual humps for low-GI foods. “It’s misinterpreted by industry experts all the time,” says Marion Franz, MS, RD, CDE, a certified diabetes educator, registered dietitian, and nutrition consultant.

In fact, when Brand-Miller plotted post-meal blood glucose levels for both high- and low-GI foods in a 2009 American Journal of Clinical Nutrition study, she found the shape of the glucose-rise lines was identical. The difference: High-GI foods led to slightly higher blood glucose levels compared with low-GI ones.

Is that reason enough to consider the GI of foods? Franz doesn’t think so, especially because short-term studies have shown mixed results when it comes to the GI’s effect on blood glucose and the longest study to date monitored results for only a year. “There are many, many more important things than the glycemic index for people to pay attention to,” Franz says.

One of the ADA’s biggest caveats to its glycemic index recommendation is the complexity of glycemic index research. Definitions of “high GI” and “low GI” often differ among studies, and in some studies it’s difficult to ascertain whether a food’s fiber content or GI is affecting a person’s glucose control. Plus, no two bodies respond exactly the same way to the same foods. Studies to determine a food’s GI are done in people without diabetes.

Some experts also point out the glycemic index’s impracticality. People eat a range of foods at a meal, which all contribute to how high a person’s blood glucose rises a couple of hours later. Dine on a mix of high- and low-GI foods, and you have a moderate-GI dinner. And that’s before adding other nutrients into the mix. Protein, fiber, and fat all affect blood glucose levels. “[The] stomach is a big sack,” says Alison Evert, MS, RD, CDE, coordinator of the diabetes education program at the University of Washington Medical Center Diabetes Care Center. “It doesn’t have a fat slot or a carb slot.”

*Weight is with core and stem removed.

SOURCES: U.S. Department of Agriculture; Diabetes Care, 2008

A food’s origins come into play, too. Where a food was grown, how it is cooked, the degree of processing, brand, ripeness—all contribute to how your body processes foods containing carbohydrate. So it can be hard to know the exact glycemic index of a food you plan to eat. Such uncertainties make critics of the glycemic index leery.

Determining whether your glycemic index list is accurate is another issue. Glycemic index values can vary from lab to lab. Two large, well-respected labs are at the University of Sydney in Australia, headed by Brand-Miller, and in Toronto, led by Wolever.

The University of Sydney’s online resource (glycemicindex.com) has a searchable database of foods and their glycemic indexes, but even following those values can’t be completely accurate because foods are made and grown differently from country to country.

That said, experts agree that the perfect shouldn’t be the enemy of the good. “We don’t need to split hairs,” says Brand-Miller. “The opponents of the GI are making it sound much more variable and complicated than it really is.” And critics who point fingers at the tool’s inexactness forget that carb counting isn’t an exact science either, Ludwig says. You know that a medium apple is about 25 grams of carbohydrate, but yours may be larger or smaller. When you eat out, chances are you estimate grams of carb based on counts from similar foods. But it’s rare to know the exact amount of carbohydrate.

The bottom line, proponents of the glycemic index say, is that following a low-GI eating plan in addition to carb counting is more effective than carb counting alone, regardless of whether your particular slice of bread has a glycemic index of 70 or 71. “The big studies consistently show that a high-GI food will consistently produce a greater [glucose] response than a low-GI food,” Ludwig says. Which means picking low-GI foods while managing total carbohydrate grams may help your efforts at blood glucose control.

The Experts’ Tips

The goal of paying attention to GI isn’t to nitpick over small differences in the glycemic indexes of foods but to learn which are considered high-, medium-, and low-GI foods, then eat accordingly.

  • Choose unprocessed foods rather than refined and processed ones. Include the edible peels of fruits and vegetables, for example. True whole-grain bread—the kind Burani describes as being dense, brown, heavy, and speckled with visible whole grains—is lower on the glycemic index than white bread or even whole grain that’s been ground into a fine flour.
  • Bypass “puffed” grain products. “By puffing up the Cheerios or the Rice Krispies, you are enlarging the surface area for the digestive enzymes to work,” Burani says. The result: The body digests puffed grains faster.
  • Avoid “instant” products. Instant grains—rice, oats, pasta, potatoes—are also higher on the index because they’ve been previously chopped, steamed, and cooked in order to make for quick prep on your part. So replace the instant rice with the longer-cooking variety, pick steel-cut oats over instant oatmeal, and look to fiber-rich grains such as quinoa or pearled barley instead of starches such as pasta and white potatoes.
  • Don’t overcook grains and starches. Just as puffing grains increases their surface area and speeds digestion, so does overcooking grains. Consider pasta and rice, which begin hard and thin but swell with water as they cook. Let it boil too long and your pasta or rice will expand in volume, be easier to digest, and have a higher GI. Typically, raw fruits and vegetables have a lower GI than cooked versions.
  • Consider cold potatoes. Because hot potatoes are higher on the glycemic index than cold ones (a GI of about 75 to 55, respectively) thanks to changes in the vegetables’ starch, Wolever suggests cooling cooked potatoes and eating them in salads or as potato salad with oil, vinegar, and other veggies.
  • Favor nearly-ripe fruits. An under-ripe banana, for example, has half the glycemic index of a ripe banana. Still, any banana is  healthier than banana-flavor ice cream.

The Plan

Putting it together—carb counting plus watching the glycemic index—can seem daunting, but it’s easier than it sounds. It’s really a matter of knowing your total carbohydrate goal for each meal or snack and then picking the most healthful carbohydrate-containing foods among your options. “I try to teach my patients how to identify a [healthful carbohydrate],” says Burani. “If it looks like it hasn’t been highly processed, that it isn’t so far removed from the grain it started as, you’re looking at something that’s at the lower end of the glycemic index.”

And keep in mind that monitoring the GI of foods isn’t for everyone. “Sometimes we make things more difficult,” says Evert, who says monitoring the glycemic index of foods on top of carb counting can be time-consuming. Practically speaking, you don’t have to eat only low-GI foods. “Have at least one low-GI food per meal,” Wolever says. To do that, all you need to do is select at least one nonstarchy vegetable, which has a negligible GI value. Or do as Evert suggests and make low-GI choices when it comes to snacks consisting of just one food.

Above all, healthfulness should come first. You shouldn’t pick a Snickers bar over a ripe banana, even though the Snickers has a lower GI. Generally, if you follow an eating plan full of fruits and nonstarchy veggies, whole grains, low-fat dairy, beans, and lean protein—the general guidelines for healthful nutrition—you’ll be eating lower-GI foods, too.



Take the Type 2
Diabetes Risk Test