Glycemic Index Confusion

My publisher wants a glycemic index (GI) chart for our upcoming book on smart glucose monitoring. I’m looking into GI and its sister, glycemic load (GL), but so many things can change GI and GL. How do you use these tools?


As Jacquie Craig, MS, RD, LD, CDE explained in an article on this site,
GI ranks carbohydrates from 0–100, based on how much and how fast they affect blood glucose levels. A higher number means the food has a larger impact on blood glucose levels. Glucose itself has a ranking of 100.

GI is important for people with Type 2 diabetes, because they often have a delayed insulin response. If glucose goes up fast, the body does not respond quickly enough, and glucose levels can get way too high after meals. So we want low-GI foods.

GI doesn’t tell you how much glucose will eventually get into your system, just how much of a blood glucose spike the food creates. To improve the GI, Dr. Walter Willett at Harvard helped develop the concept of glycemic load (GL). GL combines the GI with a measure of how much carbohydrate there is in a food.

So in theory, GL can tell you a given food’s total impact on your blood glucose levels, which should help in meal planning, insulin dosing, and food choices. But the reality of GI and GL is much more complicated.

How Are GI and GL Determined?
You can’t tell the GI or GL of a food by analyzing it in a lab. That’s because different people’s digestive systems handle different carbs differently. We also absorb and break down the same carbs into glucose differently at different times, depending on what other nutrients are being consumed.

For example, a plain pizza with tomato sauce and Parmesan cheese has very high GI, around 80. But a super-deluxe pizza with all the fixings has a low GI of 36. The GI is lower because the protein and fats in the toppings slow down the absorption of carbs and slows their turning into glucose.

Since you can’t tell GI or GL in a test tube, glycemic index values are determined by feeding 8 to 10 people a fixed portion of the food (after an overnight fast). Then samples of their blood are taken every 15–30 minutes and their glucose levels are measured. The GI values that the foods register in the participants are averaged to give a GI number.

The GL is calculated from the GI using the formula GL = (GI × Net Carbs) ÷ 100. (Net carbs are equal to the total carbohydrates minus dietary fiber.)

So if a plain pizza has a GI of 80, and 27 grams (about one ounce) of net carbs in 100 grams, its GL would be 80 multiplied by 27, divided by 100, for a total of 22. For GL, 20 and over is considered high, and 10 or below is considered low. Levels in the range of 11–19 are considered medium.

For GI, anything below 55 is low, and anything over 70 is considered high. Numbers from 55–69 are medium.

The same foods can have a very different GI and GL depending on how they are prepared. A boiled sweet potato has a low GI of 44 and a medium GL of 11. But if baked for 45 minutes, the same sweet potato has a GI of 94 and a GL of 42, both extremely high. Baking has essentially turned the sweet potato into candy.

White potatoes also have a higher GI and GL when baked. Microwaving often raises GI and GL. Have you ever noticed how sweet beets taste after baking or microwaving? That’s because much of their carbohydrate content has been converted into glucose.

Even the same cooking method can give different results. Spaghetti cooked al dente (boiled for 8 minutes) has a much lower GI and GL than soft (boiled for 20 minutes) spaghetti.

Different brands or varieties of the same food can have very different GI and GLs. Professor Jennie Brand-Miller of the University of Sydney, Australia, has created a list of thousands of foods based on her tests and the published results of others. Her list includes two types of rice milk, one with a GI of 92, another of 79. Sweet corn from New Zealand has a GI of roughly 37, while South African sweet corn gets about a 62.

This is really a wonderful, if confusing, list, and it can be seen at and an updated list is available at

Different People, Different GI/GL
Even if you could find the published GI or GL of a specific food, you couldn’t be sure of that food’s effect on your personal glucose numbers. People vary significantly in their response to foods. And in real life, foods are rarely consumed one at a time. We have drinks and other foods with them, which can affect their glucose response in the body.

It seems the only way to be sure about a particular food’s effect on you is to check your glucose after eating it. Preferably check before, too, so you can record how much change there has been.

Do you look at GI and GL numbers or think about them in meal planning or insulin dosing? What information sources do you use, and how do you use that information? Thanks in advance for any help you can give.

Learn more about the health and medical experts who who provide you with the cutting-edge resources, tools, news, and more on Diabetes Self-Management.
About Our Experts >>

  • Bob Fenton

    You do ask the hard questions! You are right that everyone varies in their reactions to foods and how it affects them. The largest fallacy of both the GI and GL is the determinations were made using healthy people. No one has done any calculations on people with diabetes or any other disease for that matter.

    I use this book – The New Glucose Revolution, New York, Marlow & Company, 349 pages, by Dr. Jenny Brand-Miller, et al. I use it only as a guide and I have found like you stated, only your meter reading from the test strips will tell you how a food affects your blood glucose.

    More research needs to be done with food combinations and how the various combinations and cooking combinations affect blood glucose levels. Until then it is at best a guessing game.

  • calgarydiabetic

    Dear David

    You should caution people that GI and GL are only suggestions as what to eat and what to avoid. The 1 hour , 2 hour and for pizza even 4 hour spikes are what is important to the individual as to type and amount of food.

  • Ted

    The GI and GL are good indicators of what carbohydrate-rich foods can do to your blood sugar and insulin levels when you are forcing yourself to eat more carbs than you should, but in the end, it’s pretty much common sense.

    Processed carbs, and natural foods that taste sweet, should be avoided all the time. Other carbs should be eaten in moderation. If you stick with that, the GI and GL are simply supporting documentation.

  • joan

    David – Thank you for the GI GL article.

    I understand the GI and GL but I have not used them very much as a Type 1 for 55 years. I know what foods will raise my blood glucose level and that is what counts more for me than any other means to figure what to do for better control.If using the GI and GL helps a person then use it as a guide.

    My theory: If it hurts, do not do that any more!

  • Dr. Harola Sussman

    This was a very illuminating piece. My son, who is a psychiatrist, told me to eat brown rice, not white rice, sweet potatoes, not white potatoes, and whole wheat bread, not white bread, due to their lower glycemic index. He did not mention that the method of preparation alters the glycemic index of various foods.
    Thank you for enlightening me since OI have Type 2 diabetes.

  • Ferne

    This is very interesting especially to find out the method of cooking varies the numbers. It is frustrating to me with recipes for diabetics that so many have rice and pasta. Any kind raises my BS even if I eat a small amount so have to just not eat it. It’s a trial and error to find out what I can or can’t eat. It’s no wonder diabetics get depressed. I had the chief of endocrinolgy tell me that being diabetic is a downward sprial the rest of my life. That is something no one should hear. Just a little encouragement, please.

  • David Spero RN


    You problems with rice and pasta are common in people with diabetes. I personally don’t think diabetic recipes should include them.

    If you want some encouragement, read my blog of a few weeks ago “Can Type 2 Diabetes be Reversed?” I’m writing a longer piece on that for the print version of Diabetes Self-Management for November.

  • Ron

    Having researched this topic for several years it is good to see the subject raised. I am a type 2 diabetic using both basal and bolus insulin.
    Although the article contains some relevant facts much of it seems simplistic or flawed.
    1. Glycemic Index
    The origin of the GI is the standard test for diabetes which is a glucose drink after fasting with measurements of BSL before drink, at one hour and at two hours. It is known that in non diabetics the pancreas will produce sufficient insulin to allow the glucose to be absorbed into muscle tissue (fat too) in two hours . In diabetics (T2) the muscle tissue becomes resistant to the insulin and allows a limited amount of glucose to reach the muscle tissue. This results in the glucose and insulin levels in the blood to become elevated.
    The question is ‘Does GI serve any other significant purpose than the test?’. In my opinion the answer is ‘no’. There simply are too many variables. Firstly consider the fact that we eat multiple food types in a meal. The GI of the meal is therefore the combination of the individual foods consumed, but calculation is near impossible for the diabetic. Remember that the goal of GI control is spike elimination, but how can we control spikes without knowing the meal GI? And when will the spike happen? Consider the Pizza mentioned. Does the plain pizza really have a GI of 80? What flour was used? Pizza bases vary in size and toppings vary too. So can we be sure that the super deluxe has a GI of 37 ( and I bet the 8 test people had pizza’s from the same supplier). Then consider the absorption rate which depends upon the type and amounts of toppings used. Further consider that you are unlikely to consume the pizza alone. Maybe coffee or tea. Milk with low GI will alter meal GI. Caffeine will alter metabolism. Maybe fruit afterwards, again low GI.
    Now consider yourself, a diabetic (T2)what is happening to your body. Chances you are on Metformin taken with meals.When will that kick in and how will it affect your BSL after taking it? This morning I had a waking BSL of 4.6, before breakfast it was 6.2. Anxiety levels, exercise and many other factors are at play that all affect the BSL. It’s not simply about what you consume.

    2) Glycemic Load
    A better form of measurement but the dependency on GI ruins the calculation and the dependency on carbs ignores the absorbtion rate. But as a tool for BSL control, rather than spike control, it has merit because it counts carbs, and the amount of carbs consumed affects BSL directly, any T1 will confirm this.

    In conclusion I would suggest that the best way for good BSL and spike control is simply a balanced diet combined with carb counting.

    I am from Australia and here dieticians don’t understand T2 diabetes. Even in hospital a few weeks ago I was fed a diabetic diet. Guess what 70% carbs. I have a max of 30 carbs per meal. They gave me 70. I ate about half. Always ask for a low carb meal rather than a diabetic meal.

  • jim snell

    Pasta and rice are big pain in derriere.

    Frrying rice in oil flattens the peaking on rice making it edible for me.

    Pasta is a huge pain. First off, the dam pasta and a saurce – usually cheese cause a huge stall in gut/intestine glucose production whereby I get a initial amount of glucose than all stops for 2 hours as intestine chews its way through pasta.

    For the person without any liver fifo glucose add riots – this is ok. For me – where I am deliberatly keeping BG above 100 and liver booted out, the system stall for 2 hours means I am adding glucose during stall and then I get get a hammering high.

    The pasta is simply huge pain in butt,

  • John_C

    RE: comment from Dr. Harola Sussman:

    “brown rice, not white rice, sweet potatoes, not white potatoes, and whole wheat bread, not white bread, due to their lower glycemic index”

    My Meter has told me that these high carb foods will raise my blood sugar up really fast! You can add a bunch of other foods to the list… you know the ones that your ADA Dietitian says are OK.

    Only your Meter will tell you what you need to avoid.

    “I had the chief of endocrinology tell me that being diabetic is a downward spiral the rest of my life.” Another garbage statement from the “baron wasteland of ignorance” (my quote 🙂

    If you want to start a good conversation ask why test strips cost so dam_ much and why they are not very accurate a lot of the time.

  • Ron

    “brown rice, not white rice, sweet potatoes, not white potatoes, and whole wheat bread, not white bread, due to their lower glycemic index”

    With all due respect this is exactly the simplistic and incorrect information we are all being duped with.

    There is nothing wrong with these foods when eaten in moderation. It’s what you consume with them that matters.
    I’ll take my small portion of boiled white potatoes with two veggies and beef over your baked sweet potato and chicken breast. I’ll take my slice of buttered white bread, boiled eggs, cereal and orange juice over your whole wheat bread for breakfast. And I will get lower spikes than you will.
    But how can you measure your spikes with a meter, say every 15 mins, if strips cost so much and are so inaccurate?
    Now as to the ADA dietician, forget them. They advise you eat unbalanced diets. They are all about “avoid, avoid, avoid” rather than “moderation”. Why they have the word ‘Diabetes’ in their title defeats me. Their website is clearly about their views of healthy eating for the public at large. Examples, Sodium, Cholesterol,fats are for all people. Fats are mentioned in detail but no mention I can find about the fact that fats reduce absorption and therefore BSL spikes. There is a growing problem with over-consumption of polyunsaturated fats. Cholesterol produces testosterone, hands up the guys who want to give up their sex lives!

    Again, eat a balanced diet and count the carbs.

  • acampbell

    Hi Ron,

    Thanks for sharing. I certainly can’t speak for all dietitians, but I think it’s a little unfair for you to state that all dietitians advise people to eat “unbalanced diets” and to “avoid” everything. While that may be the case with some dietitians, it’s not true for all, including myself. I personally take the view that, in general, most foods can be part of a diabetes eating plan. I don’t believe in telling people with diabetes not to eat carbohydrate. The issues are the type, the amount, and other foods that you eat along with them. And the concepts of glycemic index and glycemic load are just those: concepts. Each person with diabetes needs to find out how foods affect their own diabetes. Blood glucose monitoring and continuous glucose monitoring can help with that. Yes, it takes time and effort, but it can be done. Finally, the glycemic index of a food does not necessarily reflect its nutritional value. To give you an example, brown rice (a whole-grain, unrefined food) has a higher glycemic index than Uncle Ben’s converted white rice (a refined food). Good discussion, and always food for thought!

  • jim snell


    excellent points and thank you for clarifying.

  • Tyrker

    “It seems the only way to be sure about a particular food’s effect on you
    is to check your glucose after eating it. Preferably check before, too,
    so you can record how much change there has been.”

    Sound advice David.

  • snootyelites

    Sweet potato is ok both in glycemic & nutrition terms. Only problem is quantity – they are huge. Best way to cook sweet potato – spray oil one inch pieces, nuke for 3 minutes and bake for 12-13 minutes or more till they turn crispy. Salt & Pepper it. No more than 2 times a week.