Glycemic Load and Glycemic Index are related. Historically, the index came first and is a measure of the blood glucose raising potential of the carbohydrate portion of food relative to glucose. The concept of glycemic load was first coined at Harvard Medical school where a measure of the effective glycemic dose of different diets was required. The Glycemic Load was defined as:
GL = GI * wt of carbohydrate/100
GL is thus the product of a quality factor (GI) and a quantity, the weight of carbohydrate. Frequently, GL is used without any units. But it does in fact have the units of grams. The GL of a portion of food is (to a first approximation) given by the weight of glucose which will produce the same glycemic effect. In the following sections we give a mathematical derivation of the relationship of GL to GI and the glucose equivalent weight.
When you eat food containing carbohydrate there is a rise in the level of glucose in your blood followed by a fall as insulin is secreted to stabilise your blood glucose to its normal level.
The rise and fall can be easily measured on a blood glucose meter by taking a very small sample of blood from your finger every 15 minutes or so. A glycemic response curve can then be plotted and will typically look like Fig. 1

This is an actual measurement taken after eating 50g of pure glucose on an empty stomach. The glycemic response to eating pure glucose is taken as a standard reference to which other foods may be compared. The dotted line is the normal (or fasting) level. This level varies from person to person but in a normal, healthy individual it is surprisingly constant - a bit like body temperature.
The glycemic response will also vary from person to person, but the above curve is fairly typical of the response to the fairly large (and rapidly digested) glycemic load of 50g glucose. Such large glycemic loads usually give rise to the undershoot below the fasting level known as "post-prandial hypoglycaemia". Many health professionals believe that this effect is a significant factor affecting appetite and obesity.
In
order to compare the glycemic response for different foods, the area of
the curve above the fasting level is measured and compared with the area
for a test food. A typical result, on the same subject after eating three
50g GoodCarb Chocolate Brownies is shown in Fig. 2.
If we make the simplifying approximation that the areas under the curves are proportional to the weight of food consumed, we can use simple linear relationships to compare the glycemic response of the test food to that of glucose.
If A g is the area under the glucose curve for a weight of w g grams and A t the area for w t grams of test product, then we can define the glycemic load per gram (GL 1 ) of the test food as:-
![]()
The glycemic load for a weight w t of test food is simply the above expression multiplied by w t i.e.
Which has the units of grams and quite clearly expresses the meaning of glycemic load i.e.
The glycemic load of a specified weight of food is equal to the weight of glucose in grams which will give the same glycemic response.
[i.e. If A g = A t then GL = w g ]. One useful way of expressing glycemic load is as "grams glucose equivalent".
Like calories, glycemic load varies in exact proportion to the weight consumed. If you double the amount you eat, you double the glycemic load. A good way of thinking of glycemic load is as "glycemic calories". In many cases, but certainly not all, the glycemic load of a meal may be obtained simply by summing the glycemic loads of each of its components. (The reason for the proviso is that some foods strongly affect the glycemic response of others. E.g. putting butter on bread does not affect the glycemic load because butter has zero GL. However, fat reduces carbohydrate absorption and so will reduce the GL of the combination.)
Like calories, the glycemic load of a standard serving can be quoted and the glycemic properties of different foods may be compared by quoting the glycemic load per 100g in exactly the same manner as other nutrititional properties are given on food products.
grams
GL 100 allows a weight-for-weight comparison of the glycemic effect of different foodstuffs and does not require knowledge of the carbohydrate fraction of the product. As such it is a much more useful and more fundamental property of foodstuffs than the (to date) more frequently used glycemic index (GI).
Historically, the concept of glycemic index came before glycemic load. It is a means of rating the glycemic effect of the carbohydrate component of a foodstuff relative to the effect of glucose on a percentage scale where glucose is set equal to 100%. To determine the GI of a food you need not only to measure its glycemic response relative to glucose, but you also need to know its carbohydrate fraction F t .

Glycemic index is thus a measure of the glycemic quality of the carbohydrate component of a foodstuff. Clearly, GI and GL 100 are closely related via:-
Glycemic load is really by far the most important concept to grasp if you want to control your blood sugar levels. Using GI alone (as many diet books unfortunately do) can be a minefield. For example, if you just looked at GI you would conclude that eating watermelon (GI = 72) was bad for you but eating spaghetti (GI = 38) was good. But it is the glycemic load which affects your blood sugars and 100g of watermelon has a GL of 3g compared with 100g of spaghetti which has a GL of 9g.
See Also: