I copied over this section from www.nutritiondata.com. I think it really helps to put perspective on the GI and GL debate. They are immportant factors to consider, but there can be problems by using it, as well. Here it is:
Limitations of the Glycemic Index and the Glycemic Load
Some proponents of the Glycemic Index (including many diet books authors) would like you to believe that GI and GL are all that matters when selecting which foods to eat. In reality, diet is a more complex issue than that. ND agrees that the Glycemic Index is a marvelous tool for ranking carbohydrates (and much better than the old "simple" and "complex carbohydrate" designations). However, there are also many limitations to GI and GL, which are explained in this section. Consider this the warning that those diet book authors don't want you to hear...
Scarcity of GI data
Although methods for determining Glycemic Index have been in existence for more than 20 years, GI values have so far only been determined for about 5% of the foods in ND's database. Seemingly similar foods can have very different GI values, so it's not always possible to estimate GI from either food type or composition. This means that each food has to be physically tested. GI testing requires human subjects, and is both relatively expensive and time-consuming. The fact that only a very limited number of researchers currently do GI testing compounds this problem. Food manufacturers continue to introduce thousands of new foods each year. Since GI testing is neither required nor common (at least in the U.S.), this problem is likely to get worse rather than better.
(ND has derived a formula that can estimate the Glycemic Load for untested foods, based on comparative analysis with foods of similar composition. To learn more about this method, please see ND's Estimated Glycemic Load page.)
Wide variation in GI measurements
The above Glycemic Index table shows a single value of GI for each food. In reality, though, the measurements are not so precise. Reported values are generally averages of several tests. There's nothing wrong with that methodology, but individual measurements can vary a significant amount. For example, baked Russet potatoes have been tested with a GI as low as 56 and as high as 111! The GI for the same fruit has even been shown to increase as the fruit ripens. This amount of variation adds a great deal of uncertainty to GI calculations.
GI values affected by preparation method
The Glycemic Index gets even trickier when you take into account the changes in value that occur in response to differences in food preparation. Generally, any significant food processing, such as grinding or cooking, will elevate GI values for certain foods, because it makes those food quicker and easier to digest. This type of change is even seen with subtle alterations of the preparation, such as boiling pasta for 15 minutes instead of 10.
GI values affected by combination with other foods
While tests for Glycemic Index are usually done on individual foods, we often consume those foods in combination with other foods. The addition of other foods that contain fiber, protein, or fat will generally reduce the Glycemic Index of the meal. The GI of this "mixed meal" can be estimated by taking a weighted average of the GI's of the individual foods in the meal. However, this averaging method may become less accurate as the total percentage of carbohydrate decreases. Therefore, foods like pizza often create a higher glycemic response than the simple weighted average of the ingredient GI's would predict.
Individual differences in glycemic response
The rate at which different people digest carbohydrates also varies, so there are some individual differences in glycemic response from person to person. In addition it has been shown that one person's glycemic response may vary from one time of day to another. And finally, different people have different insulin responses (i.e. produce different levels of insulin), even with an identical glycemic response. This fact alone means that a diabetic can not rely completely on the Glycemic Index without monitoring his own blood sugar response. (This, of course, is a limitation of any food index, and not a specific limitation of GI.)
|