It's a bit more complicated than taking on person at one time. The numbers of subjects in each individual test are not large but each person serves as his or her own control. The rank order of effect is normally quite consistent. Moreover many tests have been done using both T1 and T2 subjects and many foods have been tested. There is a strong correlation between the GIs of common foods tested in mutliple laboratories. So for example the GI of typical factory wholemeal bread is 74+/-2 , lentils 32+/-5 and
instant porridge 67+/-5. Moreover when tested there was a significant correlation found between the ranking of 20 basic foods tested on normal subjects and people with diabetes .
http://www.ncbi.nlm.nih.gov/pubmed/18835944
The GI is a tool, like any tool it has to be used sensibly It's frankly not sensible to eat a snickers bar compared with eating something with more nutrients. (or indeed rely on a GI of a product from a different country, quite frequently the recipes and hence GIs will differ, check snickers bars from the US v those from Australia,) The overall nutrient content obviously should play a part. Eating non or minimally processed foods though eliminates this sort of food from consideration.
Where it comes into it's own is selecting between similar carb loads of similar items.
(ie it's no use comparing a slice of bread containing 15g of carb and a cup of rice cntaining 40g of carb)
If you are going to have some rice, you pick a lower GI type rather than a higher one like Jasmin or sushi. GI rankings demonstrate that grains like barley will have a lesser effect on glucose levels than for example couscous.It quite clearly shows that you are better having boiled new potatoes than a jacket potato.but that it might be even better to choose legumes or certain types of beans more frequently.
If you want to try porridge and you find it causes a spike, don't be surprised if you are using a modern instant type (or even fine 'rolled' oatmeal) there GI is high.
Bread, well it's quite clear from the GI that typical factory white and wholemeal are very similar in terms of GI. If you are going to eat any bread then it's better to go for the coarser heavier breads containg, whole seeds and rye or other grains or even to choose Industrial breads designed to be lower GI like Vogel and Burgen.
I hardly ever look at a Gi table now, I know the type of foods that have lower GIs and the properties they have that tend to make them lower GI. Learning about it in the 3 years that I use diet to 'control' my undiagnosed LADA was one of the best things I ever did. I'm convinced that and exercise kept my levels down , it also gave me a good start when beginning to use insulin. I still use those principles, do not take large amounts of insulin and have had good levels for many years.
But like anything it's best to learn for yourself about it and use it as one tool to help control your glucose levels. It is not stand alone since carbohydrates only form part of the diet and diet only a part of diabetes control.
The Sydney University Glycemic Index site
link already given describes the basics and has a data base. The GI news is also worth looking at. (don't look at the recipe link on the home page, it should go to a list of lower GI recipes on another site but it actually links to the general recipes on the site so could confuse)
The
Montignac site also already linked to, has some excellent info on how the GI of foods is altered by the type of starch they contain and methods of cooking and processing. It has a good definition of what GI and how it works which might help correct some of the misrepresentation that occurs.