I know that you have T2 but this is also something that can be important for people with T1 since we have to inject the insulin, if the body normally produces it for a food; maybe we need to inject it for that food.
In a couple of small trials they have actually used the insulin index to calculate the insulin dose for people with T1diabetes.
In mixed meals with a higher protein/fat content it was more effective than using carbohydrate counting alone.
http://www.medscape.com/viewarticle/807058#vp_1
One thing you need to be aware of with the insulin index is that it is calculated on a basis of 1000 KJ portions (ie 240 calories.) That would be approximately
70g of Muesli (original Alpen)
300g of cod
160g of fillet steak.
70g of muesli is bigger than a proper portion size but I suspect many people would eat it. The steak is around a normal portion size. I doubt many people would eat 300g of cod.
There are people with Type 1 who use quite complicated calculations to take account of glucose produced from protein (rather than insulin elicited in the non diabetic). Here are some of their discussions
http://www.tudiabetes.org/group/tagers
This US carb counting course teaches a simplified TAG system and has a chart with estimated figures for protein foods using 58% of the protein for the amount of glucose produced.
A cup of milk (any variety ) has 12g carb but 17g of total available glucose. Foods like lean beef and chicken are rated at 4g per oz. (they also give a lower TAG for foods such legumes which have both lower insulin and glycemic indices
https://healthonline.washington.edu/document/health_online/pdf/CarbCountingClassALL3_05.pdf
Certainly,I take these things into account but on more ad hoc basis ie I know the sorts of meals I eat and their effect upon me so I will adjust the insulin based on previous experience of the food. ( I find 1 egg is for me as glucose raising as almost 10g of carbohydrate, others don't seem to see a rise from them)
