It is very scientific. Aaaand, if I go too high >300 then I still need to do a correction with U-100 Novolog. The correction, according to the pump times 5 to get the right amount in my syringe. Isn't it crazy?
I am still confused about your usage. Based on what you said, you use 0.7U/hr for basal, which adds up to 16.8U/day. And then you are using an ICR of 30, assuming you eat 300 grams of carbs a day (which I expect you don't do), that is another 10U. So you are using less than 30U a day of U-500, your cartridge should last nearly a week. Are you sure we are all talking about the same things?
At breakfast, and dinner and snack, I have a different ICR for each. I am a very complicated person. I eat about 140 carbs/day, but my ICR for breakfast is 9. Lunch is 30 and dinner is 12 and snack is 20. In a perfect world, things should work with the math factors, but if I exercise or eat more then I also need to do more U-500. It is a confusing but logical answer. You should know that there is not a diabetic alive that is the same or reacts the same.
You are exactly right. Even when we think we finally understand ourselves, we turn right around and smack ourselves in the face.
Most of the U-500 regimes I have seen suggest a three times a day strategy and not even bothering attempting to bolus. Should you wish to try to peak your insulin for meals, you may need to bolus 3-4 hours before your meal. Have you been prebolusing?
I don't see why not if the site was good. I actually try for 4 days with mine. You would just have to reprime each time, but still cartridges are a lot cheaper than sets.
Zoe is right. I don't run my sets and cartridges in synch so that I almost never change my set when I change the cartridge. Just remember that you don't fill the cannula at the end of the cartridge change since it is already full.
good idea young_at_heart and zoe. Thanks.
BBach -
I don't think you're weird, just unique. While I don't have the same problems you do with insulin, you are so right in that no two diabetic persons are ever alike.