Reading the ACCORD study and the OP link
what it actually depits is that you can't set blanket Goals for diabetics to achiveve, but need to look at every individual assess several medical factors to pitch their control and treatment to there overall long term advantage...
Existing risk factors need to be taken into consideration, as it seems that for those who have pre-existing cardo problems, driving the A1c down to non-diabetic figures with high medication can actually increase the likely hood of a event happening etc..
You could be flipping a coin, increased risk of blindness or increased risk of heart attack? Less than 5% of diabetics suffer major impairment of vision, and treatment is improving with every year!
Comparing A1c's of non-diabetics to diabetics and assuming that because if they do fine on non-diabetic A1c's the diabetic automatically must etc etc... Is a load of tosh as the A1c yes can be used as an indicator of risks it can't not depict which or if a diabetic will have a complication.. There are diabetics with high A1c's that seem to avoid complications or have limtied complication, then there diabetics who have very good A1c's, but not only still get complication but also the impact is a lot higher!!!
After non-diabetics do get heart disease, circulation problems and infections that lead to amputations of limbs...