(The conclusion can not be drawn from this experiment! One is tempted to use words like morons, or idiots, but then one remembers that the objective of medical science is to sell pharmaceutical drugs through their marketing arm - medical doctors.)
I am tempted to ask the forum why the conclusion cannot be drawn, but let me go ahead and spill the beans. (In fact, those results might even be predicted, if medical researchers understood the nature of diabetes.)
The problem is the common medical mistake of incorrectly inferring cause from effect.
Even if one assumes that the data is not tainted, unbiased, accurate, etc. (if that is possible from food questionaires) the problem with diabetes, especially Type II, is that glucose (and in my opinion ascorbate) is not entering cells because of the problem with the cell membranes. (Readers may remember Glucose-Ascorbate Antagonism, and the issue of artifical trans fats "gummy" up cellular membranes.)
Ergo, diabetics have higher blood sugar. This dramatically reduces the availability of the short-lived ascorbate to enter cells, thus taking more has little benefit until the membrane disturbance is corrected.)
But to conclude that increased DEATH is caused by vitamin C, is erroneous. (I would love to know how they "adjusted" for the taking of diabetic drugs, because that would be MY explanation as the most probable factor for the higher death rate.)
I can think of countless factors that can be correlated with the death rate, and the big problem with this approach is this is exactly the hypothesis they set out to prove - that high vitamin C as a prooxidant hurts diabetes patients. It is not surprising that they found what they were looking for.