The results of the study agree with the (sadly obscure) research on
metabolic differences among humans.  Just as we know that eating extra sugar
doesn't cure hypoglycemia, we shouldn't be surprised to learn that carb
restriction may not be the most effective treatment for diabetes.

Dr. George Watson noticed that diabetics tended to have low lipid levels and
an excessively alkaline blood plasma pH, the reverse being true for
hypoglycemics.  He inferred that the rate of glycolysis (oxidation of sugar)
was unusually slow in diabetics, and that, because of this, they were
relying excessively on oxidized fat for energy.  When these individuals were
placed on a high-carb, low-fat diet, their statistics normalized.
Conversely, glycolysis runs unusually fast in hypoglycemics, causing them to
rely excessively on carbs for energy, unless a low-carb diet is adopted.

A friend of mine has also considered differences in insulin production.  As
we know, hypoglycemics produce excessive amounts of insulin in response to a
flood of carbs; this could be because they aren't adapted to a high-carb
diet.  At the opposite metabolic extreme, there are those who need to
consume a relatively large amount of carbs (or caffeine) before sufficient
insulin is produced.

However, the above is not true for _all_ diabetics.  Some diabetics were
originally hypoglycemic, but eventually became diabetic because of
pancreatic exhaustion.  These individuals are in the minority (in the U.S.),
and probably include the Audettes and most Native Americans with diabetes.
Confusing, no?

Even though the conclusion of the study seems plausible to me, I would still
like to know more.  I'm especially skeptical of the finding that "obese
women put on high-sugar and sugar-free diets that were identical in other
ways both lost the same amount of weight, and their blood sugar and fat
levels were the same."  And advocating Olestra?  Ugh.

Cheers,
-- Aaron Wieland