ichael, There have been conflicting studies on weight. One study showed that overweight people did better than normal weight, thin, or obese people. Others have shown that overweight (or maybe obese; can't remember) people did better after surgery. But other studies have shown that overweight people do worse on various things.

Weight gain tends to be normal as we age, especially in older women, so maybe there's a reason. Fat does protect the bones when we fall. But it also increases insulin resistance, especially fat around the middle.

So, as with so many things, there are pluses and minuses to being overweight. My gut feeling is that not aiming to weigh the same as you weighed in college might be good, just a little overweight. However, my endo says all the very elderly people she sees in nursing homes are thin.

However, being morbidly obese is another matter. I think that's not a healthy state. The knees wear out from the stress of bearing so much weight. Other organs are stressed. And society hurls insults at obese people.

As you say, there are also racial differences in being overweight. Blacks are healthier than whites when they're overweight. I don't know about morbidly obese. Asian Indians have metabolic consequences from just a little weight gain.

Re the ketosis-prone people, it's seems odd. But note that they said the people who improved control gained weight, not that gaining weight improved control. When you're wildly out of control (as you have to be to spill ketones), you're spilling a lot of glucose into the urine, and when you regain control, you're not, so in essence, you're eating more calories.

Type 1s have often lost weight on diagnosis and start gaining it when they get treatment.

Re Ned's package, are you referring to his software beta testing? I signed up but can't use it because I have Excel 2000. How did you know I got it?

Correlation doesn't prove causation. For example, diabetes risk is high among Native Americans. Native Americans mostly have dark hair. Hence you could do a study and show that dark hair and diabetes were correlated.

But dark hair does not cause diabetes and diabetes does not cause dark hair. They just track together in some populations.

Am I missing something in that paper? I confess, I read it quickly. So much piled up on my desk demanding that I read it.

Michael, there are so many factors involved in diabetes that it can be mind-boggling, and you're dealing with a less common form, with less research.

I just try to read as much as I can, from all points of view, and hope that eventually something will click.

One area I think has been neglected is the alpha cells and glucagon, which are too active in type 2. I don't know about ketosis-prone diabetes.