How does this proove the point? In the example given, the diabetic is expending more calories (through waste) than the non-diabetic... the model still holds valid. The diabetic has less net calories absorbed by their body.

>how do we know there aren't factors yet to be discovered?

We don't and can't (by definition), that's part of the scientific process. Part of that process requires us to find evidence to the contrary or provide a viable testable competing argument. We don't just waive our hands in the air and claim magic gremlins are making people skinny or fat since there are likely limits to human understanding and knowledge, especially at any given specific point on time.

Most the evidence out there, that I'm aware of, vehemently supports the CICO model and it makes sense as a complex system resting up very well studied chemical and physical phenomena. The only thing broken seems to be a general understanding of the model being discussed.