What is interesting here is their finding that the two specific known major causes of obesity-linked mortality that they considered (c-reactive protein and high

What is interesting here is their finding that the two specific known major causes of obesity-linked mortality that they considered (c-reactive protein and high resting insulin -- effectively "obesity-linked diabetes") explained slightly more than all of the increased mortality associated with obesity. That suggests that there is no other material contributor to mortality from obesity[1], which is a somewhat unexpected result. -- [1] it is possible that there are negative contributions to mortality associated with obesity which could potentially "hide" other positive contributors