If the study showed that incredibly "fit" people who smoked occasionally and had a sightly elevated blood pressure and/or cholesterol had a "lower adjusted death rate" (what is that, exactly?) than people who sat in front of the computer all day but had none of the above factors, that would be one thing.
If the study showed that slightly "fit" people who smoked heavily and had an extremely elevated blood pressure and/or cholesterol had a "lower adjusted death rate" than people who sat in front of the computer all day but had none of the above factors, that would be something else.
The same could be asked of the fitness vs obesity results. How fit and how obese?
Also, have the results of the study ever been replicated? Or do they conflict with the results of other studies?