The gray area is the problem. There's tons and tons of food that is likely healthy, but would meet the "ultra-processed" definition due to including certain ingredients. If those get lumped into the pile, then what is the study actually even measuring? Nothing useful in my view. If you want to make a case that a specific ingredient is harmful by itself (e.g., preservative X added to a food, all else equal, is bad) then I think that's a much stronger argument.

Right now all these studies seem to use broad strokes on the definition of ultra processed that make me think that it would be impossible to accurately remove confounding variables from the study.