Any study that mashes together processed and unprocessed meat in epidemiological setting is next to meaningless in my opinion. You can associate basically anything this way.
Guess where read meat and processed meat intersect? McDonald’s, for example. Now tell me that eating sirloins kills me.

You know it’s not even E2EE by default, and when it is it uses a homegrown algo that is not exactly well spoken of? (at least V1)