You have tapped into a logical fallacy. A is associated with B. B is associated with C. This in no way indicates that A causes C. It just doesn’t. You have to show that A and C are associated, then you have to determine causation. You can’t skip steps.

I see this all the time in nutrition “science”, and it is just bad science.