Another factor that bugs me about vegetarian-related studies like this is that they're almost always correlative and thus self-selective, but never causative.

You can just as easily study the US population of Non-White people and conclude that "Being Asian is predictive of greater income, Black of lower income". Because all the self-sorting of various other non-racial related strata already happened (the fact that whites only let in the super STEM profession Asian immigrant whitecollar classes, while Black Americans went through slavery AND a century+ of land theft before those Asians even arrived).
Obviously, concluding that "being Black is inherently inferior" from this would be factually wrong.