Yeah, I was most concerned about controlling for BMI, since ~all of the prior evidence for sex differences didn't/couldn't do it.
They controlled for BMI in cohort A, and performed an adjusted analysis in cohort B, so I'm somewhat satisfied on that front.