I agree, there are too many confounders to draw a conclusion from this, a lot of the things in this study could be just down to random chance.