Or some other correlated factor with no causative relationship. Like, obesity being tied to generally worse health outcomes so where antibiotics were less commonly used obese people died from infections more often. Like horse ownership correlating with better overall health, but the causative factor is “being wealthy enough to own horses.”