Epidemiological studies and reviews can demonstrate associations between variables, but an association is not necessarily causal. Sadly, there is no sufficient way to prove for certain that an association between a factor and a disease is a causal relationship. In 1965, Austin Bradford Hill proposed a set of criteria for assessing the causal nature of epidemiological relationships; he based these in part on Koch’s postulates. The criteria have been revised many times, so you may find different versions with different numbers of criteria.