Causation: the elusive grail of epidemiology

Abstract

The paper discusses the evolving concept of causation in epidemiology and its potential interaction with logic and scientific philosophy. Causes are contingent but the necessity which binds them to their effects relies on contrary-to-fact conditionals, i.e. conditional statements whose antecedent is false. Chance instead of determinism plays a growing role in science and, although rarely acknowledged yet, in epidemiology: causes are multiple and chancy; a prior event causes a subsequent event if the probability distribution of the subsequent event changes conditionally upon the probability of the prior event. There are no known sufficient causes in epidemiology. We merely observe tendencies toward sufficiency or tendencies toward necessity: cohort studies evaluate the first tendencies, and case-control studies the latter. In applied sciences, such as medicine and epidemiology, causes are intrinsically connected with goals and effective strategies: they are recipe which have a potential harmful or successful use; they are contrastive since they make a difference between circumstances in which they are present and those in which they are absent: causes do not explain event E but event E rather than even F. Causation is intrinsically linked with the notion of "what is pathological". Any definition of causation will inevitably collapse into the use made of epidemiologic methods. The progressive methodological sophistication of the last forty years is in perfect alignment with a gradual implicit overhaul of our concept of causation.