ABSTRACT
The basic question which the epidemiologic study tries to answer is this: is there an association between a specific exposure (in the widest sense of the word) and an outcome. If so, can this association be explained by chance, by the action of confounding factors (for a discussion see section “Confounding”), or by bias? Such associations can be estimated with different techniques, the two classical ones being the case-control (or case-referent) and cohort approaches. The first compares exposures in “cases” (the abnormal reproductive outcome under study) and “controls” (lacking that abnormal reproductive outcome). With this design, many exposures can be studied simultaneously in relation to the selected outcome. The cohort approach studies reproductive outcome in parents who were exposed in a specific way and the outcomes are compared with that of unexposed cohorts or the total population. With this approach, many different outcomes can be studied after a specific exposure. In both the methods, the following two variables must be defined: reproductive outcome(s) and exposure(s).
