[Analytical epidemiology--case-control and cohort studies].

Background: "The study of the occurrence of disease" is a short but quite precise definition of epidemiology, as it includes both descriptive and analytic approaches. Whereas descriptive studies of the distribution of diseases and their potential causal factors may generate hypotheses about disease aetiology, causal inferences may most safely be drawn from analytical testing through properly designed observational studies.

Methods: The most commonly used observational designs are the retrospective case-control and the prospective cohort studies. In some respects the two designs complement each other. Drawing on some classic epidemiological studies, their main properties in terms of what questions they may answer, what their applicabilities are, as well as their main strengths and weaknesses, are discussed.