[Epidemiologic challenges in rare diseases].

Rare diseases, often called "orphan diseases", are a special challenge for epidemiologic research. Apart from the mere logistic effort for sample collection, there are considerable implications in statistical methodology. Usually one will not find enough cases of an orphan disease in a random sample from the population at risk. Furthermore, random error plays a more important role for decreasing probability of disease. Critical issues related to total population sampling, active and passive surveillance and capture-recapture methods are discussed. Challenges in risk factor research and related to therapeutic or preventive trials are presented. Examples from epidemiologic practice are given.