[Identification and sampling of people with migration background for epidemiological studies in Germany].

In 2009, 19.6% of the population of Germany either had migrated themselves or were the offspring of people with migration experience. Migrants differ from the autochthonous German population in terms of health status, health awareness and health behaviour. To further investigate the health situation of migrants in Germany, epidemiological studies are needed. Such studies can employ existing databases which provide detailed information on migration status. Otherwise, onomastic or toponomastic procedures can be applied to identify people with migration background. If migrants have to be recruited into an epidemiological study, this can be done register-based (e. g., data from registration offices or telephone lists), based on residential location (random-route or random-walk procedure), via snowball sampling (e. g., through key persons) or via settings (e. g., school entry examination). An oversampling of people with migration background is not sufficient to avoid systematic bias in the sample due to non-participation. Additional measures have to be taken to increase access and raise participation rates. Personal contacting, multilingual instruments, multilingual interviewers and extensive public relations increase access and willingness to participate. Empirical evidence on 'successful' recruitment strategies for studies with migrants is still lacking in epidemiology and health sciences in Germany. The choice of the recruitment strategy as well as the measures to raise accessibility and willingness to participate depend on the available resources, the research question and the specific migrant target group. © Georg Thieme Verlag KG Stuttgart · New York.