Introduction: Perinatal studies provide an excellent database for public health research, to date this potential has rarely been used, however. Taking the example of the Perinatal Study in Bavaria, the objective is to demonstrate the pros and cons of this database. As it includes only very few variables on socio-economic status, an additional variable is calculated assessing the socio-economic status of the community where the mother lives. This is rarely done in Germany, and as far as we know the procedure proposed here has not been applied before.
Methods: The analyses are based on the data from 2004. They focus on three dependent variables: number of prenatal screenings, maternal smoking during pregnancy, birth weight of the baby. The following independent variables are included as well: age of the mother, nationality of the mother (e.g., German, Mediterranean countries), single mother (yes/no), occupational status of the mother, community where the mother lives (4- or 5-digit postal code). The socio-economic status of the community is assessed by the poverty rate, linking two other datasets, one for transferring the postal codes to community names, the other providing information per community. The multivariate analyses are conducted by logistic regressions.
Results: Information was available from about 76 000 births. Concerning the variable 'few prenatal screenings', the analyses show an increased risk for mothers from Eastern Europe and from the Mediterranean countries, for single mothers and for mothers with low occupational status. The risk factor 'maternal smoking during pregnancy' is increased for mothers from the Mediterranean countries and for single mothers. It is especially high, however, for low status blue collar workers: compared with white collar workers their smoking prevalence is 4.67-times (large cities) or even 6.14-times (smaller communities) higher. The risk factor 'low birth weight of the baby' is again increased for single mothers and for mothers with low occupational status. An association with the poverty rate is mainly seen for the variable 'maternal smoking during pregnancy', with higher smoking prevalences in the poor communities.
Discussion: The results demonstrate that the data from the perinatal studies are important for public health research. Concerning the risk factors analysed here, large social differences can be observed. In order to show time trends, it would be important to repeat these analyses on a routine basis. From a methodological point of view, it can be stressed that regional differences in health and health care have rarely been looked at in Germany, and that the procedure proposed here provides a new starting point for closing this research gap.