[Population-based health monitoring via computer-assisted telephone interviews in Bavaria]

From June 1999 until May 2000, 2051 computer-assisted telephone interviews concerning health and health related behaviours were assessed in Bavarian households. The aim of the study was the evaluation of the validity and representativity of this data by a comparison of selected variables from the Bavarian Mikrozensus-survey 1999. The distribution patterns of the marital status, the size of the households, the gainful employment, the household-netto-income, the school qualification and of the smoking status correspond well with those of the Mikrozensus 1999. As often found in questionnaire based surveys, a disproportionately high rate of participation of highly educated persons was observed. This led to an effect only in terms of a too small rate of persons with a low household-netto-income. Within the study, two different designs were compared. The commitment of the telephone numbers to a gender for an equal quotation of men and women led to an under-representation of single-households and therefore also of unmarried persons, but it had no effect on the other characteristics. The results of this study show, that the method of computer assisted-telephone interviews is a valid and cheap basis for the establishment of an exhaustive health surveillance system in Germany.