Evaluation des Gesundheitsmonitorings in Bayern mit computer-assistierten Telefoninterviews (CATI) durch den Vergleich mit dem Bundes-Gesundheitssurvey 1998 des Robert Koch-Instituts

Between June 1999 and May 2000, 2051 residents of Bavaria were interviewed on health and health-related behaviour. The objective of the study was to assess the representativeness and validity of data acquired by computer-assisted telephone interviewing (CATI) as compared to results of the German National Health Examination Survey 1998. The following variables showed excellent agreement: subjective health status, lifetime disease prevalences, such as myocardial infarction, stroke, cancer, asthma, diabetes, hypertension, hypercholesterolaemia, utilisation of medical services, and smoking habits (prevalence of present and former smoking, average cigarette consumption per day). Minor differences were found with respect to the frequency of consulting naturopathic doctors. Self-reported anthropometric data from the present study differed considerably from the results of the objective measurements during the German Health Survey. We conclude that, in terms of data quality, CATI compares well with traditional methods such as face-to-face interviewing or mailed questionnaires. In view of several practical advantages of the method, we recommend CATI as a future basic methodology of a German health surveillance system consisting of different coordinated health data collections.