To improve efficiently the outcome of population based screening for diabetes, a preceding questionnaire could be useful to reduce the number of subjects for the chemical diabetes test. During the Austrian Diabetes-Bus-Tour, a questionnaire, proved in the U.S., was completed in 16537 subjects without known diabetes. Further, the questionnaires were analysed by two different aspects: following the original American system and secondly, a system pronouncing the hereditary component of type 2 diabetes. In comparison of the measured blood glucose values und the questionnaire, in 28% of investigated subjects, at least 50% of expected newly diagnosed type 2 could be identified by the original American system. The point system with hereditary emphasis was not equally useful. [ABSTRACT FROM AUTHOR]