Small area variation in childhood diabetes mellitus in Austria: links to population density, 1989 to 1999.

We analyzed the associations of environmental factors with the regional distribution of Type 1 diabetes mellitus in Austria. All newly diagnosed cases (n=1449) from 1989 to 1999 were allocated to districts using the postal code. Nitrate content of the water was measured by the Austrian Federal Environmental Agency. Data on infant mortality, population density, and percentage of employment by industry were derived from Statistics Austria. An inverse effect was seen between the proportion of children younger than 15 years of age and the risk ratio (P<.01). Infant mortality, population density, and percentage of persons with employment in industry were not of significant influence. The mean nitrate level was positively associated (P=.07). In regions with a higher percentage of children younger than 15 years of age, fewer children developed diabetes, which is in agreement with the observation that early social mixing is a protective factor. Nitrate levels may have a confounding effect.