Incidence of type 1 diabetes in age groups above 15 years: facts, hypothesis and prospects for future epidemiologic research.

Although onset of type 1 diabetes can occur in adulthood, epidemiological data are scarce, limiting our potential to identify unknown determinants of the disease. Paucity of registries expanding the recruitment of incident cases up to adulthood, atypical clinical features of type 1 diabetes at onset, misclassification of type 1 as type 2 diabetes and little use of markers of β-cell autoimmunity represents major obstacles in studying the risk of type 1 diabetes in adults. New strategies in study design, data collection and analyses may overcome these problems in the future. Population-based surveys and registries including adulthood; use of etiological rather than clinical criteria to define type 1 diabetes; availability of electronic health records as prescription data sources to avoid missing data; and application of proper statistical methods will be instrumental to gain better insight on the epidemiology and natural history of the disease.