Hospital incidental diagnosis of diabetes: A population study

Aims To identify incidental previously unrecognized diabetes (IPUD) among hospitalized patients and corresponding mortality risk in comparison with individuals with known diabetes (KDM). Methods Out of 214,991 individuals discharged in year 2011 from all hospitals of Tuscany, Italy we retrospectively identified IPUD as individuals with no known diabetes and/or previous antidiabetic medication, receiving at least two prescriptions of glucose-lowering-drugs over the next 6 months after discharge. Two-year (2012-2013) adjusted mortality risk was tested by a Cox-regression-analysis, comparing IPUD and KDM patients with at least one hospital admission in 2011. Results 974 patients with IPUD (375.6 × 100,000 hospitalized people) have been identified. IPUD risk was associated with aging, male gender and greater burden of co-morbidities, was higher in migrants of non-Italian ancestry and was reduced among patients of family physicians adhering to guidelines resulting in a proactive model of care delivery. In IPUD patients alive at 1st January 2012, (n = 865) the adjusted risk of two-year mortality was similar to that of KDM subjects (HR = 1.08; 95% CI: 0.92-1.26; p = NS). Conclusions IPUD occurs more commonly in older male subjects, migrants of non-Italian ancestry, and among patients of physicians non-adhering to a shared diabetes care model. People with IPUD have similar two-year-mortality risk compared with KDM individuals