BMC Endocr Disord

BackgroundTo evaluate the U.K. Prospective Diabetes Study (UKPDS) and Framingham risk equations for predicting short-term risk of coronary heart disease (CHD) events among adults with long-standing type 2 diabetes, including those with and without preexisting CHD.MethodsProspective cohort of U.S. managed care enrollees aged \ue2\u2030\ua5 18\ue2\u20ac\u2030years and mean diabetes duration of more than 10\ue2\u20ac\u2030years, participating in the Translating Research into Action for Diabetes (TRIAD) study, was followed for the first occurrence of CHD events from 2000 to 2003. The UKPDS and Framingham risk equations were evaluated for discriminating power and calibration.ResultsA total of 8303 TRIAD participants, were identified to evaluate the UKPDS (n\ue2\u20ac\u2030=\ue2\u20ac\u20305914, 120 events), Framingham-initial (n\ue2\u20ac\u2030=\ue2\u20ac\u20305914, 218 events) and Framingham-secondary (n\ue2\u20ac\u2030=\ue2\u20ac\u20302389, 374 events) risk equations, according to their prior CHD history. All of these equations exhibited low discriminating power with Harrell\ue2\u20ac\u2122s c-index