Validity of an adaptation of the Framingham cardiovascular risk function: the VERIFICA study

• DOI: 10.1136/jech.2005.038505
• Publication Date: 2006-12-20T23:43:19Z
• Authors: J. Marrugat; I. Subirana; E. Comin; C. Cabezas; J. Vila; R. Elosua; B.-H. Nam; R. Ramos; J. Sala; P. Solanas; F. Cordon; J. Gene-Badia; R. B D'Agostino
• Keywords: Framingham Heart Study; Cumulative incidence

<b>Background:</b> To assess the reliability and accuracy of Framingham coronary heart disease (CHD) risk function adapted by Registre Gironí del Cor (REGICOR) investigators in Spain. <b>Methods:</b> A 5-year follow-up study was completed 5732 participants aged 35–74 years. The adaptation consisted using average population factor prevalence cumulative incidence observed Spain instead those from a Cox proportional hazards model. Reliability estimating were tested with area under curve comparison goodness-of-fit test, respectively. <b>Results:</b> Kaplan–Meier CHD during 4.0% men 1.7% women. original REGICOR estimates 10.4% 4.8%, 3.6% 2.0%, REGICOR-adapted function's estimate did not differ cumulated (goodness fit men, p = 0.078, women, 0.256), whereas all differed significantly (p&lt;0.001). Reliabilities best model data similar (area receiver operator characteristic 0.68 0.69, respectively, 0.273), fitted better women (0.73 0.81, p&lt;0.001). <b>Conclusion:</b> to local characteristics accurately reliably predicted for patients years, contrast function, which consistently overestimated actual risk.