Abstract 16585: Prediction of Atherosclerotic Cardiovascular Disease Mortality in a Nationally Representative Cohort Using Set of Risk Factors From Pooled Cohort Risk Equations
Introduction: The American College of Cardiology/American Heart Association Pooled Cohort risk equations were developed to estimate atherosclerotic cardiovascular disease (ASCVD) risk. It is unclear if the set of risk factors might predict ASCVD mortality in a nationally representative cohort. Hypothesis: We assessed the performance of the set of risk factors developed by the Pooled Cohort equations in the prediction of ASCVD mortality by sex and race-ethnicity. Methods: We used the Third National Health and Nutrition Examination Survey Linked Mortality File (NHANES III 1988-2006, n=7,042) and included participants 40-79 years of age without CVD at baseline. We used Cox proportional hazard models to estimate the predicted probability of ASCVD death at 10-year by sex and race-ethnicity (non-Hispanic white (NHW), non-Hispanic black (NHB), and Mexican American (MA)). We calculated the discrimination and calibration for each sex-race-ethnicity model. Results: We documented 409 ASCVD deaths during 62,335 person years of follow-up (212 men and 197 women). The Pooled Cohort equations’ set of risk factors demonstrated moderate discrimination ability for ASCVD mortality with modified c-statistics of 0.658 (95% CI: 0.602-0.713), 0.661 (0.589-0.732), and 0.703 (0.624-0.782) for NHW, NHB and MA men respectively. The corresponding c-statistics for women were 0.695 (0.631-0.760), 0.695 (0.624-0.765), and 0.687 (0.588-0.786). Modified Hosmer-Lemeshow χ 2 suggested adequate calibration for NHW men and women (p-value 0.13 and 0.06 respectively). The calibration was borderline adequate for NHB women with p-value 0.04, and inadequate for MA men and women with p-value <0.001 and 0.004 respectively. The moderate discrimination and inadequate calibration for some prediction models might be partly due to the limited sample size. Conclusions: In conclusion, in this nationally representative cohort, the set of risk factors from the Pooled Cohort equations perform adequately to predict 10-year ASCVD mortality for NHW men and women with observed and predicted 10-year ASCVD deaths being similar, and demonstrated moderate discrimination for all sex-race/ethnicity groups.