Abstract 15997: Pooled Cohort Risk Equations: A Comparison of Their Predictions With Five Other Cardiovascular Risk Scores in Five Peruvian Sites

Abstract

Introduction: Cardiovascular disease (CVD) risk scores are used to estimate an individual’s risk of developing a disease or death from a cardiovascular event. Recently, the American College of Cardiology/American Heart Association (ACC/AHA) introduced the Pooled Cohort risk equations (ACC/AHA model). It is important to know how comparable CVD risk predictions are in low-middle income countries (LMIC). Hypothesis: ACC/AHA model has a poor concordance with any other CVD risk score. Methods: We used secondary data from two Peruvian, age and sex-matched, population-based studies across five geographical sites. The ACC/AHA model was compared to five other CVD risk prediction tools: two versions of the Framingham Risk Score (FRS-Lipids and FRS-BMI), Reynolds Risk Score (RRS), four versions of the Systematic Coronary Risk Evaluation (SCORE 1-4), World Health Organization risk chart (WHO), and Lancet Chronic Disease risk chart (LCD). We calculated predicted risk as a continuous variable and used Lin’s concordance correlation coefficient (CCC). We also compared the high predicted risk prevalence between all the scores using the cut-off levels suggested by each score’s guidelines. Results: We included 2183 subjects in the risk scores age range of 45-65 years (mean age 54.3 (SD±5.6) years). CCC agreement values found in this study were generally poor. The highest concordance was observed between the ACC/AHA model and the risk scores derived from the Framingham Study (40% with FRS-BMI and 44% with FRS-Lipids). ACC/AHA model depicted the highest proportion of people with predicted high-risk of 10-year CVD, at 29.0% (95%CI 26.9-31.0%) and the same tendency was observed in all study sites. Conclusions: In Peruvian population-based samples, agreement between ACC/AHA model and five other CVD risk scores was generally poor. There is an urgent need to use an appropriate risk score for CVD in LMIC. In an ideal scenario, it would be significant to have a proper CVD risk score for LMIC.