Abstract 17705: Overestimation by Common Cardiovascular Risk Scores Amongst Primary Prevention Patients in the Cholesterol Treatment Trialists’ Collaboration (CTTC) Database
Introduction: Cardiovascular risk calculators are increasingly used to guide treatment recommendations. We examined the accuracy of popular risk calculators amongst primary prevention, control arm patients of statin trials in the CTTC database
Methods: Using individual patient data we compared our previously modelled and highly correlated (r=0.997) five year derived vascular risk of participants in the control arms of 22 randomised statin trials without known vascular disease at baseline with 5 year equivalent Framingham, QRisk2 and Pooled Cohort Equations (PCE) scores for each patient.
Results: Data was available for 31050 patients with a CTTC derived median five year absolute risk of 6.2% (IQR 4.0-10.2). There was a similar paired sample correlation between CTTC derived risk and each risk score (Framingham 0.68, QRisk2 0.65, PCE 0.68, all p10%) was 85% for Framingham, 64% for QRisk2 and 55% for PCE scores. Fewer than half of all patients had an absolute difference between predicted and CTTC five year risk of less than 2.5% (Table 2).
Conclusions: QRisk2 and Pooled Cohort Equations scores outperform Framingham scores when applied to primary prevention control arm patients in the Cholesterol Treatment Trialists’ database. However for all risk calculators accuracy is limited and risk overestimation occurs in the majority of individuals. This may have implications when predicted risk levels are used to guide recommendations and estimate cost effectiveness for cardiovascular therapies.
Author Disclosures: J. Fulcher: Honoraria; Modest; Pfizer, Bristol Myers Squibb. R. O'Connell: None. A. Keech: Honoraria; Modest; Astra-Zeneca, Bristol Myers Squibb, Eli Lilly, Pfizer, Sanofi. Honoraria; Significant; Abbott.
- © 2015 by American Heart Association, Inc.