Abstract
INTRODUCTION: Accurate risk assessment is critical in cardiovascular (CV) prevention, yet physicians often underestimate CV risk, leading to inadequate preventive measures. AIM: This study evaluates the concordance between physician-perceived CV risk and calculated CV risk in a primary prevention setting. METHODS: This cross-sectional study included primary prevention patients from the Cardiology Outpatient Clinic of Caserta Hospital, Italy. Two independent cardiologists evaluated the physician-perceived risk, and a third resolved discrepancies. CV risk was calculated using SCORE2 for patients with 70 years or less and SCORE2-OP for those with more than 70 years. The concordance between perceived and calculated risks was assessed using Cohen's kappa coefficient. Multivariate logistic regression analysis was performed to examine the influence of risk estimation on achieving low-density lipoprotein cholesterol (LDL-C) targets recommended by the ESC. RESULTS: 389 patients had complete data for CV risk calculation. Physician-perceived risk categorized 8.7% of patients as low/moderate, 37.8% as high, and 53.5% as very-high risk. In contrast, calculated CV risk according to the SCORE2/SCORE2-OP classified 8% as low/moderate, 5.7% as high, and 86.4% as very-high risk. The concordance between perceived and calculated CV risk was poor (Cohen's kappa 0.208, p < 0.001). Underestimated patients reached LDL-C targets in 16% of cases, well-estimated in 34.5%, and overestimated in 76.9%. Statin use was significantly lower in patients with underestimated CV risk (29.2%) compared to well-estimated (50%) and overestimated (76.9%) groups (p < 0.001). Multivariate analysis showed that patients with overestimated risk were more likely to achieve LDL-C targets (OR 5.33, CI 1.33-21.42, p = 0.018), whereas underestimated patients were 47% less likely (OR 0.53, CI 0.3-0.93, p = 0.027). CONCLUSIONS: A significant discrepancy exists between physician-perceived and calculated CV risk, leading to risk underestimation in over one-third of patients. This underestimation is associated with lower LDL-C target achievement and reduced statin use.