Background and objective: During the last decade, recommendations for coronary heart disease prevention have emphasized the need of an assessment of coronary risk. The objective of this study was to assess the usefulness of the Framingham risk function for the estimation of coronary risk in our population, assessing the accuracy of the coronary risk as a diagnostic test (assuming that a coronary risk >20% is a positive diagnosis test of coronary disease).
Patients and method: Prospective case-control study, carried out in a public tertiary hospital.
Results: 138 cases were identified and matched, according to age and sex, with 223 controls. The level of risk > or =20% was the one which better classified the patients (70% of patients correctly classified as cases and controls) with an area under the ROC curve of 73.5% (moderate accuracy). Patients with a coronary risk >20% had a six times higher probability of being a case than a control (OR=6.09; 95% CI, 3.30-11.22).
Conclusions: The utilization of the Framingham function for the prediction of coronary disease is a useful method to identify high risk patients. It is recommended to use it in clinical practice until we have our own model derived from populations with a baseline risk similar to ours allowing to predict the risk with higher accuracy.