The ability to accurately predict coronary heart disease (CHD) is useful in designing health intervention programs. The predictive model resulting from the Framingham Heart Study is a widely accepted and validated model for estimating CHD risk. However, the Framingham model is limited since it was started long before many newer risk factors were identified.
Based on a novel algorithm called Synthesis Analysis, a new evidence-based CHD prediction model was developed. This model includes more risk factors than the Framingham model. First, the univariate association of each new risk factor with the clinical endpoint is assessed using meta-analysis. Then, the association of the complete set of variables with the endpoint is determined using Synthesis Analysis. This method accounts for the colinearity within the array of predictive variables.
The risk probabilities predicted by the new model were compared with the Framingham model using the First National Health and Nutritional Examination Survey (NHANES-I) data. Five year’s CHD mortality was used as the outcome for comparison. The correlation between the risk probabilities predicted by the new model and the Framingham model was 0.962. The sensitivities of the new model vs. the Framingham model were 0.41 vs. 0.34 in female, and 0.20 vs. 0.19 in male. Specificity of 0.9 was observed in both model and in both gender. Areas under the ROC curves of the new model vs. the Framingham model were 0.835 vs. 0.807 in female and 0.727 vs. 0.718 in male. The new evidence-based predictive model consistently outperforms the Framingham model.
See www.knowyournumber.comLearning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to: 1. Acquire the knowledge of a new technology in the evidence-based medicine 2. Assess the important risk factors to predict CHD morbidity and mortality 3. Articulate the procedure for assessing the risks of CHD 4. Evaluate and apply the study results to health promotion and/or other preventive programs
Keywords: Evidence Based Practice, Risk Assessment
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: BioSignia, Inc.
I have a significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.
Relationship: employment