OBJECTIVE: To compare metabolic syndrome (MS) score with the 10-year-Framingham risk score (FRS) to predict the occurrence of cardiovascular disease (CVD). METHODS: MS score for prediction of CVD was developed based on the 10-year FRS. Cox proportional hazard model and receiver-operating characteristic (ROC) curves were used to compare the predictive effects, based on data from a cohort study on the prevention of multiple metabolic disorders and MS in Jiangsu province. RESULTS: Area under the curve (AUC) increased after changing MS components into continuous variables. AUC of MS score/MS components aggregation was 0.70/0.65, P < 0.05 and sensitivity of MS score/MS components aggregation was 80.5%/74.4% for a given specificity. After mutually adjusted risk factors of MS score and the FRS, when age was exclusively excluded, AUC of the FRS decreased from 0.78 to 0.65 (P < 0.05). However, when age was included, the AUC of MS score increased to 0.78 (sensitivity of MS score including the age/the FRS: 90.2% vs. 87.8%); In Cox proportional hazards multiple risk factors analysis, MS score including age appeared greater association with CVD than FRS on the same exposed subjects. CONCLUSION: The new developed MS score with age included was a valid tool for predicting CVD and its predictive ability was as good as the FRS.
OBJECTIVE: To compare metabolic syndrome (MS) score with the 10-year-Framingham risk score (FRS) to predict the occurrence of cardiovascular disease (CVD). METHODS: MS score for prediction of CVD was developed based on the 10-year FRS. Cox proportional hazard model and receiver-operating characteristic (ROC) curves were used to compare the predictive effects, based on data from a cohort study on the prevention of multiple metabolic disorders and MS in Jiangsu province. RESULTS: Area under the curve (AUC) increased after changing MS components into continuous variables. AUC of MS score/MS components aggregation was 0.70/0.65, P < 0.05 and sensitivity of MS score/MS components aggregation was 80.5%/74.4% for a given specificity. After mutually adjusted risk factors of MS score and the FRS, when age was exclusively excluded, AUC of the FRS decreased from 0.78 to 0.65 (P < 0.05). However, when age was included, the AUC of MS score increased to 0.78 (sensitivity of MS score including the age/the FRS: 90.2% vs. 87.8%); In Cox proportional hazards multiple risk factors analysis, MS score including age appeared greater association with CVD than FRS on the same exposed subjects. CONCLUSION: The new developed MS score with age included was a valid tool for predicting CVD and its predictive ability was as good as the FRS.