New risk prediction model of coronary heart disease in participants with and without diabetes: Assessments of the Framingham risk and Suita scores in 3-year longitudinal database in a Japanese population

Figures

• 

  Figure 1. ROC curve of Suita scores (TC and LDL-C), FRS (TC), and NCEP-ATPIII FRS (TC) for coronary heart disease (CHD) prediction in total participants, men and women (a) and in men and women with and without diabetes. (b) Receiver operating characteristic (ROC) curves in Suita scores total cholesterol (TC, black line) and low-density-lipoprotein cholesterol (LDL-C, yellow line), Framingham risk score (FRS) TC (FRS TC, red line), and NCEP-ATPIII FRS TC (green line) are shown. The area under the curve (AUCs) of new-onset coronary heart disease (CHD) and cutoff values (sensitivity, 1 – specificity) and P values are shown in the lower left panel. P values for model comparisons are also shown in the lower right panel.

• 

  Figure 2. Decision curve analysis (DCA) for coronary heart disease (CHD) prediction in total participants, men and women (a) and in men and women with and without diabetes. (b) Horizontal dotted lines are net benefit of treating no participants, assuming that all would not develop CHD; gray line is net benefit of treating all participants similarly, assuming that all would develop CHD; net benefit of treating participants based on Suita score (TC) is black line and those based on FRS (TC) is red line. Dotted lines are 95% confidential intervals.

• 

  Figure 3. ROC curves of FRS, Suita score (TC) and its modified scores for coronary heart disease (CHD) prediction in total participants, men and women (a) and in men and women with and without diabetes. (b) Receiver operating characteristic (ROC) curves in FRS (light green lines), Suita score (TC) (black lines), New risk prediction model 1 (orange lines), Suita score (TC) new coefficients (blue lines), New risk prediction model 2 (light blue lines), and New risk prediction model 3 are shown. The area under the curve (AUCs) of new-onset coronary heart disease (CHD) and cutoff values (sensitivity, 1 – specificity) and P values are shown in the lower left panel. P values for model comparisons are also shown in the lower right panel. New risk prediction model 1: Suita score (TC) + new covariates*, New risk prediction model 2: Suita score (TC) new coefficients + new covariates*, New risk prediction model 3: Suita score (TC) new coefficients + new covariates 2**, *new covariates: Waist circumference (≥85 cm) in men and proteinuria (≥1+) in women were included, **new covariates 2: Triglyceride, drug of diabetes, hypertension, and dyslipidemia, waist circumference (≥85 cm), and proteinuria (≥+−)…

• 

  Figure 4. Decision curve analysis (DCA) for coronary heart disease (CHD) prediction in men and women with and without diabetes. Horizontal dotted lines are net benefit of treating no participants, assuming that all would not develop CHD; gray line is net benefit of treating all participants similarly, assuming that all would develop CHD; net benefit of treating participants based on Suita score (TC) is black line and those based on New risk prediction model 1 (orange), Suita score (TC) new coefficients (blue lines), New risk prediction model 2 (light blue lines), and New risk prediction model 3 (light pink lines)are also shown. Dotted lines are 95% confidential intervals. (New risk prediction model 1: Suita score (TC) + new covariates*, New risk prediction model 2: Suita score (TC) new coefficients + new covariates*, New risk prediction model 3: Suita score (TC) new coefficients + new covariates 2**, *new covariates: Waist circumference (≥85 cm) in men and proteinuria (≥1+) in women were included, **new covariates 2: Triglyceride, drug of diabetes, hypertension, and dyslipidemia, waist circumference (≥85 cm), and proteinuria (≥+−) in men, and triglyceride, drug of diabetes, hypertension,…