Developing a Prediction Model for 7-year and 10-year All-cause Mortality Risk in Type 2 Diabetes Using a Hospital-based Prospective Cohort Study

Figure S1: Proposed study scenario and application; Figure S2: Flow diagram for study subjects; Figure S3(A): Proportional hazards assumption checking for the 7-year follow- up; Figure S3(B): Proportional hazards assumption checking for the 10-year follow- up; Figure S4: Flowchart for model cross-validation; Figure S5(A): Cumulative all-cause mortality for the 7-year follow-up based on 9101 validation subjects; Figure S5(B): Cumulative all-cause mortality for the 10-year follow-up based on 9101 validation subjects; Figure S6(A): ROC and AUC for cross-validation at the 2-year time point based on the 7-year follow-up; Figure S6(B): ROC and AUC for cross-validation at the 4-year time point based on the 7-year follow-up; Figure S6(C): ROC and AUC for cross-validation at the 6-year time point based on the 7-year follow-up. Figure S7(A): ROC and AUC for cross-validation at the 2-year time point based on the 10-year follow-up; Figure S7(B): ROC and AUC for cross-validation at the 4-year time point based on the 10-year follow-up; Figure S7(C): ROC and AUC for cross-validation at the 6-year time point based on the 10-year follow-up; Table S1: Distribution of patient characteristics and risk factors by sex; Figure S7(D): ROC and AUC for cross-validation at the 8-year time point based on the 10-year follow-up; Table S1: Description of variables, candidate predictors and definition in this study; Table S2: Distribution of patient characteristics and risk factors by sex; Table S3: Numbers and causes of deaths by the 7-year and 10-year follow-ups; Table S4: AIC for model selection; Table S5: Distribution of patient characteristics and risk factors by training and validation data; Table S6: Harrell's C statistic by sex based on 7-year anFigure S1: Proposed study scenario and application; Figure S2: Flow diagram for study subjects; Figure S3(A): Proportional hazards assumption checking for the 7-year follow- up; Figure S3(B): Proportional hazards assumption checking for the 10-year follow- up; Figure S4: Flowchart for model cross-validation; Figure S5(A): Cumulative all-cause mortality for the 7-year follow-up based on 9101 validation subjects; Figure S5(B): Cumulative all-cause mortality for the 10-year follow-up based on 9101 validation subjects; Figure S6(A): ROC and AUC for cross-validation at the 2-year time point based on the 7-year follow-up; Figure S6(B): ROC and AUC for cross-validation at the 4-year time point based on the 7-year follow-up; Figure S6(C): ROC and AUC for cross-validation at the 6-year time point based on the 7-year follow-up. Figure S7(A): ROC and AUC for cross-validation at the 2-year time point based on the 10-year follow-up; Figure S7(B): ROC and AUC for cross-validation at the 4-year time point based on the 10-year follow-up; Figure S7(C): ROC and AUC for cross-validation at the 6-year time point based on the 10-year follow-up; Table S1: Distribution of patient characteristics and risk factors by sex; Figure S7(D): ROC and AUC for cross-validation at the 8-year time point based on the 10-year follow-up; Table S1: Description of variables, candidate predictors and definition in this study; Table S2: Distribution of patient characteristics and risk factors by sex; Table S3: Numbers and causes of deaths by the 7-year and 10-year follow-ups; Table S4: AIC for model selection; Table S5: Distribution of patient characteristics and risk factors by training and validation data; Table S6: Harrell's C statistic by sex based on 7-year and 10-year follow-ups.d 10-year follow-ups