Figure 4.

Risk Score Predictive Accuracy in Cancer Groups vs Matched Controls for Specific Outcomes

The AUC, balanced accuracy, and sensitivity for atrial fibrillation, heart failure, and stroke are compared between cancer groups and their matched control cohorts. Different outcome–risk score pairs are shown down the right-hand side, and different cancer groups are shown on the left-hand side. Predictive performance results for cancer survivors are displayed in red, with matched controls in dark blue. Cancer groups not significantly different from their noncancer control group are shown in the lighter color. AUC results are from time-dependent analyses accounting for the competing risk of death at the 10-year follow-up. Balanced accuracy and sensitivity are from static performance scoring of 10-year outcomes, with 95% CIs derived from bootstrapping and permutation testing with 1,000 replicates. Abbreviations as in Figures 1 and 2.