Supplementary Table 4. Association, discriminative capacity and calibration statistics of PRECISE-DAPT score and PARIS bleeding score for predicting clinically relevant bleeding at follow-up.
| PRECISE-DAPT score | PARIS bleeding score | |
|---|---|---|
| HR (95% CI) ＊ | 1.036 (1.024-1.048) | 1.131 (1.074-1.191) |
| P ＜0.001 | P ＜0.001 | |
| Hosmer-Lemeshow χ 2 | 10.84 | 4.45 |
| P = 0.211 | P = 0.487 | |
| Greenwood-Nam-D’Agostino χ 2 | 7.97 | 4.40 |
| P = 0.537 | P = 0.623 | |
| Harrel’s C Statistic | 0.585 (0.550-0.619) | 0.580 (0.546-0.615) |
＊ Using the total risk scores as a global prognostic indicator (i.e., as a continuous variable). For Greenwood-Nam-D’Agostino test, patients were categorized into deciles based on predicted probability of incident clinically relevant bleeding. Greenwood-Nam-D’Agostino test for calibration across deciles requires at least 2 events in each group; thus, deciles were collapsed as needed if ＜2 clinically relevant bleeding events occurred.