SHR-weighted scoring for the development of the CoBRA score
| . | SHR . | 95% CI . | P value . | Scoring∗ . |
|---|---|---|---|---|
| Age ≥55 y | 1.81 | 1.54-2.14 | <.001 | 2 |
| HCT-CI ≥3 | 1.51 | 1.29-1.76 | <.001 | 2 |
| Male recipient | 1.40 | 1.21-1.63 | <.001 | 1 |
| GVHD prophylaxis other than Tac/MTX | 1.36 | 1.16-1.58 | <.001 | 1 |
| PS 2-4 | 1.29 | 1.03-1.62 | .028 | 1 |
| HLA mismatch ≥2 | 1.22 | 0.94-1.59 | .14 | 1 |
| Refined DRI high-risk | 1.18 | 1.02-1.36 | .027 | 1 |
| MAC | 1.15 | 0.98-1.35 | .088 | 1 |
| CD34+ cell doses <0.82 × 105/kg | 1.12 | 0.98-1.29 | .10 | 1 |
| . | SHR . | 95% CI . | P value . | Scoring∗ . |
|---|---|---|---|---|
| Age ≥55 y | 1.81 | 1.54-2.14 | <.001 | 2 |
| HCT-CI ≥3 | 1.51 | 1.29-1.76 | <.001 | 2 |
| Male recipient | 1.40 | 1.21-1.63 | <.001 | 1 |
| GVHD prophylaxis other than Tac/MTX | 1.36 | 1.16-1.58 | <.001 | 1 |
| PS 2-4 | 1.29 | 1.03-1.62 | .028 | 1 |
| HLA mismatch ≥2 | 1.22 | 0.94-1.59 | .14 | 1 |
| Refined DRI high-risk | 1.18 | 1.02-1.36 | .027 | 1 |
| MAC | 1.15 | 0.98-1.35 | .088 | 1 |
| CD34+ cell doses <0.82 × 105/kg | 1.12 | 0.98-1.29 | .10 | 1 |
SHR, subdistribution hazard ratio.
We constructed the CoBRA score using the backward elimination method. Eliminating disease and CFU-GM from the multivariable model with all aforementioned variables could result in a decreased AIC. Thus, we determined to exclude these factors from the model, and performed a multivariable analysis for NRM using the remaining covariates. Finally, SHR-weighted scoring was assigned to the variables.