Univariate Cox Proportional Hazards Regression for the Risk of Death
| Covariate . | Values . | Relative Risk . | P Value . | 95% Confidence Interval . |
|---|---|---|---|---|
| Age at second transplant (years) | <18 | 1.0 | — | — |
| ≥18 | 5.39 | .017 | (1.36, 21.4) | |
| Gender | Female | 1.0 | — | — |
| Male | 1.60 | .50 | (0.41, 6.2) | |
| Marrow cell dose | <3 × 108/kg | 1.0 | — | — |
| ≥3 × 108/kg | 0.33 | .11 | (0.08, 1.28) | |
| Intertransplant interval | Δ Months | 0.76 | .044 | (0.59, 0.99) |
| Year of second transplant | <1982 | 1.0 | — | — |
| ≥1982 | 0.16 | .020 | (0.03, 0.75) | |
| GVHD prophylaxis | MTX/CSP | 0.08 | .017 | (0.01, 0.64) |
| MTX | 1.0 | — | — | |
| Buffy coat after first transplant | No | 1.0 | — | — |
| Yes | 2.65 | .13 | (0.74, 9.47) | |
| Buffy coat after second transplant | No | 1.0 | — | — |
| Yes | 20.6* | .003* | (2.5, ∞)* | |
| Transfusion before first transplant | No | 1.0 | — | — |
| Yes | 0.91 | .91 | (0.19, 4.32) | |
| F/M donor/patient sex mismatch | No | 1.0 | — | — |
| Yes | 0.95 | .95 | (0.20, 4.51) |
| Covariate . | Values . | Relative Risk . | P Value . | 95% Confidence Interval . |
|---|---|---|---|---|
| Age at second transplant (years) | <18 | 1.0 | — | — |
| ≥18 | 5.39 | .017 | (1.36, 21.4) | |
| Gender | Female | 1.0 | — | — |
| Male | 1.60 | .50 | (0.41, 6.2) | |
| Marrow cell dose | <3 × 108/kg | 1.0 | — | — |
| ≥3 × 108/kg | 0.33 | .11 | (0.08, 1.28) | |
| Intertransplant interval | Δ Months | 0.76 | .044 | (0.59, 0.99) |
| Year of second transplant | <1982 | 1.0 | — | — |
| ≥1982 | 0.16 | .020 | (0.03, 0.75) | |
| GVHD prophylaxis | MTX/CSP | 0.08 | .017 | (0.01, 0.64) |
| MTX | 1.0 | — | — | |
| Buffy coat after first transplant | No | 1.0 | — | — |
| Yes | 2.65 | .13 | (0.74, 9.47) | |
| Buffy coat after second transplant | No | 1.0 | — | — |
| Yes | 20.6* | .003* | (2.5, ∞)* | |
| Transfusion before first transplant | No | 1.0 | — | — |
| Yes | 0.91 | .91 | (0.19, 4.32) | |
| F/M donor/patient sex mismatch | No | 1.0 | — | — |
| Yes | 0.95 | .95 | (0.20, 4.51) |
*Estimates and P values based on exact logistic regression methods. Cox proportional hazards model was not estimated, because all patients who did not get buffy coat survived.