Univariate analysis and logistic regression analysis of the timing of convalescent plasma administration with 28-day mortality as the dependent outcome stratified by site of admission
| Site of plasma administration | Timing of admission (days) | Time (days) of admission-plasma infusion Median (IQR) | Patients receiving convalescent plasma n (%) | 28-day mortality 100/patients | Unadjusted OR (95% CI) | Adjusted* OR (95% CI) |
| General ward (n=3036) | >7 | 10.8 (10) | 130 (4.3) | 26.2 | 1 | 1 |
| 3–7 | 3.2 (2) | 814 (26.8) | 18.4 | 0.64 (0.42 to 0.98) | 0.67 (0.43 to 1.05) | |
| <3 | 0.2 (1) | 2092 (68.9) | 12.6 | 0.41 (0.27 to 0.62) | 0.44 (0.28 to 0.67) | |
| Intensive care unit (n=1171) | >7 | 8.5 (6) | 52 (4.4) | 55.8 | 1 | 1 |
| 3–7 | 3.3 (2) | 383 (32.7) | 37.3 | 0.47 (0.26 to 0.84) | 0.56 (0.30 to 1.03) | |
| <3 | 0.3 (1) | 736 (62.9) | 26.2 | 0.28 (0.16 to 0.50) | 0.35 (0.20 to 0.64) | |
| ICU admission with requirement of mechanical ventilation (n=512) | >7 | 9.50 (7) | 44 (8.8) | 51.7 | 1 | 1 |
| 3–7 | 3.5 (2) | 183 (35.7) | 65.5 | 1.68 (0.86 to 3.23) | 1.98 (0.99 to 3.96) | |
| <3 | 0.1 (1) | 285 (55.7) | 37.9 | 0.47 (0.25 to 0.90) | 0.52 (0.27 to 1.01) |
*Adjusted for age, gender, obesity, diabetes, hypertension, cardiovascular disease, chronic obstructive pulmonary disease and immunodeficiencies
ICU, intensive care unit.;