Timeliness of model
| Cohort, (n) | Correctly identifies, n (%) | Time from intubation, median (IQR), hours | Before intubation, n (%) | After intubation, n (%) | Time from ARDS label, median (IQR), hours | Before ARDS, n (%) | After ARDS, n (%) | Time from death median (IQR), hours |
| MV non-COVID-19 cohort | ||||||||
| ARDS, (204) | 166 (81.4) | 0 (−12.8 to 26.0) | 87 (52.4) | 79 (47.6) | 0 (−43.8 to 12.0) | 115 (69.3) | 51 (30.7) | |
| Death, (69) | 60 (87) | 225.5 (−461.3 to –101.3) | ||||||
| ARDS and death, (116) | 108 (93.1) | -1 (−38.8 to 9.3) | 68 (63.0) | 40 (37.0) | 20 (−115.5 to 0.3) | 81 (75.0) | 27 (25.0) | 314 (−588.5 to –127.8) |
| ARDS or death, (389) | 274 (70.4) | 1 (−17.8 to 15.0) | 155 (56.6) | 119 (43.4) | 10.0 (−75.5 to 4.0) | 196 (71.5) | 78 (28.5) | 225.5 (−461.3 to –101.3) |
| No ARDS or death, (392) | 223 (56.9) | |||||||
| COVID-19 cohort | ||||||||
| ARDS, (328) | 318 (97) | 3 (−8.8 to 11.0) | 136 (42.8) | 182 (57.2) | 0 (−16 to 9.0) | 141 (44.3) | 128 (40.3) | |
| Death, (652) | 308 (47.2) | 58 (−112 to –20) | ||||||
| ARDS and death, (255) | 237 (92.9) | 4 (−1 to 18) | 86 (36.3) | 156 (65.8) | 0 (−12 to 10) | 125 (52.7) | 112 (47.3) | 112 (−211.3 to –52) |
| ARDS or death, (1235) | 555 (44.9) | 3 (−3.5 to 13.0) | 222 (40.0) | 333 (60.0) | 0 (−14.0, 10.0) | 266 (47.9) | 240 (43.2) | 58 (−112 to –20) |
| No ARDS or death, (4437) | 3724 (83.9) | |||||||
| MV COVID-19 subcohort | ||||||||
| ARDS, (328) | 318 (97) | 3 (−8.8 to 11.0) | 136 (42.8) | 182 (57.2) | 0 (−16.0 to 9.0) | 141 (44.3) | 128 (40.3) | |
| Death, (163) | 128 (78.5) | 37 (−87 to –11) | ||||||
| ARDS and death, (255) | 237 (92.9) | 4 (−1, 18) | 86 (36.3) | 156 (65.8) | 0 (−12 to 10) | 125 (52.7) | 112 (47.3) | 112 (−211. 3 to –52) |
| ARDS or death, (746) | 555 (74.4) | 3 (−3.5 to 13.0) | 222 (40.0) | 333 (60.0) | 0 (−14.0 to 10.0) | 266 (47.9) | 240 (43.2) | 37 (−87 to –11) |
| No ARDS or death, (57) | 13 (22.8) |
ARDS, acute respiratory distress syndrome; MV, mechanically ventilated.