Table 4. Sensitivity Analyses Using Alternative Definitions of the Study Cohort, Exposure Status, and Outcome.
Sensitivity analysis | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 |
---|---|---|---|---|
Expanding definition of ICU to include admission to telemetry or stepdown units | ||||
Events/total (%) | 125/5034 (2.5) | 167/5545 (3.0) | 213/6408 (3.3) | 282/6964 (4.0) |
aOR (95% CI)a | 1 [Reference] | 1.12 (0.87-1.44) | 1.17 (0.91-1.50) | 1.39 (1.09-1.78) |
Expanding cohort to include patients with nonthoracic injuries with AIS score ≤2 | ||||
Events/total (%) | 294/8778 (3.3) | 452/10 539 (4.3) | 525/11 561 (4.5) | 706/15 726 (4.5) |
aOR (95% CI)a | 1 [Reference] | 1.06 (0.87-1.31) | 0.92 (0.75-1.13) | 0.71 (0.58-0.86) |
Including patients admitted to observation status, considered non-ICU | ||||
Events/total (%) | 139/5522 (2.5) | 194/5906 (3.3) | 194/5590 (3.5) | 268/7910 (3.4) |
aOR (95% CI)a | 1 [Reference] | 1.07 (0.83-1.38) | 0.99 (0.76-1.28) | 0.69 (0.54-0.89) |
Restricting cohort to patients with ≥2 fractured ribs | ||||
Events/total (%) | 112/4205 (2.7) | 182/4689 (3.9) | 180/4832 (3.7) | 234/6348 (3.7) |
aOR (95% CI)a | 1 [Reference] | 1.20 (0.92-1.58) | 0.96 (0.73-1.27) | 0.73 (0.55-0.95) |
Restricting cohort to patients with ≥3 fractured ribs | ||||
Events/total (%) | 98/3565 (2.7) | 147/3438 (4.3) | 150/4058 (3.7) | 212/5076 (4.2) |
aOR (95% CI)a | 1 [Reference] | 1.35 (1.01-1.81) | 0.98 (0.73-1.31) | 0.79 (0.60-1.04) |
Restricting cohort to hospitals with ≥20 eligible patients during the study period | ||||
Events/total (%) | 117/4838 (2.4) | 193/5142 (3.8) | 183/5365 (3.4) | 234/6522 (3.6) |
aOR (95% CI)a | 1 [Reference] | 1.22 (0.92-1.60) | 0.94 (0.71-1.24) | 0.71 (0.54-0.94) |
Expanding definition of the composite outcome to include discharge to hospice | ||||
Events/total | 177/5336 (3.3) | 244/5548 (4.4) | 249/5597 (4.4) | 338/7470 (4.5) |
aOR (95% CI)a | 1 [Reference] | 1.10 (0.87-1.39) | 0.96 (0.76-1.22) | 0.74 (0.59-0.93) |
Abbreviations: AIS, Abbreviated Injury Scale; aOR, adjusted odds ratio; ICU, intensive care unit.
Adjusted for propensity scores based on patient-level and hospital-level factors associated with ICU admission (eAppendix 2 in the Supplement).