Table 3.
Variable | Unadjusted | Adjusted | |||
---|---|---|---|---|---|
OR [95% CI] | p | OR [95% CI] | Wald | p | |
In-hospital death | |||||
Age (0 = 67; per 10 years) | 1.46 [1.40–1.53] | < 0.001 | 1.55 [1.47–1.63] | 271.41 | < 0.001 |
Temperature (0 = 98.4) | 0.86 [0.83–0.89] | < 0.001 | 0.90 [0.86–0.93] | 32.35 | < 0.001 |
Bilirubin > 2 mg/dL | 2.70 [2.22–3.28] | < 0.001 | 2.33 [1.88–2.90] | 58.21 | < 0.001 |
History of heart failure | 1.23 [1.06–1.43] | 0.006 | 1.01 [0.86–1.19] | 0.02 | 0.889 |
Diabetes | 0.63 [0.55–0.73] | < 0.001 | 0.65 [0.56–0.76] | 28.68 | < 0.001 |
Lactate (per 1 mmol/L) | 1.26 [1.23–1.29] | < 0.001 | 1.21 [1.17–1.24] | 171.84 | < 0.001 |
MAP (per 10 mmHg) | 0.83 [0.80–0.85] | < 0.001 | 0.88 [0.85–0.91] | 54.45 | < 0.001 |
Heart rate (per 10 bpm)a | |||||
≤ 95 bpm | 0.87 [0.82–0.93] | < 0.001 | – | – | – |
> 95 bpm | 1.07 [1.02–1.12] | 0.005 | – | – | – |
Slope difference | – | < 0.001 | – | – | 0.058 |
Overall | – | – | 1.06 [1.03–1.09] | 12.31 | < 0.001 |
Variable | Unadjusted | Adjusted | |||
---|---|---|---|---|---|
HR [95% CI] | p | HR [95% CI] | Wald | p | |
Discharge from the ICU | |||||
Age (0 = 67; per 10 years) | 0.95 [0.93–0.97] | < 0.001 | 0.95 [0.93–0.97] | 27.70 | < 0.001 |
Temperature (0 = 98.4) | 1.07 [1.06–1.09] | < 0.001 | 1.06 [1.05–1.08] | 58.95 | < 0.001 |
Bilirubin > 2 mg/dL | 0.65 [0.58–0.73] | < 0.001 | 0.70 [0.62–0.79] | 35.30 | < 0.001 |
History of heart failure | 0.88 [0.82–0.94] | < 0.001 | 0.88 [0.82–0.95] | 12.20 | < 0.001 |
Diabetes | 1.11 [1.04–1.18] | 0.001 | 1.10 [1.03–1.17] | 8.89 | 0.003 |
Lactate (per 1 mmol/L) | 0.95 [0.93–0.96] | < 0.001 | 0.96 [0.95–0.98] | 28.12 | < 0.001 |
MAP (per 10 mmHg) | 1.03 [1.01–1.04] | < 0.001 | 1.02 [1.01–1.03] | 6.04 | 0.014 |
Heart rate (per 10 bpm) | 1.01 [1.00–1.03] | 0.024 | 0.99 [0.98–0.01] | 1.62 | 0.203 |
Both the unadjusted and adjusted models included facility indicator variables as fixed effects to account for the clustering of patients within facilities.
aAfter adjusting for covariates, the slope difference in the log-odds of in-hospital death before vs. after HR of 95 was no longer statistically significant (p = 0.058); as such, the adjusted model includes a single overall effect of HR
OR odds ratio, HR hazard ratio, MAP mean arterial pressure