Table 2.
Primary and secondary outcome analyses with different models for cohort
| Models | P -value | Result |
|---|---|---|
| 28-day mortality | ||
| Propensity score matching model [OR (95% CI)]a | < 0.05 | 1.46 (1.07, 2.00) |
| Multivariable logistic model adjusted with all covariates [OR (95% CI)]a | < 0.05 | 1.28 (1.04, 1.57) |
| Multivariable logistic model adjusted with covariates selected by Boruta [OR (95% CI)]a | < 0.05 | 1.3 (1.06, 1.60) |
| Multivariable logistic model adjusted with all covariates using IPTW [OR (95% CI)]a | < 0.01 | 1.29 (1.11, 1.51) |
| Doubly robust estimation with all covariates [OR (95% CI)]a | < 0.05 | 1.29 (1.04, 1.61) |
| In-hospital mortality | ||
| Propensity score matching model [OR (95% CI)]a | < 0.01 | 1.61 (1.16, 2.23) |
| Multivariable logistic model adjusted with all covariates [OR (95% CI)]a | < 0.05 | 1.28 (1.03, 1.59) |
| Multivariable logistic model adjusted with covariates selected by Boruta [OR (95% CI)]a | < 0.05 | 1.3 (1.05, 1.60) |
| Multivariable logistic model adjusted with all covariates using IPTW [OR (95% CI)]a | < 0.01 | 1.28 (1.09, 1.50) |
| Doubly robust estimation with all covariates [OR (95% CI)]a | < 0.05 | 1.28 (1.02, 1.60) |
Statistical analyses of different models with P -value < 0.05 were displayed in bold
aOR Odds Ratio, CI Confidence Interval