Skip to main content
. 2020 Sep 24;15(9):e0239519. doi: 10.1371/journal.pone.0239519

Table 7. Two-factor logistic regression model for survival prediction.

Variable β SE OR 95% CI for OR p value
Age (years) -0.082 0.029 0.921 0.863–0.979 0.004
pO2 (%) 0.188 0.061 1.207 1.085–1.390 0.002
Intercept -9.902 5.181 - - 0.056
ROC AUC SE 95% CI p value
0.857 0.05 0.752–0.963 <0.0001
Characteristics* Sensitivity Specificity Positive PV Negative PV Accuracy
98% 53% 89% 89% 89%
Log-likelihood ratio test Statistic p value
25.71 <0.0001

Abbreviations: CI, confidence interval; OR, odds ratio; SE, standard error; ROC, receiver operating curve. *Model characters in predicting survival in obtained dataset, not validated with testing dataset.