Do heart and respiratory rate variability improve prediction of extubation outcomes in critically ill patients?

Table 3 Prognostic accuracy comparison

Model	ROC AUC	Sensitivity*	Specificity*	PPV*	NPV*	NRI*
Single logistic regression: heart rate	0.51	0.5	0.55	0.11	0.87	0.22
Single logistic regression: RSBI	0.61	0.5	0.72	0.18	0.91	0.04
Single logistic regression: respiratory rate	0.63	0.5	0.66	0.17	0.91	0.04
Ensemble of three univariate logistic regressions:	0.62	0.5	0.69	0.17	0.90	0.07
Heart rate, respiratory rate, RSBI	0.62	0.5	0.69	0.17	0.90	0.07
WAVE score	0.69	0.75	0.59	0.18	0.94	-

*Positive test for probability of failure equal or above 0.5. ROC AUC, area under the receiver operating characteristic curve; PPV, positive predictive value; NPV, negative predictive value; NRI, net reclassification improvement; RSBI, rapid shallow breathing index; WAVE, weaning and variability evaluation.

ISSN: 1364-8535