Logistic regression model can reduce unnecessary artificial liver support in hepatitis B virus-associated acute-on-chronic liver failure: decision curve analysis

Table 4 Net Benefit for ALSS for All ACLF Patients or According to LRM, Using a Threshold of Pt

Pt (%)	Net Benefit		Advantage of LRM-guided ALSS
Pt (%)	ALSS for All	LRM-guided ALSS	Net Benefit	Reduction in Avoidable ALSS per 100 Patients
16	0.430	0.430	0	0
20	0.402	0.414	0.012	5
30	0.317	0.365	0.048	11
40	0.203	0.299	0.096	14
50	0.043	0.207	0.164	16
60	−0.197	0.069	0.266	18
64	−0.329	−0.008	0.321	18

Note: The reduction in the number of unnecessary ALSS per 100 patients is calculated as follows: (net benefit of the model – net benefit of treat all)/[Pt/(1 – Pt)] × 100
ALSS artificial liver support system, ACLF, acute-on-chronic liver failure, LRM logistic regression model, Pt threshold probability of risk

ISSN: 1472-6947