Skip to main content
. Author manuscript; available in PMC: 2010 Oct 1.
Published in final edited form as: Aliment Pharmacol Ther. 2009 Jul 13;30(7):741–748. doi: 10.1111/j.1365-2036.2009.04096.x

Table 3.

Cox Regression Models Predicting Death at One-Year

Model Predictor HR (95% CI) p-value C-index (95% CI) C-index improvementc IDIc
Ascites only Ascitesa 6.08 (3.62–10.19) <0.0005 0.653 (0.58–0.72) - -
MELD only MELD 1.21 (1.16–1.26) <0.0005 0.730 (0.66–0.80) - -
MELDNa only MELDNa 1.22 (1.17–1.27) <0.0005 0.770 (0.70–0.83) - -
MELDNa + Ascites MELDNa 1.24 (1.18–1.29) <0.0005 0.804 (0.75–0.85) 3.4% (p=0.036) 1.8% (p=0.016)
Ascitesb 1.20 (1.11–1.30) <0.0005
a

Ascites coded as a binary predictor.

b

Linear spline, equal to 21-MELDNa for patients with ascites and MELDNa scores < 21, and zero for all other patients.

c

The improvement in model performance compares a MELDNa model against MELDNa + Ascites