Table 3 Comparison of Area Under Curve (AUC) and overall model quality (OMQ): 4 or 5 randomized variables as a novel co-variable (Variable weight assignment).

From: A longitudinal observational retrospective study on risk factors and predictive model of PICC associated thrombosis in cancer patients

Variables

Cutoff

OMQ

AUC

S.e.m

P

95% CI

Sensitivity

1-Specificity

H + P + D + L + N

8.5

0.76

0.822

0.031

0.000

0.762–0.883

0.729

0.250

P + D + L + N

7.5

0.76

0.821

0.031

0.000

0.759–0.882

0.644

0.171

H + P + D + L

6.5

0.72

0.783

0.032

0.000

0.720–0.846

0.763

0.386

H + P + D + N

7.5

0.75

0.812

0.032

0.000

0.749–0.875

0.749

0.644

H + D + L + N

6.5

0.75

0.805

0.030

0.000

0.747–0.863

0.847

0.407

H + L + N + P

6.5

0.71

0.784

0.036

0.000

0.712–0.855

0.644

0.232

Back to article page