. Author manuscript; available in PMC: 2018 Mar 1.

Published in final edited form as: J Surg Res. 2016 Oct 5;209:168–173. doi: 10.1016/j.jss.2016.09.058

Table 2.

Bayesian network AUC for ACS-NSQIP SSI identification.

Feature used	AUC	95%	p-value^†

a. pre/intra-op risk factors	0.643	0.5767 – 0.7075	NA
b. risk factors	0.721	0.6526 – 0.7880	0.002 (a vs. b)
c. keywords	0.799	0.7349 – 0.8638	0.071 (b vs. c)
d. risk factors & keywords	0.827	0.7695 – 0.8823	0.305 (c vs. d)

^†

p-value of the difference between two AUCs