Model Performance
Nomograms derived from the multivariable model to predict the
probability of cesarean delivery were shown in Figure 1. The sum of the
assigned points for each variable in the nomograms predicted the overall
risk. The discrimination of this
prediction model was evaluated using the AUC (0.85, 95% CI: 0.79-0.91;P < 0.001) (Figure 2a). The calibration plots showed
good agreement between prediction and observation with a calibration
slope of 1.01 (Figure 2b). Additionally, the
Hosmer-Lemeshow test yielded a
nonsignificant statistic (P = 0.472), which indicated a
reasonable model fit.