Fig. 3

Apparent performance of the predictive nomogram in the cohort. Calibration curves of the pain severity nomogram prediction in the training cohort (a) and the validation cohort (d): The x-axis represents the predicted pain severity of CP/CPPS patients. The y-axis represents the actual pain severity of CP/CPPS patients. Receiver operating characteristic (ROC) curves of the nomogram in the training cohort (b) and the validation cohort (e): The ROC curve is displayed in solid line, and the reference is displayed in dotted line. The ROCs of the predictive nomogram in the training and validation cohorts, with the AUC of 0.781 and 0.735, respectively. Decision curve analysis (DCA) of the nomogram in the training cohort (c) and the validation cohort (f): The y-axis measures the net benefit. The blue solid line represents the pain severity predictive nomogram. The thin solid line represents the hypothesis that all patients are mild pain. The thin thick solid line represents the assumption that patients are moderate to severe pain. The DCA showed that if the threshold probability of a patient and a doctor are > 25% and < 83% in training cohort (c) and > 16% and < 78% in the validation cohort (f), respectively. Using this predictive nomogram to predict the pain severity of CP/CPPS patients adds more benefit than the intervention-all-patients scheme or the intervention-none scheme