The performance of pediatric risk of mortality score in pediatric patients with severe hand, foot and mouth disease

Objective To comment the severity of severe hand, foot and mouth disease(HFMD) by pediatric risk of mortality score(PRISM), and assess the performance of PRISM in predicting mortality or complication probability in HFMD. Methods Four hundred and twenty-four severe HFMD pediatric patients were recruited in the study from 1th Jan 2010 to 31th June 2013.Information on the outcome and the variables required to calculate PRISM score were collected.The logistic regression model developed in the learning sample was evaluated in the test sample by calculating the area under the receiver operating characteristic(ROC) curve to assess discrimination pneumorrhagia and death.Calibration across deciles of risk was evaluated using the Hosmer-Lemeshow goodness-of-fit χ2 test. Results The area under the ROC curve were 0.87(95%CI 0.80~0.94)for PRISM in predicting pneumorrhagia probability.The area under the ROC curve were 0.87(95%CI 0.80~0.95)for PRISM in predicting mortality probability.The PRISM in observed and expected pneumorrhagia did not demonstrate good calibration at ten mortality risk intervals(χ2=36.66, P<0.001). The PRISM in observed and expected mortality did not demonstrate good calibration at ten mortality risk intervals(χ2=41.11, P<0.001). Conclusion The PRISM score is demonstrated good discrimination of pneumorrhagia and death in HFMD pediatric patients, but the performance of calibration is not good. Key words: Pediatric risk of mortality score; Severe hand, foot and mouth disease; Receiver operating characteristic; Hosmer-Lemeshow goodness-of-fit χ2 test