Objectives: To assess the performance of the Pediatric Index of Mortality (PIM) 2 score in Italian pediatric intensive care units (PICUs).
Design: Prospective, observational, multicenter, 1-year study.
Setting: Eighteen medical-surgical PICUs.
Patients: Consecutive patients (3266) aged 0-16 years admitted between 1 March 2004 and 28 February 2005.
Measurements and main results: To assess the performance of the PIM2 score, discrimination and calibration measures were applied to all children admitted to the 18 PICUs, in the entire population and in different groups divided for deciles of risk, age and admission diagnosis. There was good discrimination, with an area under the receiver operating characteristic (ROC) curve of 0.89 (95% CI 0.86-0.91) and good calibration of the scoring system [non-significant differences between observed and predicted deaths when the population was stratified according to deciles of risk (chi2 9.86; 8 df, p = 0.26) for the whole population].
Conclusions: The PIM2 score performed well in this sample of the Italian pediatric intensive care population. It may need to be reassessed in the injury and postoperative groups in larger studies.