Comparison of Pediatric Risk of Mortality (PRISM III) score with Pediatric Index of Mortality (PIM III) score in Pediatric Intensive Care Unit: A Single center, prospective observational study from South India

Abstract

Abstract Background: Pediatric risk of mortality (PRISM III) score and pediatric index of mortality (PIM III) scores are most recent versions of PRISM and PIM score used for mortality prediction in the Pediatric intensive care units (PICUs). This study aims to compare the performance of these two scores in children admitted to a tertiary PICU of South India. Subjects and Methods: This single-center, prospective observational study was conducted in PICU at a Tertiary Care Hospital in South India over a period of 20 months. A total of 214 cases admitted between the age groups of 1 month and 18 years were included in this study. PIM III and PRISM III scores were calculated. Quantitative correlation between the results of the score was analyzed by spearman test. Discrimination was assessed using the area under receiver operating characteristics curve and calibration using Hosmer–Lemeshow goodness-of-fit test. Results: Out of 214 children, observed mortality was 11 (5.1%). Estimated mortality using PRISM III and PIM III was 4 (1.86%) and 2 (0.93%), respectively. Although both scoring systems had good discrimination performance, PIM III score (0.934 [confidence interval (CI) 95%, 0.886–0.982]) had higher discrimination than PRISM III score(0.888 [CI 95%, 0.797–0.980]). PRISM III score (χ 2 = 4.469 and P = 0.484) had better calibration as compared to PIM III score (χ 2 = 16.093 and P = 0.041). Conclusions: Both PRISM III and PIM III score showed good discriminatory performance, however PRISM III had better calibration than PIM III score.