Risk of mortality in pediatric intensive care unit, assessed by PRISM-III

Pak J Biol Sci. 2009 Mar 15;12(6):480-5. doi: 10.3923/pjbs.2009.480.485.

Abstract

This study aimed at evaluating the mortality rate in a PICU applying PRISM-III. Two hundred and twenty one infants and children consecutively admitted to PICU of Tabriz Children's Hospital were studied during a 13 months period of time. Data required for calculating the PRISM-III score were collected during the first 24 h of PICU stay in all patients. The prediction of actual mortality by PRISM-III scoring was evaluated by the Hosmer and Lemeshow goodness-of-fit test. Receiver Operating Characteristic (ROC) curve was constructed, as well. The observed (O) short-term (during hospital stay) mortality rate was compared with the expected (E) figures as the O/E ratio. The mean value of the PRISM-III score was 14.22 +/- 9.57(2-42). ROC analysis indicated a strong predictive power for the PRISM-III (area under the curve = 0.898) and the test was well fit to the designed study (goodness-of-fit p-value = 0.161). The observed short-term mortality rate was 9.05% and the expected mortality rate by the PRISM-III scoring was 9% (O/E ratio = 1.005). The PRISM-III scoring system was highly calibrated in our institute.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Child
  • Child, Preschool
  • Cross-Sectional Studies
  • Female
  • Hospital Mortality*
  • Humans
  • Infant
  • Intensive Care Units, Pediatric*
  • Iran
  • Male
  • Models, Theoretical*
  • Outcome Assessment, Health Care
  • ROC Curve
  • Risk Adjustment / methods*
  • Risk Assessment / methods*
  • Risk Factors
  • Treatment Outcome