Internal validation of models for predicting paediatric survival and trends in serious paediatric hospitalised injury in Australia

Vu Quang Do, Hsuen Pei Ting, Kate Curtis, Rebecca Mitchell*

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Background: Adult injury severity metrics are often applied to paediatric populations despite differences in anatomy, physiological and mortality risk. Measures to assess paediatric mortality have not been conducted on a population-wide basis. Purpose: To determine the predictive ability of the International Classification of Injury Severity Score (ICISS) in assessing 30-day mortality in a paediatric population, and to examine temporal trends for serious injury for common paediatric injury mechanisms. Methods: A retrospective analysis of linked injury hospitalisation and mortality data of children aged ≤16 years during 1 July 2002 to 30 June 2012 in Australia was conducted. Both multiplicative-injury and single-worst-injury ICISS were calculated. Logistic regression examined 30-day mortality with a range of predictor variables. The models were assessed in terms of their ability to discriminate survivors and non-survivors, model fit, and ability to explain outcome variance. Results: There were 728,584 index injury admissions and 1,064 (0.15%) deaths within 30-days of hospital admission. The multiplicative-injury ICISS was identified as a better predictor of 30-day mortality than the single worst-injury ICISS; and the best model included age group, gender, all comorbidities, trauma centre type, injury mechanism, and nature of injury as covariates. Temporal trends for serious injury have remained relatively constant over the 10-year period. Examination of specific injury mechanisms showed a significant decline in road trauma and drowning hospitalisations. In comparison, hospitalisations due to fall and self-harm injuries within adolescents increased. Conclusion: ICISS was shown to be excellent indicator for predicting 30-day mortality for all paediatric hospital admissions within a national jurisdiction. The rate of paediatric hospitalisation due to all-cause serious injury has not changed over a 10-year period despite being a national public health area of high priority.

Original languageEnglish
Pages (from-to)1769-1776
Number of pages8
JournalInjury
Volume51
Issue number8
DOIs
Publication statusPublished - 1 Aug 2020

Keywords

  • Child
  • Injury severity, 30-day mortality
  • Trauma

Fingerprint Dive into the research topics of 'Internal validation of models for predicting paediatric survival and trends in serious paediatric hospitalised injury in Australia'. Together they form a unique fingerprint.

  • Cite this