Using crash scene variables to predict the need for trauma center care in older persons

Linda J. Scheetz, Juan Zhang, John Kolassa

Research output: Contribution to journalArticlepeer-review

18 Scopus citations

Abstract

Current trauma triage protocols lack sensitivity to occult injuries in older persons, resulting in unacceptable undertriage rates. We identified crash scene information that could be used by emergency personnel to identify the need for trauma center care in older persons injured in motor vehicle crashes. Crash records of 7,883 persons 65 years and older were explored using classification and regression trees (CART) analysis. CART analysis of 26 crash scene variables resulted in two classification trees from which triage decision rules were stated for persons with severe and moderate injuries. Sensitivity and specificity of the rules were 95.15% and 76.47% for severe injury and 83.1% and 81.5% for moderate injury.

Original languageEnglish (US)
Pages (from-to)399-412
Number of pages14
JournalResearch in Nursing and Health
Volume30
Issue number4
DOIs
StatePublished - Aug 1 2007

All Science Journal Classification (ASJC) codes

  • Nursing(all)

Keywords

  • Artificial intelligence
  • Classification trees
  • Data mining
  • Elderly
  • Emergency medical services
  • Prehospital triage
  • Trauma

Fingerprint Dive into the research topics of 'Using crash scene variables to predict the need for trauma center care in older persons'. Together they form a unique fingerprint.

Cite this