Using quantitative methods to improve the diagnostic workup for abdominal pain in children

David A. Lanning, Ronald L. Thomas, Kim D. Rood, Michael D. Klein

Research output: Contribution to journalArticlepeer-review

Abstract

Introduction: The differential diagnosis of abdominal pain in children can be challenging. We applied quantitative decision-making methods to this process and sought to determine if their use provided measurable benefit. Methods: After obtaining institutional review board approval, we recorded key elements of the history, physical examination, laboratory, and imaging evaluations along with the cost and the time spent in the emergency department (ED) for children presenting with abdominal pain. Initially, data were collected (group 1, n = 1366 patients) and then presented to the ED pediatricians. For subsequent patients, ED physicians received a sheet specific to that patient's age and sex reporting the most common diagnoses and the elements of the evaluation that had proven most useful (group 2, n = 624 patients). We compared the difference in length of stay and costs before and after intervention, between study groups, by age groups, and separately by sex using a 2-factor analysis of variance. Results: The diagnostic workup cost less in boys aged 2 to 12 years after the intervention. In boys and girls older than 12 years, the cost trended lower. Discussion: This study demonstrates that ED physicians equipped with specific information were able to complete their diagnostic evaluation of children presenting with abdominal pain at a lower cost.

Original languageEnglish
Pages (from-to)949-954
Number of pages6
JournalJournal of Pediatric Surgery
Volume40
Issue number6
DOIs
StatePublished - Jun 2005

Keywords

  • Abdominal pain
  • Appendicitis
  • Bayes theorem
  • Posttest probability
  • Pretest probability
  • Quantitative analysis

Fingerprint

Dive into the research topics of 'Using quantitative methods to improve the diagnostic workup for abdominal pain in children'. Together they form a unique fingerprint.

Cite this