Discrepancies in interpretation of ED body computed tomographic scans by radiology residents

Nelson Tieng, Diana Grinberg, Siu Fai Li

Research output: Contribution to journalArticlepeer-review

55 Scopus citations

Abstract

Objective: In academic institutions, radiology residents are often relied on for providing preliminary reports of imaging studies done in the ED. We examined the prevalence of discrepant interpretations of body computed tomographic (CT) scans in our institution. Methods: We conducted a retrospective study on a consecutive series of body CT scans at an urban ED. We compared the preliminary interpretation by radiology residents with the final interpretation by radiology attending physicians. An interpretation was characterized as having no discrepancy, minor discrepancy, or major discrepancy. A major discrepancy was defined as a discrepancy that resulted in a change in diagnosis, treatment, or disposition. Results: Two hundred three body CT scans were identified during the study period. Of these CT scans, 20 had major discrepancies (10%), 40 had minor discrepancies (20%), and 143 had no discrepancy (70%). Major discrepancies included missed appendicitis, normal appendix, missed bowel obstruction, and missed colon cancer. Computed tomographic scans with abnormal findings were more likely to contain major discrepancies (relative risk = 6.0; 95% confidence interval = 1.8-2.0). Conclusion: Discrepancies between radiology residents and radiology attending physicians were common at our institution. Emergency department physicians should exercise caution when relying on residents' interpretation of body CT scans.

Original languageEnglish (US)
Pages (from-to)45-48
Number of pages4
JournalAmerican Journal of Emergency Medicine
Volume25
Issue number1
DOIs
StatePublished - Jan 2007

ASJC Scopus subject areas

  • Emergency Medicine

Fingerprint

Dive into the research topics of 'Discrepancies in interpretation of ED body computed tomographic scans by radiology residents'. Together they form a unique fingerprint.

Cite this