First year radiology residents not taking call: will there be a difference?

Emerg Radiol. 2007 Feb;13(5):231-5. doi: 10.1007/s10140-007-0571-7. Epub 2007 Jan 25.


Currently, there is a debate in the academic radiology community about whether or not first year residents should take overnight call. The purpose of this study was to track discrepancies on overnight resident preliminary reads on radiographs from the emergency department to see if the experience level of the resident makes a difference. From October 1, 2005 to September 22, 2006, 13,213 radiographs were prospectively interpreted by residents at night at a Level I Trauma Center. Discrepancies were documented after review of the films with the staff radiologist in the morning. The patient's medical record was then examined to determine if there was any adverse clinical outcome as a result of the reading. Of the 13,184 radiographs interpreted, 120 total discrepancies were identified (overall discrepancy rate 0.9%). First year residents showed a discrepancy rate of 1.59%, higher than other residents, which were ranged from 0.39 to 0.56%. Of the 54 patients with follow-up imaging, the abnormality that was felt to be present by staff persisted on follow-up imaging in 22 cases; however, the abnormality was not present on follow up of the other 32 patients (59.2% of discrepancies with follow-up imaging). Although there is higher rate of discrepancy among reports generated by first year residents, the difference compared to the other levels of experience is small, and its overall significance can be debated. Follow-up imaging often showed that staff interpretations were false positives when there was a discrepancy reported.

MeSH terms

  • Clinical Competence
  • Diagnostic Errors
  • Humans
  • Internship and Residency / statistics & numerical data*
  • Observer Variation
  • Personnel Staffing and Scheduling
  • Professional Autonomy
  • Radiology Department, Hospital / statistics & numerical data*
  • Radiology* / methods
  • Trauma Centers / statistics & numerical data