Study finds low overall discrepancy rate between residents, attending radiologists

Academic radiology departments have traditionally asked trainees to provide after-hours interpretations, with an attending radiologists confirming the findings the following morning. But concerns over quality have led some facilities to employ attending radiologists around the clock.

According to a study published by the Journal of the American College of Radiology, however, one institution's overall discrepancy rate between residents and attendings was just 1.4 percent.

“Despite a trend toward attending radiologist staffing after hours, trainees will likely issue preliminary interpretations in many departments for the foreseeable future,” wrote Vincent Mellnick, MD, Mallinckrodt Institute of Radiology at the Washington University School of Medicine in St. Louis, and colleagues. “This practice is supported by low discrepancy rates, but requires continued data tracking to ensure quality and safety and patient-centered communication of results, particularly when discrepancies exist.”

Mellnick et al. reviewed discrepancy data from January 2013 to December 2014. Radiology residents interpreted more than 153,000 exams during that time, resulting in 2,169 discrepancies (1.4 percent) between residents and attendings.

The data revealed that there were technically more discrepancies as the residents moved further along in the residency. While the discrepancy rate for postgraduate year (PGY)-5 residents was 1.88 percent, the rate was 1.65 percent for PGY-4 residents and 1.31 percent for PGY-3 residents.

“This, at first inspection, seems counterintuitive and in contrast to other studies that have shown decreasing discrepancy rates with increasing resident experience,” the authors wrote. “However, this is likely related to graduated call responsibilities in our coverage model. The patient population in our study consisted of 85.9 percent emergency department patients and 14.1 percent inpatients, the latter showing a higher discrepancy rate of 1.8 percent; these were primarily interpreted by our senior residents.”

More than 76 percent of the patients who had discrepancies were still admitted or in the emergency department when the discrepancy was found. The other patients had already been discharged, and of those discharged patients, two had “critical findings.”

“Neither patient was adversely affected by the discrepancy,” the authors wrote.