Both radiology residents and attending providers make more interpretive mistakes during evening and overnight hours.
Everyone makes mistakes. And, according to a new study, both radiology residents and full attending providers make more interpretation errors with body CT scans during the evening and overnight hours.
In a study published Aug. 18 in Radiology, investigators from the Mayo Clinic Arizona in Phoenix, take the first look at all providers – not just trainees – perform during off-hours. Their findings, said lead author and radiologist Anika Patel, M.D., reveal that fatigue and circadian misalignment factor heavily into these mistakes, highlighting the need to carefully management workload during these time periods.
“If diminished diagnostic accuracy during night assignments is a function of circadian misalignment and fatigue, our works shows it extends beyond residents to radiologists who have completed residency,” he said. “These findings have implications for patient care and quality assurance efforts, and for designing processes to deliver the highest quality of care at lower cost.”
Related Content: Diagnostic Errors: Lessons Learned and Mitigation Strategies
Given that the higher off-hours error rate is known, he added, that knowledge can be used to identify which studies might benefit from more increased quality scrutiny, as well as how to best implement that oversight.
A, Abdominal and, B, pelvic reconstructed CT images in a 66-year-old woman with cirrhosis obtained in the emergency department (ED) for abdominal pain show gallstones (arrowhead) detected by the off-hours in-house radiologist during night assignment at 12:30 am. This patient was discharged from the ED with a diagnosis of biliary colic. An abdominal division attending physician (A.10) revised the report at 6:30 am and noted that “the endometrium is abnormally thickened up to 14 mm in the anteroposterior diameter with enhancing intracavitary observation” (arrow) and documented discussion with the ED. The ED contacted the patient later that day with recommendation and appointment for outpatient gynecologic consultation. Over the subsequent 6 weeks, the patient underwent gynecologic consultation, pelvic US, and eventual hysteroscopic resection of a benign endometrial polyp. Courtesy: Radiology
Previous studies examining overnight radiology performance has focused on trainees, but Patel’s team investigated whether the same diagnostic performance problems plague more established providers. Therefore, they followed radiology fellows during both day and night shifts.
In a study that ran from June 2014 to June 2018, the team evaluated the interpretative performance of 32 fellows who read 10,090 CT scans during off hours. They concentrated on errors that affected follow-up care and compared mistakes that happened during daytime hours (7 a.m. to 5:59 p.m.) to those that occurred during evening and overnight (6 p.m. to 6:59 a.m.).
According to their analysis, Patel’s team identified a 2-percent error rate with daytime reading and a 3-percent error rate with evening and overnight. Moreover, they found 69 percent of readers made more mistakes during night hours with most of the mistakes occurring during the second part of the night shift – 3.7 percent versus 2.5 percent, respectively.
It is possible to reduce both the number and the impact of these interpretation errors, however, the team said. Double reading has proven effective, and adding additional radiologists to the night shift can also alleviate some of the workflow burden, potentially reducing errors.
Overall, the team said, their findings show that interpretation errors that occur during the evening and overnight hours do not only plague trainees. Michael Bruno, M.D., radiology professor and director of quality management services at Penn State Milton S. Hershey Medical Center, agreed in an accompanying editorial, noting that these results have widespread implications throughout radiology.
“This article puts to rest the wrongheaded assumption that only radiology trainees will experience errors and have diminished performance when working after hours,” he wrote. “To the contrary, we are all subject to the same biologic and neurocognitive limitations as part of our shared humanity.”
Consequently, he added, the profession must pay close attention to the circadian rhythms of its providers and put strategies and practices in place to help prevent any associated interpretation errors.
Can CT-Based Deep Learning Bolster Prognostic Assessments of Ground-Glass Nodules?
June 19th 2025Emerging research shows that a multiple time-series deep learning model assessment of CT images provides 20 percent higher sensitivity than a delta radiomic model and 56 percent higher sensitivity than a clinical model for prognostic evaluation of ground-glass nodules.
How to Successfully Launch a CCTA Program at Your Hospital or Practice
June 11th 2025Emphasizing increasing recognition of the capability of coronary computed tomography angiography (CCTA) for the evaluation of acute and stable chest pain, this author defuses common misperceptions and reviews key considerations for implementation of a CCTA program.