Radiology practices in emergency departments associated with pediatric residency training programs

摘要

To examine the current radiology practices in academic emergency departments (EDs) serving children, a postal survey was done of 116 directors of EDs that serve as routine teaching sites for pediatric residents. One hundred three ED directors (89percnt;) completed the survey, representing 75 pediatric-only EDs and 28 combined pediatric/adult EDs. Thirty-four of these EDs offer a pediatric emergency medicine fellowship. Hospitals were self-categorized as children's hospitals in 41 and non-children's hospitals in 62. Radiologists immediately read every study in 8percnt; of the 103 EDs. Overall, 66percnt; of the EDs have a radiology resident in-house overnight, which is significantly more likely in non-children's hospitals than in children's (79 vs 46percnt;, P 0.001). Overnight, ED radiographs of children may at times be solely interpreted by emergency attending physicians in 57percnt; or emergency house staff in 44percnt;. In EDs that allow their house staff to interpret solo overnight, emergency attending physicians are readily available to help with these interpretations less than half the time. A radiologist's second opinion overnight is readily or usually available in 63percnt; of EDs. When any emergency physician interprets a radiograph solo overnight, the interpretation is almost always or often available later to the radiologist during the official interpretation only 40percnt; of the time. Overnight, pediatric cervical spine studies are cleared, at times, solely by emergency attending physicians in 46percnt; and by emergency house staff in 4percnt;. Only 18percnt; of programs have a daily or weekly ED radiograph review with radiologists. An ED atlas of common radiographic variants or a pediatric radiology textbook is available in 69percnt; of EDs. Quality assurance records of discordant emergency physician readings are kept in 64percnt; of programs, with significantly higher rates for children's hospitals (83 vs 48percnt;P 0.001) and for pediatric emergency medicine fellowship programs (82 vs 52percnt;,P 0.003). Official radiograph interpretations are returned within 24 hours (including weekends) in 59percnt; of programs. The rates of imaging modalities being readily or usually available to the ED overnight are as follows: head computed tomography, 99percnt; body computed tomography, 97percnt;: portable radiographs, 97percnt; fluoroscopy, 69percnt; ultrasound, 61percnt; (including only 11 of 20 EDs with more than 50,000 pediatric visits per year); nuclear medicine, 39percnt;; and magnetic resonance imaging, 24percnt;.Emergency house staff often interpret pediatric radiographs overnight, frequently without adequate backup, feedback, resources, or ongoing instruction by radiologists to improve interpretation skills. Recognition of these practices and specific educational interventions are needed. Improved turnaround time of official radiologist ED radiograph interpretations in children and of the rates of ED recording of discordant interpretations is needed. Because of the increasing importance of certain imaging modalities (particularly ultrasound) in diagnosing acute pediatric conditions overnight or on weekends, increasing the availability of such modulaties should be considered.