Clinical impact of radiograph misinterpretation in a pediatric ED and the effect of physician training level

Radiograph interpretation in the pediatric emergency department (ED) is commonly performed by pediatric emergency medicine (PEM) attendings or physicians-in-training. This study examines the effect of physician training level on radiograph interpretation and the clinical impact of false-negative radiograph interpretations. Data were collected on 1,471 radiographs of the chest, abdomen, extremity, lateral neck, and cervical spine interpreted by PEM attendings, one PEM fellow, one physician assistant, and emergency medicine, pediatric and family practice residents. Two hundred radiographs (14%) were misinterpreted, including 141 chest (16%), 24 extremity (8%), 20 abdomen (12%), 14 lateral neck (18%), and 1 cervical spine radiograph (2%). Physicians-in-training misinterpreted 16% of their radiographs versus 11% for PEM attendings (P = .01). Twenty (1.4%) radiographs had clinically significant (false-negative) misinterpretations, including 1.7% of physician-in-training and 0.8% of attending interpretations (P = 0.15). No morbidity resulted from the delay in correct interpretation. Radiograph misinterpretation by ED physicians occurs but is unlikely to result in significant morbidity.