Clinical effect of a quality assurance system for radiographs in a pediatric emergency department

• Published in: Pediatric emergency care Vol. 11; no. 6; p. 351
• Main Authors: Shirm, S W, Graham, C J, Seibert, J J, Fiser, D, Scholle, S H, Dick, R M
• Format: Journal Article
• Language: English
• Published: United States 01.12.1995
• Subjects: Arkansas; Emergency Medicine - standards; Emergency Service, Hospital; Evaluation Studies as Topic; Humans; Observer Variation; Pediatrics - standards; Quality Assurance, Health Care - standards; Radiology - standards; Arkansas
• ISSN: 0749-5161
• DOI: 10.1097/00006565-199512000-00005

The objective of this study was to determine the clinical relevance and effectiveness of a quality assurance system used to identify discordant x-ray interpretations between emergency department pediatricians and pediatric radiologists in the emergency department of a large pediatric hospital. Of 5862 patients who underwent 8174 radiographic studies during a one year period, 286 patients with discordant radiographic diagnoses were identified. The incidence of discordant radiograph interpretation was determined to be 3.5% (286/8174 studies). Of those patients with discordant diagnoses, 11.5% (33/286 discordant diagnoses) received immediate intervention by the emergency department, 64.0% (183) received subsequent intervention at their follow-up appointment or by the inpatient team caring for them, 9.4% (27) required no intervention, and 15.0% (43) had no evidence of necessary intervention documented on their medical record. While only 33/5862 (0.6%) patients receiving radiographs after routine working hours required immediate intervention by the emergency department, this intervention was potentially lifesaving. No adverse outcomes were identified in this group of patients who did not receive immediate interpretation of their radiographs by a radiologist. When 24-hour in-house radiology coverage is not provided, a quality assurance system that recalls patients identified with discordant radiographic diagnoses, who may require a change in management, appears to be an effective method of patient management only when discordant interpretations are identified and promptly acted upon.