Pediatric emergency physician interpretation of plain radiographs: Is routine review by a radiologist necessary and cost-effective?

STUDY OBJECTIVE

To determine the concordance rate of plain radiograph interpretations by pediatric emergency physicians and pediatric radiologists, to evaluate the effect of incorrect radiologic diagnosis on patient management, and to evaluate the necessity and cost-effectiveness of routine follow-up review of all plain radiographs by a radiologist.

METHODS

We assembled a prospective series of all patients who presented to the emergency department of an urban tertiary care children's hospital and underwent plain radiography between October 1 and October 31, 1994. Pediatric emergency physicians documented their interpretations. Within 24 hours, films were reviewed by a pediatric radiologist. The two interpretations were classified as concordant or discordant and were further assessed for medical significance and subsequent change in management.

RESULTS

During the study period, 707 radiographic examinations were performed: chest, 56%; skeletal excluding spine, 20.1%; abdomen, 11.9%; sinus, 4.2%; spine, 3.6%; and other, 4%. The accuracy or concordance rate was 90.2% (638 of 707) for pediatric emergency physician interpretations; clinical management was unchanged in 96.9% (685 of 707) of the cases. Of the 69 discordant interpretations, 48 were clinically significant, with 22 requiring changes in management. They included 9 false-negative interpretations by pediatric emergency physicians: (5 fractures, 2 cases of pneumonia, 1 case of sinusitis, 1 case of cardiomegaly); 10 false-positive interpretations by pediatric emergency physicians (5 fractures, 4 cases of pneumonia, 1 case of sinusitis), and 3 false-positive interpretations by radiologists (1 case of C-2 spine subluxation, 1 retropharyngeal abscess, and 1 case of necrotizing enterocolitis). No adverse outcomes resulted from these misinterpretations. Routine review of all plain radiographs by a radiologist represents an estimated $210,000 annual cost to the patients and payers.

CONCLUSION

Radiograph interpretations by pediatric emergency physicians were generally accurate, and no adverse outcomes occurred as a result of misinterpretation. Clinical assessment probably assisted these physicians in interpreting the radiographs of high-risk patients. Judicious consultation with a radiologist during the initial presentation of a high-risk patient, when deemed warranted by the pediatric emergency physician, will help the emergency physician deliver high-quality, cost-effective health care. Given the overall clinical accuracy rate of radiograph interpretations by the pediatric emergency physicians and the cost of routine review of all plain radiographs in the ED by a radiologist, routine review versus selective specialty consultation must be further evaluated.