B G Minnes1, T Sutcliffe, T P Klassen. 1. Division of Emergency Medicine, Children's Hospital of Eastern Ontario, Ottawa, Canada.
Abstract
OBJECTIVES: To measure agreement beyond chance (kappa) for comparison interpretations of extremity radiographs by pediatric radiologists and emergency physicians (EPs) and to identify factors associated with disagreement. METHODS: A random sample of 205 radiographs was selected from 1,016 patients having x-rays of their extremities in the emergency and radiology departments of a tertiary care pediatric hospital. Interpretations by the "official" reporting pediatric radiologist (ORPR), the treating EP, and a pediatric radiologist blinded to all clinical information (BPR) were compared for three categories: "abnormal" (one or more of fracture, dislocation, or effusion); "possibly abnormal"; and "normal." RESULTS: The overall weighted kappa (Kw) for the ORPRs and the EPs was 0.55. For fractures alone, the Kw for the ORPRs vs the EPs was 0.77; and for effusions alone, the value was 0.34. The Kw for the ORPRs vs the BPR was 0.63 (range 0.43-0.83 for individual ORPRs). The main areas of disagreement were in the identification of joint effusions and of nondisplaced fractures of the phalanges, elbow joint, tarsals, or metatarsals. CONCLUSIONS: There is good agreement between EPs and pediatric radiologists in interpreting extremity radiographs of injured children and adolescents. Disagreement occurs mainly for effusions or minor fractures and for the elbow region. Because of the importance of recognizing abnormalities in this region, an educational intervention to improve this area of deficiency is recommended.
OBJECTIVES: To measure agreement beyond chance (kappa) for comparison interpretations of extremity radiographs by pediatric radiologists and emergency physicians (EPs) and to identify factors associated with disagreement. METHODS: A random sample of 205 radiographs was selected from 1,016 patients having x-rays of their extremities in the emergency and radiology departments of a tertiary care pediatric hospital. Interpretations by the "official" reporting pediatric radiologist (ORPR), the treating EP, and a pediatric radiologist blinded to all clinical information (BPR) were compared for three categories: "abnormal" (one or more of fracture, dislocation, or effusion); "possibly abnormal"; and "normal." RESULTS: The overall weighted kappa (Kw) for the ORPRs and the EPs was 0.55. For fractures alone, the Kw for the ORPRs vs the EPs was 0.77; and for effusions alone, the value was 0.34. The Kw for the ORPRs vs the BPR was 0.63 (range 0.43-0.83 for individual ORPRs). The main areas of disagreement were in the identification of joint effusions and of nondisplaced fractures of the phalanges, elbow joint, tarsals, or metatarsals. CONCLUSIONS: There is good agreement between EPs and pediatric radiologists in interpreting extremity radiographs of injured children and adolescents. Disagreement occurs mainly for effusions or minor fractures and for the elbow region. Because of the importance of recognizing abnormalities in this region, an educational intervention to improve this area of deficiency is recommended.
Authors: C James Holliman; Terrence M Mulligan; Robert E Suter; Peter Cameron; Lee Wallis; Philip D Anderson; Kathleen Clem Journal: Int J Emerg Med Date: 2011-07-22