Errors in fracture diagnoses in the emergency department – characteristics of patients and diurnal variation

BACKGROUND Evaluation of the circumstances related to errors in diagnosis of fractures at an Emergency Department may suggest ways to reduce the incidence of such errors. METHODS Retrospective analysis of all cases during a two year period (2002-2004) where a fracture had been overlooked or an injury had been erroneously diagnosed as a fracture (n = 61). 100 random selected patients with correctly diagnosed fractures served as control group. RESULTS In the two year period 5879 patients visited the ED with injuries. 1% of all visits to the ED resulted in an error in fracture diagnosis and 3.1% of all fractures were not diagnosed at the initial visit to the ED. 86% of such errors had consequences for treatment. No patient characteristics could be identified as risk factors for a misdiagnosis of a fracture. There was a peak in errors in fracture diagnoses between 8 pm and 2 am (47% against 20% in controls, p < 0.005). CONCLUSION A considerable number of fractures were not correctly diagnosed at the initial ED visit. There was a diurnal variation in the rate of misdiagnosis of fractures with a significant peak from 8 pm to 2 am. Where there was an error in fracture diagnosis, the patients did not appear to have a characteristic profile as regarding e.g. age, sex or capability to communicate with the ED staff. Increased consultancy service in radiology may reduce the frequency of errors in diagnosis, particularly in the evenings between 8 pm and 2 am.