Two specially trained radiographers took part in the radiologists' rota for "cold" reporting skeletal radiographs of patients who attended the Accident and Emergency (A&E) Department at times when there was no "hot" reporting service operating, and who were not admitted for treatment or referred to fracture clinics for follow-up. These radiographs had initially been seen by A&E medical staff. At the end of an 18 month period during which the two radiographers reported on 11,322 skeletal examinations, a retrospective search was made to detect interpretive errors. The radiological history of all 11,322 patients was reviewed over a follow-up period of at least 3 months subsequent to the attendance reported by the radiographers. 48% of patients did not re-attend, 42% re-attended for unrelated examinations, and 10% re-attended for repeat examinations of the same anatomical area, or for different procedures (bone scintigraphy or CT) related to the original injury. The second (or subsequent) report was discrepant with that of the first attendance in only 29 cases out of 1103. Of these 29 patients, 13 had occult fractures which were undetectable at first attendance even in retrospect, six had new injuries accounting for the new findings, six had been the subject of false positive calls at an earlier visit, one had a fracture of the hamate missed at first attendance, and in three cases no consensus could be reached as to the cause of the discrepancy, owing to incomplete films or insufficient clinical data. Analysis of the patients' postcodes showed at least 89% were of local origin and only 1% were from outside Yorkshire, suggesting that the review should have identified the great majority of erroneous reports. It is concluded that appropriately trained and supervised radiographers can successfully undertake diagnostic reporting of selected skeletal examinations on A&E patients.