Consistency in nuclear medicine reporting--a pilot study using bone scans

In a pilot study of consistency in nuclear medicine reporting, a panel of three observers evaluated 60 bone scans carried out for detection of metastases. The bone scans had originated from 10 departments. Difficulties arose in evaluation of the quality of the images by the panel members because of the differing criteria each chose to adopt. This highlights the need to define strictly, in advance, the parameters to be included in quality assessment. All three panel members agreed on the lesions present in 54 out of 60 studies. The observers' consensus report showed material differences from the original departmental report in the number of lesions present in 10 out of 60 studies. Material differences in the interpretation of the study occurred between panel members in two cases and between the panel and the original report in four cases. The pilot study has shown an acceptable level of consistency in bone scan reporting in the departments contributing images. A national study assessing consistency of reporting would be difficult to organize but alternative, more local, schemes are suggested.