Abdominal radiograph preliminary clinical evaluation image test bank project

Abstract

Introduction
Literature documenting preliminary clinical evaluation (PCE) commonly focuses on the evaluation of musculoskeletal radiographs. Despite this, the professional body suggest that a diagnostic radiographer should be able to provide a PCE for any radiograph.

Methods
An image bank of 30 abdominal radiographs was designed comprising of 17 abnormal cases with a range of pathologies which one could expect to encounter in the emergency department (ED). Participants’ were asked to select one of four taxonomies to represent their PCE for each radiograph. Participants’ answers were compared to a gold standard PCE taxonomy based on the radiological report. Inferential statistics were applied to assess for any significant different in accuracy between NHS pay bands of the participants.

Results
On average participants selected an abdominal radiograph PCE taxonomy with a sensitivity of 75.2% and a specificity of 75.7%. Whilst band 7 radiographers selected the most accurate abdominal radiograph PCE and had the highest area under curve (AUC), no significant difference was found in the PCE categorisation of abdominal radiographs by radiographers of all pay bands.

Conclusion
Participants’ have shown good sensitivity in recognising prominent findings on abdominal radiographs. This sensitivity is however reduced when assessing less obvious radiographic appearances, illustrating areas where additional training would be beneficial. The study provides evidence towards the consideration of an expansion of current practice regarding the implementation of a scheme of abdominal radiograph PCE. Further research with a larger cohort of participants’ and a lower abnormal case prevalence would be beneficial to the limited research base.