Development of a valid and reliable test to assess trauma radiograph interpretation performance

The purpose of this investigation was to develop and examine the preliminary validity and reliability among radiographers of a test to assess trauma radiograph interpretation performance suitable for use among health professionals.

Stage 1 examined 14,159 consecutive appendicular and axial examinations from a hospital emergency department over a 12 month period to quantify a typical anatomical region case-mix of trauma radiographs. A sample of radiographic cases representative of affected anatomical regions was then developed into the Image Interpretation Test (IIT). Stage 2 involved prospective investigations of the IIT's reliability (inter-rater, intra-rater, internal consistency) and validity (concurrent) among 41 radiographers.

The IIT included 60 cases. The median (interquartile range) clinical experience of participants was 5 (2-10) years. Case scores were internally consistent (Cronbach's alpha = 0.90). Favourable inter-rater reliability (kappa > 0.70 for 58/60 cases, Intra-class correlation coefficient (ICC) > 0.99 for total score) and intra-rater reliability (kappa > 0.90 for 60/60 cases, ICC > 0.99 for total score) was observed. There was a positive association between radiographers' confidence in image interpretation and IIT score (coefficient = 1.52, r-squared = 0.60, p < 0.001).

The IIT developed during this investigation included a selection of radiographic cases consistent with anatomical regions represented in an adult trauma case-mix. This study has also provided foundational preliminary evidence to support the reliability and validity of the IIT among radiographers. The findings suggest that it is possible to assess image interpretation performance of adult trauma radiographs with this test.