Novel training approach to improve a cohort of radiographers' image interpretation skills of trauma chest radiographs

Introduction

Zambia is experiencing a critical shortage of radiologists responsible for interpreting X-ray images. Nine radiologists serve the entire population of over 18 million people. Consequently, referring physicians can receive reports late and often receive X-ray images without radiological reports attached, which may lead to delayed diagnoses and treatment of critically injured patients. This challenge could be alleviated if radiographers could assist with interpreting X-ray images. This study was undertaken to subject a cohort of Zambian radiographers to a training intervention, however, the COVID-19 pandemic necessitated using a novel approach to the intervention by delivering the training mainly through social media but also through face-to-face lectures.

Methods

A cohort of 27 radiographers employed at eight public hospitals in the Copperbelt Province of Zambia undertook a training intervention using face-to-face training and image discussions on the social media WhatsApp® platform. The participants underwent a pre-and post-test in which they were asked to interpret 20 adult trauma CXR images. For the training intervention, the radiographers attended a face-to-face image interpretation lecture, after which they received training images with a radiologist report weekly for eight weeks via the WhatsApp® platform. Participants were encouraged to discuss and pose questions via the platform.

Results

The cohort of radiographers (n = 27) showed an improvement in their interpretation skills for trauma CXR images. The interpretation median scores ranged from approximately 82% to 93% in the pre-test and 85% to 97% in the post-test. The Wilcoxon signed-rank tests revealed significant differences in the interpretation ability skills for 12 of the 20 CXR images after the 8-week training, demonstrating the successful implementation of the program. When comparing three categories of radiographers' years of experience (1–5; >5–10; and >10 years), the Kruskal Wallis test could not identify significant differences in the CXR image interpretation skills among the different categories of experience (P = 0.1616). When comparing the interpretation skills of radiographers working at the three different hospital levels (Level 3 with a full-time radiologist and more than ten radiographers; Level 1 and 2 without a full-time radiologist; Level 2 with six to ten radiographers; and Level 1 with five or less radiographers), the Kruskal Wallis test revealed that the level of the hospital where the radiographers were employed significantly influenced their skills to interpret the CXR images (P = 0.0323).

Conclusion

This type of novel training intervention is urgently required in the Copperbelt Province of Zambia. The results show that the training process was implemented successfully to improve radiographers' image interpretation skills of adult trauma CXR images.

Implications for practice

Promoting radiographers' involvement in image interpretation will likely improve imaging services in Zambia, considering the critical shortage of radiologists.