Article Text
PDF
Abstract
Aims and Objectives CT Head scans are commonly requested in the Emergency Department (ED), but the increasing demand has led to longer radiology report turnaround times, affecting ED flow. This study aims to assess the accuracy of ED clinicians in interpreting CT head images, evaluate the impact of an online training simulation, and estimate the effects of clinician-led interpretation on patient flow.
Method and Design Methods A multicentre randomized controlled trial was conducted across six hospitals in the Thames Valley Emergency Medicine Research Network. Emergency medicine clinicians of various grades and backgrounds participated. In the online phase, participants completed a blinded baseline assessment of accuracy by interpreting 50 CT Head scans. Non-control participants received online training and practiced with 50 cases. They then retook the assessment to measure changes in reporting accuracy and confidence. Follow-up assessments were conducted at 3 and 6 months to assess knowledge retention. Training and assessment were conducted through the online platform www.raiqc.com. In the prospective phase, participants interpreted 30 CT head scans during their clinical practice, and their findings and interpretation times were compared to radiology reports.
Registration ClinicalTrials.gov ID NCT05427838
Results and Conclusion Results
A total of 206 participants took part in the study. The online phase showed a significant increase in pooled sensitivity (73.3% to 83%) and specificity (65.8% to 89.1%) in detecting acute abnormalities. Similar improvements were observed across all pathology subgroups. At six months post training, overall diagnostic performance remained elevated (sensitivity 80.6%, specificity 79.1%). In the prospective phase, 4,815 CT Head interpretations with linked radiology reports were recorded, and data analysis is ongoing.
Discussion and Conclusion Dedicated training significantly improved the interpretation accuracy of ED clinicians. Web-based self-directed simulation-based learning platforms can effectively deliver training, particularly in departments with high staff turnover. Further analysis of prospective results will assess accuracy and impact in a live ED setting.
