|ECR 2018 / C-0669|
|Radiology residents' skill level in chest x-ray reading|
A full consensus between experts was obtained in 24 out of 40 CXR, including 10 CXR from the detection category, 10 CXR from the interpretation category and 4 CXR from the normal category.
All residents accepted to participate and all completed the experiment. There were 37% PGY 1-2 and 63% PGY 3-5 residents. Twelve out of 81 (14.8%) residents attended at least one CXR training session during their residency. Thirty-one out of 81 (38.3%) residents read on average more than 10 CXR per week. Seventy-four (91.4%) residents expressed a need for further formation in CXR during their residency. Full information about residents is reported in Table 1.
Residents’ overall success rate:
The overall success rate of residents in evaluating the normality of radiographs was of 90.4%, including a success rate of 92.6% for abnormal CXR (i.e. detection and interpretation categories) and of 79.6% for normal CXR (Table 2). When they were asked whether a complementary CT study was necessary, the residents overall succeeded in 76.6% of cases (74.3% and 88.6% for abnormal and normal CXR, respectively). A true final diagnosis was found in 52.7% of all cases, including a success rate of 47.4% for abnormal CXR and of 79.6% for normal CXR.
Residents’ success rate according to detection and interpretation CXR categories:
Regarding the normality of CXR and the need for additional CT study, residents significantly performed better in the interpretation CXR category than in the detection CXR category (Table 3). For the third question related to final diagnosis, there was no significant difference of residents’ performances between the detection CXR category and the interpretation CXR category (49% vs 45.7%, respectively; P= 0.156).
Factors influencing diagnostic performances of residents:
The results of this section are presented in Table 4.
Final diagnosis was most frequently found by PGY 3-5 than PGY 1-2 residents in the detection CXR category (52.2% vs 43.7%, respectively; P= 0.025), but there was no significant difference in the interpretation CXR category (Figure 3). The attendance at a CXR training course during residency did not influence residents’ success rate whether for deciding on radiograph normality or for asking complementary CT. For providing final diagnosis, training course was associated with better success rate in the interpretation CXR category only (56.7% vs 43.6%; P= 0.031) (Figure 4). There was no influence of the number of CXR read per week, whatever the CXR category or the level of diagnosis (Figure 5). Lastly, there was no significant difference in success rate for any of the three questions according to residents’ university hospital.
Thematically related posters
ECR 2018 / C-0918
Can Radiological Technologists Serve as Primary Screeners in Low-dose Lung Cancer CT Screening? (Analysis Based on the Results of Certifying Examinations over a 10-year Period)