British Journal of Radiology (2004) 77, 775-776
© 2004 British Institute of Radiology
doi: 10.1259/bjr/66388556
Training improves medical student performance in image interpretation
T J W Dawes, MA, MB, BChir
1
S L Vowler, MSc
2
C M C Allen, MD, FRCP
1 and
A K Dixon, MD, FRCP, FRCR, FMedSci
3
1 School of Clinical Medicine, University of Cambridge, Addenbrooke's NHS Trust, Cambridge CB2 2QQ, 2 The Centre for Applied Medical Statistics, University Forvie Site, Robinson Way, Cambridge CB2 2SR and 3 Department of Radiology, University of Cambridge, Box 219, Addenbrooke's NHS Trust, Cambridge CB2 2QQ, UK
Correspondence: Dr T J W Dawes, The Old Farm, Long Sutton, Hook, Hampshire RG29 1SX, UK
 |
Abstract
|
|---|
The aim of this study was to objectively quantify the effects of radiological teaching on student performance in interpreting radiological images, and to establish whether training location affects performance. 114 clinical students were asked to anonymously interpret radiological cases before and after 26 weeks clinical training. Improvements were assessed by comparing performance on the median question in each assessment using the Kruskal-Wallis analysis. Variations according to different placements were assessed by the Mann-Whitney U-test. There was a highly significant (p<0.001) improvement in the performance of the group with training. The proportion of correct answers to the median question improved from 8% to 43%. Differences between training placements were non-significant. Our study suggests that radiology teaching significantly improves student performance. Future work should look to define the contribution of "clinical exposure" towards this improvement, as well as the various teaching methods employed.
|
Introduction
|
|---|
There are few truly objective measures of the improvement in medical student performance during clinical training, let alone radiological training [1]. Our clinical course provides a relatively large amount of radiological exposure, both formal and informal (e.g. clinicoradiological meetings [2]). This study is one of the few assessments of students' performance in interpreting images before and after training.
|
Method
|
|---|
During the first week of their clinical training 114 students were divided into two groups and each asked to interpret five of 10 images. Each image was accompanied by the salient history and examination findings (e.g. Figure 1
) and overall covered a similar, balanced range of radiological investigations including plain films, ultrasound, CT, MRI and nuclear medicine. No answers were provided at that time.
View larger version (91K):
[in this window]
[in a new window]
|
Figure 1. An example of a typical question. History: 50-year-old patient who presents with a painless lump in the neck. A. What is this procedure? (1 mark) B. What is the structure marked by a cross? (1 mark) C. What is the abnormality? (1 mark) D. Name one possible cause. (1 mark) E. Name another site affected by this process? (1 mark).
|
|
Student teaching
All students then underwent 26 weeks of clinical training, including 8 weeks at both teaching and at district general hospitals during which students received formal pre-announced group teaching, and informal teaching by clinical firms.
At the end of training, 133 students were presented with all 10 radiological cases, 5 seen and 5 unseen. The cases were then presented and answered by a consultant radiologist, maximizing the educational experience of the "examination" [3].
Students remained anonymous throughout, contributing towards a group performance for each question rather than a performance for each student. The same questions were used so that performance on each question before and after training could then be compared (Kruskal-Wallis analysis). Variations according to different placements were assessed by the non-parametric Mann-Whitney U-test.
|
Results
|
|---|
The percentage of correct answers to the median question improved from 8% to 43% (Figure 2) – a highly significant (p<0.001) improvement in student performance.
View larger version (45K):
[in this window]
[in a new window]
|
Figure 2. The distribution curve showing the median scores for each question (n=50) in order of increasing performance before and after training.
|
|
After training, both groups of students performed significantly better on the images which they had seen before training (p<0.001). Their performance on the new images was significantly better than their performance on the pre-training images (p<0.001). By splitting the year in two and giving each half a different half of the question pool we were able to avoid the possibility that the post-training images were simply easier to interpret. Both halves of the student pool showed highly significant improvements on seen (p=0.01, p<0.001) and unseen (p<0.001, p=0.01) questions.
Differences in student performance depending on the hospital were seen, but the probability that these differences were due to chance alone is high (p=0.51 for surgery locations, p=0.24 for medicine locations).
|
Discussion
|
|---|
This study is unusual in that it provides objective, quantitative evidence that training improves students' skills in clinical radiology. What cannot be determined is the direct effect of the radiological teaching itself. Students' results could have improved by exposure to clinical practice. Provision of extra radiology teaching in one group of students, or gathering the same data sets in another medical school with less radiological teaching might distinguish this.
It is reassuring that minor variations in the site and manner of training show no significant differences as it is difficult to provide identical training opportunities to students.
Since radiological images, unlike a patient, can be standardized for all candidates they are useful to assessors and amenable to setting standards in student performance. These are reasonable aims of any assessment scheme, and among the aims set out by the General Medical Council. As such, they could become an increasingly useful surrogate for future medical examinations [4]. However, a large proportion of student learning is informal and depends on enthusiasm so the use of other less quantitative assessment techniques is crucial; making radiology an attractive option for students may be just as beneficial as the amount, or mode, of formal teaching.
|
Acknowledgments
|
|---|
We thank the clinical students who participated in this evaluation, Julian Evans, research technician, for his electronic skills and all those who contribute to radiological teaching in our programme.
Funding: in house; Competing interests: none declared.
Received for publication April 17, 2003.
Revision received March 2, 2004.
Accepted for publication April 20, 2004.
|
References
|
|---|
- Marchiori DM. Developing a clinical competency examination in radiology: parts I & II – test results. J Manipulative Physiol Ther 1999;22:63–74.[Medline]
- Leung DPY, Dixon AK. Clinicoradiological meetings: are they worthwhile? Clin Radiol 1992;46:279–80.[Medline]
- Morag E, Lieberman G, Volkan K, Shaffer K, Novelline R, Lang EV. Clinical competence assessment in radiology: introduction of an objective structured clinical examination in the medical school curriculum. Acad Radiol 2001;8:74–81.[Medline]
- Mullins ME, Will M, Mehta A, Novelline RA. Evaluating medical students on radiology clerkships in a filmless environment: use of an electronic test prepared from PACS and digital teaching collection images. Acad Radiol 2001;8:514–9.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
C.R. Lamb, D.U. Pfeiffer, and P. Mantis
Errors in Radiographic Interpretation Made by Veterinary Students
J Vet Med Educ,
January 1, 2007;
34(2):
157 - 159.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
