Key words radiology - reaching - education - COVID-19 - E-Learning
Introduction
On 1/30/2019 the World Health Organization (WHO) declared COVID-19a global emergency
[1 ]. In spite of significant efforts, SARS-CoV-2 spread around the globe resulting in
at least 102,177,365 cases and 2,209,313 confirmed deaths globally on the first anniversary
of the pandemic [2 ]. Contact restrictions and social distancing are necessary to break the chain of
infection. Consequently, at the height of the first wave in Europe, up to 85 % of
universities were not able to provide in-person learning because they could not meet
the strict hygiene requirements [3 ]. In addition to the significant impact on everyday life, the pandemic also resulted
in significant restrictions in medical education [4 ]. To avoid jeopardizing the education of students in the medium term, higher education
in Germany was forced “from the status quo into the digital world without proper preparation”
[5 ]
[6 ]. High-quality and effective digital teaching requires well-thought-out digital course
offerings and an adequate technical infrastructure [7 ]. However, the development and use of corresponding structures vary between individual
sites [8 ]. To provide ad hoc support to German-speaking medical faculties in this situation,
the German Radiological Society developed a course system tailored to the main university
course content called “CoRad-19” at the start of the pandemic [9 ]
[10 ]. The nine modular courses are comprised of a combination of lectures, theoretical
questions, and interactive cases and can be implemented by universities individually
or as a complete package to supplement any already offered courses. 13 medical schools
in Germany, Switzerland, and Austria decided to implement CoRad-19. While some universities
only used individual course modules, others replaced their entire radiology program
with the CoRad-19 courses. The goal of this study was to systematically determine
how students feel about e-learning, particularly in regard to radiology, and whether
measurable learning gains can be achieved with CoRad-19.
Materials and Methods
Target group
The target group included the participants in the CoRad-19 course modules at all participating
universities. The study includes the period from April 1, 2020 to October 1, 2020.
In terms of demographic data, we recorded the participants' native language, gender,
and age.
Survey and course modules
We asked participants about their prior experience and contact with radiology and
about their general opinion of e-learning and if they had previously participated
in digital courses. Participants were asked immediately before and after completion
of a course module to anonymously evaluate their personal performance. A 4-point Likert
scale (1 = disagree, 2 = tend to disagree, 3 = tend to agree, 4 = agree) was used
for the self-evaluation regarding every learning objective of the nine course modules
(see [Table 1 ]). Due to the anonymous nature of the survey, it was not necessary to obtain ethics
committee approval.
Table 1
Course modules and learning objectives.
Course modules and learning objectives
Thematic self-evaluations
Technology & radiation protection
Sonography, radiation protection, layout and function of an X-ray tube, layout of
a CT scanner, Hounsfield scale,
functionality of an MRI scanner, T1 and T2 weighting
8
Thorax radiology
Anatomy, CT, morphological pathologies in pulmonary artery embolism and pneumothorax
6
Abdominal radiology
Anatomy, morphological aspects of cysts and changes in the aorta
5
Angiography & interventions
Anatomy, biopsy and intervention, general pathology, and indications
4
Pediatric radiology
Anatomy, systematics, typical pathologies
4
Gynecological radiology
Systematics, sensitivity, typical pathologies
4
Musculoskeletal radiology
Bones in different modalities, arthrosis, fractures, tumors
5
Neuroradiology
Ischemia, disc prolapse
3
Final course “Radiological Routine”, mixed topics
Anatomy, systematics, typical pathologies, interest in the field of radiology
4
= Total
43
Statistics
Statistical analyses were performed with IBM SPSS Statistics Version 27 for Windows
(Armonk, NY, USA). Normally distributed variables are given as mean ± standard deviation
(SD), not normally distributed variables as median and interquartile range (IQR).
To improve accuracy, we focused on intraindividual comparisons in the individual modules.
Data sets with missing values in the individual categories were excluded. Normally
distributed variables were analyzed with a one-way repeated measure ANOVA, and not
normally distributed variables were analyzed with the Friedman test. An alpha correction
according to Dunn-Bonferroni was performed for the post-hoc tests. A p-value of < 0.05
was considered statistically significant. We calculated the Pearson's correlation
coefficient as a measure of the effect size (r ). Values from 0.1 to 0.3 indicated a small effect size, from 0.3 to 0.5a moderate
effect size, and ≥ 0.5a significant effect size.
For better comparability and optimized representation, we summarized the self-evaluations
regarding individual learning objectives for each course. In addition, we calculated
what percentage of participants gave a positive response (3 or 4) before and after
the modules and calculated the difference as an indicator of learning gains.
Results
Target group
At the time of the analysis, a total of 994 students had completed the self-evaluation.
451 complete data sets were included and evaluated intraindividually (45 %). The average
age of the participants was 25 ± 4 years.. The gender distribution was as follows:
152 male, 273 female, 26 not specified. German was specified as the native language
among 87 % of the participants, Italian among approximately 4 %, French among 3 %,
English among 1 %, and other languages among 5 %. See [Fig. 1 ] for further details.
Fig. 1 Data sets, previous experience, e-learning offerings used, contact with radiology.
Opinions regarding e-learning
E-learning was rated as “very useful” both before and after participation in CoRad-19
(4 [IQR 3–4], p = 0.527, r = 0.16). E-learning as a method was also rated as a “very good” medium both before
and after participation (4 [IQR 3–4], p = 0.414, r = 0.17). However, it is noteworthy that significantly more students rated radiology
as particularly suited for digital teaching after participation (before: 3 [IQR 3–4]
vs. after 4 [IQR 3–4], p = 0.005, r = 0.6). [Fig. 2 ] shows a graphic of the opinions of students regarding e-learning before and after
participation in our courses.
Fig. 2 Opinions regarding e-learning.
Courses
The intraindividual evaluation of the course modules showed that the self-evaluation
by participants was significantly higher after completing the courses than before
(≥ 10 % learning gain, Friedman Test: χ²(1) > 6.8, p ≤ 0.009, n ≥ 120). There were no differences between participants regarding native language,
gender, age, and previous experience in radiology (p = 0.861). The greatest learning gains were achieved in the course modules “pediatric
radiology” (47 %, p < 0.001, r = 0.67) and “musculoskeletal radiology” (44 %, p < 0.001,
r = 0.66). The observed effect was low (r = 0.24) only in the thematically mixed final
course “radiological routine”, which included various cases from all areas. However,
the self-evaluation prior to participation in the final course was significantly higher
compared to the self-evaluation prior to the start of the other courses (p < 0.025).
See [Fig. 3 ] for further details.
Fig. 3 Summarized self-evaluation regarding course learning objectives (n = assessments
* learning objectives).
Post-hoc analysis of the learning objectives
Particularly high subjective learning effects were achieved for the learning objective
“pathologies and tumors” in the module “musculoskeletal radiology” (72 %) and for
the learning objective “typical pathologies” in the module “pediatric radiology” (68 %)(in
each case p < 0.001, r > 0.7; see suppl. Table 1 ). We saw the lowest effect (10 % learning gain) in the thematically mixed final course
“radiological routine” (before 3 [IQR 2–3], after 3 [IQR 3–3], r = 0.24) in spite
of significant improvement. In this module the post-hoc analysis did not show any
changes regarding the learning objectives “anatomy” (before 3 [IQR 3–3], after 3 [IQR
3–3]; p = 0.763), “pathology” (before 3 [IQR 2–3], after 3 [IQR 3–3]; p = 0.285), and “systematics” (before 3 [IQR 2–3], after 3 [IQR 3–3]; p = 0.109). However, a significant improvement was observed regrading participants'
general interest in the field of radiology (before 3 [IQR 2–3], after 3 [IQR 3–4];
p = 0.02). [Fig. 4 ] shows the self-evaluations regarding the learning objectives in the final course
“radiological routine” before and after participation in the course module.
Fig. 4 Post-hoc analysis “Radiological Routine” before and after completion of the course.
Discussion
E-learning is an integral part of modern teaching. Even before the COVID-19 pandemic,
many students participated in digital learning on a supplementary basis. However,
the focus in medical education was on in-person learning. Because of the pandemic,
universities around the world had to fundamentally change their courses without warning.
The goal of this representative survey was to assess the self-evaluations of medical
students who had completed the CoRad-19 course system provided by the German Radiological
Society. We evaluated how the participants feel about e-learning and whether the modular
course system CoRad-19 results in measurable learning gains. The students we surveyed
had a highly positive view of digital teaching both before and after participation
in CoRad-19. Other current studies also reflect a positive attitude on the part of
students toward digital teaching. For example, Sud et al. showed that 97.2 % of the
students they surveyed see Web-based teaching methods as an adequate alternative to
in-person learning [11 ]. After participation in CoRad-19, significantly more participants were convinced
that radiology is particularly suitable for digital teaching. This coincides with
the results of other studies. For example, Häusler et al. were able to show that radiology
lectures and seminars can be implemented particularly effectively using a digital
format [12 ]. However, as in other studies, Häusler et al. came to the conclusion that digital
lectures and seminars are less suitable for teaching practical skills due to the low
level of interactivity [13 ]
[14 ]. The reporting of findings, one of the most important practical skills in radiology,
is performed, however, almost exclusively digitally in the daily routine. Therefore,
with corresponding interactivity, radiology reporting can be effectively taught and
learned on a digital basis. Nevertheless, digital teaching formats in radiology must
comply with data security and structural requirements. Therefore, in comparison to
other disciplines, much larger volumes of data per patient and examination must be
able to be stored and must also be able to be retrieved as dynamically and interactively
as possible [15 ]. In publications in other disciplines, the conversion to purely digital teaching
was described as particularly challenging due to insufficient technical infrastructure
[16 ]. In contrast, the radiological technical infrastructure was already comparably well-established
prior to the pandemic because it is a necessity in radiology. This presumably greatly
facilitated the seamless development and implementation of CoRad-19 [15 ]. Nonetheless, optimism tends to be mixed with caution with respect to the prompt
implementation of digitalization in other disciplines in Germany [6 ]. We conclude that radiology is suitable for digital teaching. However, it should
be noted that radiologyʼs technical requirements made it particularly well prepared
for a conversion to digital teaching. Like other research groups, we were also able
to establish that digital teaching results in measurable subjective learning gains.
For example, Kaur et al. were able to show that digital teaching methods were almost
exactly as effective among medicine students during the pandemic as classic in-person
learning [17 ]. Backhaus et al. indicated that students with a digital affinity have a significantly
more difficult time adjusting to traditional lecture formats than less digitally oriented
students [18 ]. Given the general increases in digitalization, future generations could see greater
implementation of digital teaching. However, Wilcha et al. correctly criticized that
exclusively digital courses are associated with less interaction between students
and between students and teachers [5 ]. Yet, working on cases together in learning groups could actually increase discipline-specific
interaction compared to traditional in-person learning. Häusler et al. concluded in
their study that a digital teaching unit should be thematically strictly limited to
the learning objectives [12 ]. Our results showed significant subjective learning gains in all topic-specific
course modules. The effect size of these significant improvements in the thematically
mixed final module “radiological routine” was the lowest among all courses. However,
this comparatively less pronounced effect can be explained by the fact that the self-evaluation
was significantly higher in this module than in all other modules already before participation.
Therefore, it can be argued that the other courses had prepared the participants for
the questions in the final course. The significantly increased by comparatively lower
learning gains in this module are presumably due to its position as the final course.
Participation in the CoRad-19 course program resulted in an increase in the interest
of participants in the field of radiology. In light of the current talent shortage
in all medical fields and professional societies, good digital teaching should be
prioritized – not just for our own professional society but also for other medical
associations and organizations. This study has a few limitations. The individual faculties
were able to decide for themselves which course modules to implement and how to adapt
their own curriculum to the courses. This resulted in unequal use of the individual
modules. A standardized, cross-location structure would certainly have resulted in
fewer incomplete data sets in the intraindividual evaluation. Furthermore, the subjective
self-evaluations were performed before and after each course module. As a result,
the long-term success may have been overestimated since the gained knowledge had just
been acquired. Evaluations completed with a time delay or for multiple modules could
have further limited this factor. The objectiveness of the subjective self-evaluations
assessed in this study could be increased in the future using surveys that include
multiple modules and are completed with a time delay. In summary, due to its extensive
experience with digitalization and networking, radiology was able to quickly provide
a very good digital curriculum during the pandemic. The e-learning courses were well
received and resulted in measurable learning gains.
Finally, it can be concluded that the German Radiological Society was able to offer
German-speaking medical faculties important ad hoc support in the form of CoRad-19
so that excellent teaching could be maintained during the ongoing COVID-19 crisis.
In addition, it should be noted that with these courses the German Radiological Society
was able to increase student interest in radiology in spite of the pandemic.
