Before the experiment began students were asked to acknowledge their willingness to participate in the study by signing a consent form. Anonymity of the study participants was ensured by a simple identity encoding system. The study was approved by the ethics committee of Jagiellonian University.
During the course students from the experimental group were granted access to a VP module on the CASUS® platform [13
]. The web-based VP shell developed by LMU University of Munich and Instruct AG has frequently been tested in large scale projects and studies in various medical fields [11
]. The CASUS system enables the authoring of VP cases, course management, as well as the tracing of students' activities. All students were given technical instruction on how to use the CASUS platform and practised it several weeks before the study began in the Basics of Computer Science course, working for at least one hour on non-BLS relevant cases.
In order to evaluate the course's educational outcome, all students were tested in BLS-AED knowledge and skills before the course started and after the last meeting of the course. Both tests were unannounced.
The pre- and post-course knowledge tests consisted of 60 true-or-false BLS-AED questions, randomly selected from a pool of 120 questions verified by subject matter experts. The questions were randomised in order to impede the students from memorising them. For each correct answer one point was scored. Both tests were carried out using the Blackboard Academic SuiteTM system and were supervised by instructors to prevent students from communicating or using external knowledge sources while taking the test.
The pre- and post-course BLS-AED skills were evaluated using recommendations taken from the standardised Cardiff Test [16
]. Each student was given the same scenario: „You are in a shopping mall, suddenly you hear a noise and see a 60-year old woman who collapses in front of you. While falling she knocks a few tins from a shelf”. Additional information for the instructor was: (a) When the student shouts for help a helper immediately appears; (b) The victim is unconscious, not breathing and pulseless when checked; (c) On request AED is available at once. The scenario was terminated after three minutes, including about two minutes BLS time. In cases in which an AED was used the scenario was terminated after two minutes of BLS following first shock delivery. The initial rhythm was always shockable. At the end of their performance students were asked to list criteria for the termination of BLS. Chest compressions and ventilations were performed and recorded on Laerdal Resusci Anne Skill Reporter, on request students received the Powerheart AED G3 Semi Automatic training device by Cardiac Science.
Usage of the VP cases was voluntary, but students were encouraged in the lectures to utilise the module to prepare themselves in advance for traditional face-to-face classes. To access the cases students needed to apply for an individual, anonymous account on the platform (no personal data was stored in the system). VPs accompanied a traditional face-to-face instructor-led BLS with AED course consisting of five classes (60 min each). The e-learning module initially contained one single VP case. A new VP case was added to the module every week of the course corresponding to the topics to be taught in the current week. By the end of the course six VPs were available. Table
presents the key learning objectives of the cases. All virtual patients were “linear string of pearls” [2
] cases consisting of 8 to 17 screen cards with text, images, videos, pop-up windows with additional comments and expert advice, as well as formative questions with feedback. The animations and videos presented key BLS-AED interventions which were also demonstrated in the traditional face-to-face classes. The theoretical foundations of multimedia learning by R. Mayer [17
] were followed in the design of the module: text was broken into small segments (screen-cards), essential concepts were highlighted and the text was illustrated by relevant images and animations. Figure
presents a sample screen card in the VPs module. It is important to emphasize that the VP module did not contain topics which were not presented in the lectures or the traditional face-to-face classes. Instructors were blinded regarding the study group allocations. In case of technical problems with the VP module usage, students were advised to contact the local CASUS system administrator by e-mail or directly in his office.
Usage of virtual patient cases by students from the experimental group
Figure 1 A sample screen card in the VPs module. A sample screen card in the VPs module implemented using the CASUS system. A question has been incorrectly answered and feedback with correct answer is displayed. The VP content is in Polish, user interface is selected (more ...)
Student satisfaction was measured with questions from a survey proposed by Huwendiek et al.
to evaluate curricular integration of VPs [18
Statistical analysis was carried out using the Statsoft Inc., STATISTICA 10 package. A significance level α of 0.05 was chosen. Knowledge test results in the study groups were compared using ANOVA for repeated measures test. When a significant effect was found, post hoc comparisons with Newman-Keulus test were applied. Dichotomous results of skill assessment were analysed using χ2 for 3×2 contingency tables. Effect size (ES) was calculated using Cohen’s d with pooled standard deviations. Correlation between data was checked with Spearman’s rank correlation coefficient.
All free-text opinions regarding VP integration given by students were carefully read. Using these opinions as its basis, a list of common themes was collated, keywords were assigned, and this was used to index the opinions. Based on the results, less frequent themes (n
3) were excluded or merged to form more general one.