Teaching small animal reproduction via virtual patients

The project QuerVet focuses on improving the interdisciplinary lectures offered to veterinary students at the Faculty of Veterinary Medicine, Freie Universität Berlin, by presenting relevant virtual patients and cases in veterinary public health (VPH). In the German regulations of licensing to veterinary medicine (TAppV, 2006), the curriculum time available for interdisciplinary lectures is 196 hr; most of it in study semesters 6–8. These interdisciplinary lectures are supposed to offer case-based content that is relevant in veterinary practice. The listed clinical subjects, for example, should address health problems related to internal medicine, surgery and reproductive medicine with consideration of pathology, animal nutrition and anatomy. The clinical focus should lie on identifying the origin, diagnosis and therapy of diseases through specific cases (TAppV, 2006). Until 2016, the interdisciplinary lectures offered at our faculty did not follow a coherent course concept. Lecture slots were assigned to the different institutions and lecturers, and the selection of topics was heterogeneous. Most lectures were given as an in-class lecture with a duration of three hours, and less than half of the topics were presented in an interdisciplinary or case-based structure. In a Received: 1 October 2019 | Revised: 4 November 2019 | Accepted: 24 November 2019 DOI: 10.1111/rda.13598

survey in 2014, over 50% of the interviewed students evaluated the interdisciplinary lectures as not efficient for teaching interdisciplinary topics (Schunter, 2016).
Education of veterinary students must include practical clinical training on live animals. For this, students have to be prepared through a learning cascade starting with demonstration of certain practical procedures, repetition of those procedures under supervision and the refinement of the procedure. Since the use of animals in education is classified as an animal experiment according to the European Directive 2010/63/EU on the protection of animals used for scientific purposes (Directive, 2010/63/EU, 2010 and this has been implemented into the German Animal welfare act in 2013 (German Animal Welfare Act, 2006), it has to be applied for authorization to the competent authorities. Everyone using animals for educational purposes is obligated to implement the 3R principles, which means to replace animal use by alternative teaching methods wherever possible (Replace). If animal use is unavoidable, the number of animals has to be reduced to the minimum requisite (Reduce) and techniques should be used which pose the minimal burden to the animal (Refine).
Virtual patients are 'interactive computer simulations of real-life clinical scenarios for the purpose of medical training, education, or assessment' (Ellaway, Candler, Greene, & Valerie, 2006).
It has been shown that the interactive format of virtual patients enhances diagnostic and clinical reasoning skills of veterinary students (Kleinsorgen et al., 2018) and that virtual patients make students feel better prepared for real patients with similar clinical symptoms (Lehmann et al., 2015). The combination of virtual patients with face-to-face teaching is important for their successful integration, which is called 'blended learning' (Lehmann et al., 2015). More benefits of virtual patients are avoiding risks for patients and learners, standardizing learning experiences and mediating learning experiences using practice-like situations (Ellaway, Topps, Lee, & Armson, 2015).
Virtual patients, therefore, could offer a feasible solution to the problems of interdisciplinary education as well as help to further refine and reduce animal experiments in practical veterinary education.
However, to this extend virtual patients have not yet been systematically integrated in veterinary education for teaching interdisciplinary lectures in a mandatory blended learning format in Germany.
Our aim was to develop and implement an interdisciplinary course with virtual cases starting with a small animal reproduction module consisting of three virtual patients. They were implemented in a blended learning concept, adapted to the existing curriculum and addressed topics relevant in veterinary practice.
The evaluation of this reproduction module by students focussed on the acceptance of virtual patients. For this context, we examined the didactic use of media and the usability and the self-assessed learning outcome. Also, we analysed the assessment of the blended learning concept in an overview of the complete semester with several modules.

| First step-Choice of Topics and learning objectives
Relevant clinical topics for the virtual patients were identified in a survey sent to veterinarians working at Freie Universität Berlin and other veterinary clinics or practices in Germany. From this list, topics were chosen in consultation with the clinical lecturers at our faculty for the blended learning approach. According to the specifications for interdisciplinary lectures (TAppV, 2006), we created a general case structure and then aligned it to real scenarios with specific diagnoses. We defined learning objectives for each case and made them available to the students at the beginning of the case. The learning objectives were categorized according to Bloom's Taxonomy for learning objects for systematic classification of the process of learning (Bloom, 1976). In addition, for each specific topic, we created individual modules containing different virtual patients or cases.

| Second step-Curriculum and implementation
The curriculum was analysed for the most appropriate time to implement the specific topics, so that the students are able to work on each module while having enough prior knowledge. This analysis was a first step to establish a learning spiral, that is that specific information will be refreshed and enhanced during the curriculum.
Furthermore, the relevant prior knowledge necessary to understand and work on the case was listed at the beginning of each case. For the implementation of the modules within this blended learning approach, free slots were provided in the timetable for students to work on the online cases. However, they could also use other times during the respective time periods, generating a high flexibility in time and place. Students had usually one week to work on each case.
To increase autonomous and active learning in the online phase, the students received virtual patients with necessary background information and if needed the opportunity to get support for technicalor content-related problems. To enable face-to-face discussions on a higher level during the in-class lecture, we scheduled one 90-min lecture after the completion of the reproduction module with three online cases.

| Third Step-Learning platform and case structure
We used the learning platform tet.folio, an online web application developed at Freie Universität Berlin, Department of Physics, Physics education unit (Haase, Kirstein, & Nordmeier, 2016). It was created with an emphasis on customized interactive content. Students work page by page through the online course-book. Answers, notes and other activities are always saved individually. The platform was integrated into the learning management system Blackboard (Washington DC, USA) used at the Freie Universität Berlin. This had the advantage that students had a familiar platform to access the modules, that tet.folio activities were automatically transferred into the Blackboard grade centre and that users only needed to login once (single-sign-on).
Within the case structure, realistic scenarios of decision-making and assessing clinical findings were simulated. In addition, feedback tools were implemented to provide the students with visual and textual feedback depending on the correctness of the given answer or decision. For each virtual patient, a glossary with relevant background information including hands-on videos and animations regarding the patient and its diagnosis was created and selectively linked to the suitable position within the case. Also, students had the opportunity to use a virtual notepad to keep their notes about the patient at hand all the time. After completion of the case, students were given access to download and print a modified glossary.

| Fourth step-Material and media
Whenever possible, already existing material such as radiographs, ultrasound images or videos of patients and their symptoms was used to build realistic virtual patients.
In addition, educational videos including videos showing practical procedures (hands-on videos) or short lectures as well as animations were created to give the students more insight into or understanding of specific topics regarding the patient.
Interactive elements such as an abdominal ultrasound examination and a swab-sample taking were created in a close cooperation with the Department of Physics. For the interactive element of the abdominal ultrasound, for example, the movable picture of an ultrasound probe was linked to the playback of a sagittal abdominal ultrasound of a bitch using a specific html5 programming routine.
The students were asked to move the probe with the mouse cursor to control the video which would move accordingly to the student's mouse cursor motion ( Figure 1).

| Fifth step-Realization
For the clinical part, one pilot case concerning small animal reproduction and a related in-class lecture for further discussion were implemented. In a short introduction prior to the implemented case, the concept and web application were explained, and students had the opportunity to log in with our assistance and ask further questions.
The following year, two additional cases were added to this reproduction module. After completion and implementation of this reproduction module, students worked on three virtual patients in a three-week online period and finished the module with an interactive in-class lecture for discussing questions and problems ( Figure 2).

| Ethic votum and animals in research
An ethics proposal for the evaluation by the students was submitted with the application number EA4/125/18 and approved by the Ethics Committee of the Charité at Campus Benjamin Franklin.
All animal-patient-related examinations, procedures and handling protocols used in the respective modules were approved by the local competent authorities, which is in Berlin the Regional Office for Health and Social Affairs (LAGeSo, L 0001/17). All applicable national and institutional guidelines for the use and care of animals used in education were respected. Experimental procedures were assessed as minor discomfort experienced by the animals, and therefore the severity was categorized as mild.

| Evaluation
For the pilot case, we chose a more complex survey, which, in addi- The results of the surveys were analysed and used for improvement of the following cases.
All surveys were anonymous and voluntary.
After having conducted the pilot case, newly developed e-learning cases were continuously evaluated with both paper-based and online surveys depending on the phase within the modules, but the size of the surveys was reduced.

F I G U R E 1
Screenshot of an interactive element in the pilot case. By moving the probe up and down with the mouse cursor the ultrasound-video on the right moves accordingly. Students were asked to analyse the ultrasound findings and determine if the bitch is pregnant F I G U R E 2 Chart showing the blended learning concept of the reproduction module with an introductory course, an online phase for working autonomously on the three virtual patients and an in-class lecture for further questions

| Statistics
Responses to Likert-type questions were expressed as overall, positive and negative proportions per item. The internal reliability was calculated with Cronbach's α coefficients, and a value greater than 0.60 for all scales was valued for a good reliability (McKinley, Manku-Scott, Hastings, French, & Baker, 1997). The binominal 95% confidence intervals were calculated after Wilson (Agresti & Coull, 1998). All answer proportions were considered significantly different from values outside their confidence intervals. For all statistical calculations and the creation of related figures, we used the software R version 3.6.0 (R Foundation for Statistical Computing) using the package sjPlot version 2.7.2 (Lüdecke, 2018).

| RE SULTS
The compilation of the first virtual patient started in July 2016 as a pilot case and was first released and extensively evaluated with students of  the 7th semester in February 2017. The assignment was an ovulation timing and pregnancy diagnosis with suspected luteal insufficiency.
The two additional virtual patients addressed pyometra and benign prostatic hyperplasia and were added to the module of reproduction in February 2018. Significantly more than 90% of the students found the following for the support contact button with a percentage of negative answers of 28% (Figure 4).

| Usability
Usability was analysed for the pilot case, respectively, the en- Significantly more than 90% of the students agreed on a high usability in all five items. Most of the students chose strongly agree.
The internal reliability of the scale usability with five items was also high with Cronbach's alpha at α = .87 ( Figure 5). The level of difficulty of the implemented exercises was also evaluated with a response scale consisting of items of the Likert type (1 = far too easy to 5 = far too difficult). On this scale, more than 88% of the students agreed on exactly the right degree of difficulty.

| Self-assessed learning outcome
The  The work on this online case was associated with clear tasks and objectives. (n=292) All important contents were made available for the work on the tasks in the online case. (n=297) The texts in this online case were well structured. (n=294) The texts in the online case were easy to read and understand. (n=298) I have found all necessary information in the glossary of this online case. (n=299) 100% 80% 60% 40% 20% 0% 20% strongly disagree disagree slightly disagree slightly agree agree strongly agree

| Blended learning concept
The showed that significantly more than 70% of students were satisfied with the new concept and its implementation, preferred the blended learning over in-class lectures and hoped for expansion of the concept. The second statement holds an opposite significance as to revise the student's awareness. A blended learning subscale was not calculated, as the items covered many different dimensions (Figure 7).

| D ISCUSS I ON
The interdisciplinary lectures at the Freie Universität Berlin in 2014 were evaluated as being suboptimal and have not been based on a coherent course concept. Apart from that, the requirements for clinical veterinary training in general regarding animal welfare have changed.
Therefore, the task was not only to successfully create a consistent course concept that fulfilled the teaching requirements but also to implement virtual patients in a blended learning format.
For this reason, an extensive course concept was developed. In addition, we assessed the acceptance of this format by students and evaluated whether virtual patients were suitable for teaching interdisciplinary lectures in a mandatory blended learning format.
According to the evaluation results, students found video lectures, hands-on videos, animations and the glossary very useful, which showed that students appreciate this kind of supporting media for virtual patients. The videos and glossary not only give background information to a patient, but also support understanding contexts and applying knowledge to relevant situations. The lowest number of votes and lowest rating concerning usefulness was given to the button for support. The lowest rating can presumably be explained by the fact, that the button was not needed or used very often. This can also be the reason for the few votes as well. Students  The face−to−face sessions about the online cases was sensibly linked to the contents of the online cases.
I hope that in the future, attendance dates will be increasingly replaced by online activities in other courses as well. (n=180) Overall, I learned more through the combination of face−to−face sessions and online activities. (n=181) I would have learned more if the topics and tasks had been covered in face−to−face sessions, instead of online. (n=180) The overall concept of the course (online cases and associated face−to−face sessions) was sufficiently explained (e.g. allocation of tasks to face−to−face sessions and online activities The usability of the three cases was graded as very high concerning the comprehensibility, structure, availability of necessary information, the presence of clear tasks and aims and the difficulty of the exercises. In this context, the usability of case-based learning modules is essential for a good acceptance of virtual patients in education. When students were asked to assess their learning outcome ( Figure 6), the first category (knowledge) of Bloom's Taxonomy received the highest agreements. Knowledge means remembering facts and is considered being the lowest category. It may be concluded that the presented teaching concepts mainly focus on knowledge acquisition. However, also the second (comprehension) and even the third category (application) received agreements of more than 80%. These categories are regarded as relatively high concerning the level of understanding and application. It was our aim to enhance autonomous and active learning on higher levels with this format of education. In terms of a spiral curriculum, the information learned in this project will be revised and enhanced during the clinical year. In future studies, it should be assessed if teaching on virtual patients will support the learning outcome of real-life clinical patients learning.
The blended learning concept received a high acceptance and over 80% of the students support an extension of the concept to other courses. This concept was implemented to give students more flexibility as well as face-to-face discussions on a higher level.
Results show that teaching reproduction with virtual patients is

ACK N OWLED G EM ENTS
The authors would like to thank the initiators and authors of the Berlin.

CO N FLI C T O F I NTE R E S T
None of the authors have any conflict of interest to declare.

AUTH O R CO NTR I B UTI O N S
MD, LV, VD, CH, SB designed the study. LV and MLW planned and submitted the request for the animal experiments. LV, VD, CH, PH, SA and SH carried out the study. LV, AB, SA and MLW drafted the paper. LV and AB analysed the data. All authors discussed the results and contributed to the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
For privacy reasons, the raw data cannot be shared.