Introduction

The traditional didactic forums that dominate veterinary medical education during pre-clinical studies are being reformed [1] through systematic curriculum planning that supports integrated teaching, core curricula with electives, and problem-based and case-based learning [2, 3]. These modifications have shifted healthcare education toward more constructive and collaborative learning [2, 3]. As a result, the focus on basic science courses during the pre-clinical veterinary curriculum has recently incorporated the horizontal integration of basic sciences with case-based learning (CBL) [4].

CBL has been defined as a teaching and learning approach that intends to prepare both pre-clinical medical and veterinary students for clinical practice through the use of real clinical cases [5,6,7]. CBL follows the same premises of being a learning platform that is student-centered and problem-centered with involvement of clinical reasoning skills and self-directed learning as in problem-based learning (PBL). Differentiations include that students in CBL receive additional guidance during deliberate in-class discussions whereas PBL encourages students to figure out what they need to know and to seek out appropriate resources [3]. Thus, CBL uses a guided inquiry method and provides more structure during small-group sessions whereas PBL uses an open inquiry approach where facilitators play a minimal role and do not guide the discussion [3, 8].

Case-based learning is delivered in small groups led by multiple facilitators where learners are presented with a clinical problem and the group focuses on problem solving. In this format, both students and instructors share responsibility for determining learning outcomes [4, 9]. However, this type of format is not practical in some educational institutions [4]. Commonly, within pre-clinical medical education, lecture-based CBL delivery methods have been reported [10]. In pre-clinical veterinary education, CBL has been successfully incorporated into first year veterinary curricula using active learning approaches delivered in lecture based sessions [4].

CBL has been used to integrate different subject matters delivered in the pre-clinical curriculum as well as help students apply learned information [11]. Another major objective of CBL is the development of students’ clinical reasoning skills [7, 12] where knowledge is commonly developed in science subjects and gradually integrated in CBL [13].

Clinical reasoning is defined as applying knowledge and expertise to develop a solution regarding the prevention, diagnosis, or treatment of a clinical problem in a specific patient [14]. The clinical reasoning process involves problem-centered learning, practice through repeated exposure and feedback. Clinical reasoning process skills must be developed in association with the acquisition of basic science and clinical information ensuring that problem-solving and knowledge are concurrently applied in the clinical setting [12]. As such, exposure to CBL for students in both medical and veterinary curricula enhances their development of clinical reasoning skills [15, 16]. Moreover, when pre-clinical veterinary students have repeated opportunities to apply the clinical reasoning process, their self-confidence increases significantly [7].

The objective of this study was to assess the effectiveness of case-based studies (CBS) in developing first year pre-clinical veterinary students’ clinical reasoning process skills through assessment and time spent.

Materials and Methods

The Ross University School of Veterinary Medicine (RUSVM) curriculum consists of 10 total semesters, with three, 15-week semesters per year (a semester begins in January, May and September). Following seven semesters of pre-clinical curriculum on St. Kitts, students complete their clinical curriculum at an affiliated American Association of Veterinary Medicine (AVMA) accredited University.

In January 2015, case-based studies’ (CBS) courses were introduced to the RUSVM pre-clinical curriculum. CBS is a required course and supports the horizontal integration of basic and clinical sciences by incorporating clinical cases commonly seen in the practice of veterinary medicine. CBS courses are pass/fail and a pass grade is awarded based solely on attendance.

Student Participants

Two cohorts of 77 third semester students volunteered to participate in the study. The control group consisted of 46 students from the January 2015 semester, as they were not previously exposed to CBS. The intervention group included 31 students from the September 2015 semester that enrolled and successfully completed both CBS I and CBS II. During an orientation meeting, all participants signed informed consent forms and completed a professional profile questionnaire that addressed student participants’ previous veterinary experiences in a veterinary hospital setting. Moreover, participants completed a self-assessment of their understanding surrounding the clinical reasoning process and specified how often they apply the clinical reasoning process. After completion of this profile, participants were provided a handout which included general history of the clinical case, three clinical reasoning assignments, a time-table and a schedule detailing due dates of each assignment.

Clinical Case

The case for this study presented a 3-year-old spayed female Australian Cattle Dog seen at a referral clinic with a history of nasal discharge, sneezing and lethargy. The case was selected by a board certified small animal practitioner and the principal investigator (PI) based on the following criteria: sufficient long and short-term history, physical examination outcomes, diagnostic results and the capability to reach a definitive diagnosis. All patient and client information was de-identified. The case diagnosis was nasal aspergillosis. For grading standardization between cohorts, one clinical case was selected.

Clinical Reasoning Rubric

Following the principles of Tanner’s clinical judgment model, and the Lasater clinical judgment rubric (LCJR) [17, 18], we developed a seven-item clinical reasoning rubric specific to the selected case. The rubric scored student participants’ skill levels on the following items: (1) gathering historical information, (2) gathering physical examination findings (3) prioritizing patient’s problems, (4) identifying differential diagnoses, (5) ranking differential diagnoses, (6) developing a diagnostic plan and (7) offering explanations based on the reported diagnosis. Each rubric level was scored following a five point Likert Scale [0-unsatisfactory to 4 exemplary] (Appendix 1 in Electtronic Supplementary material).

Clinical Reasoning Assignments

Based on the clinical case, each student was required to complete 3 assignments and submit via email to the facilitator (PI) (Appendix 2 in Electtronic Supplementary material). Assignment 1 required students to report (1) a comprehensive list of historical questions and (2) questions surrounding the patient’s physical examination findings. Subsequently, each student was able to present follow up historical questions based on the answered material received from their first completed assignment. Assignment 2 engaged students in prioritizing patient’s problems, identifying and ranking the differential diagnoses for each problem and providing a diagnostic plan. Based on the submitted information, requested and available diagnostic results were shared by email to each participant. The third and final assignment involved an explanation of the reported diagnosis. Furthermore, all participants were asked to complete a timetable that recorded the time spent on completing each of the 3 assignments. Upon completing all assignments, participants had the option to attend a one-hour presentation that discussed the selected case.

This study was completed in December 2015 and received research ethics approval the RUSVM Institutional Review Board.

Statistical Analysis

Data analyses were performed using the software R [19]. Participant information from the professional profile was analyzed using frequency distributions. Rubric scores were graded on a 0–4 scale and time was recorded as minutes. All explanatory variables were considered as quantitative variables. Variables on knowledge and application of clinical reasoning were considered as ordinal and also treated as quantitative variables. The Wilcoxon rank sum test was used to compare distribution of these variables between the two student cohorts. A power analysis was completed to support the non-significant data. The Pearson correlation test was used to evaluate the association between time and rubric scores. Associations were considered significant where p was ≤ 0.05.

Results

A total of 77 students participated in this study. Of these, 69% of participants (n = 53) have worked or volunteered in a veterinary hospital setting for less than 5 years, while 22 (29%) have worked for more than 5 years. Of the 75 (97%) students with reported experience in a veterinary hospital setting, 75% worked in predominantly small animal practice (n = 56).

While there was no difference in the previous veterinary experience between intervention and control groups (p > 0.1), a difference was observed in the self-reported assessment of knowledge and previous use of the clinical reasoning process. The CBS-exposed group reported a higher perceived knowledge (p = 0.04) and application (p = 0.03) of the clinical reasoning process (Table 1).

Table 1 Profile results showing students’ self-reported knowledge and application of the clinical reasoning process of both cohorts
Full size table

The total score of the clinical reasoning assignment between the intervention and control groups was not significantly different (p = 0.1). The control group had a significantly higher score on the explanation of diagnosis (p = 0.02). The predicted power analysis confirms that the sample size was adequate to support the non-significant data except for the items associated with prioritizing patient’s problems and identifying differential diagnoses (Table 2).

Table 2 Mean rubric item scores of both cohorts
Full size table

Additionally, results demonstrated that rubric item scores for both cohorts were significantly positively associated with the time spent completing assignment 2: specifically for the items of prioritizing patient’s problems (r = 0.27, p = 0.02), identifying differential diagnoses (r = 0.23, p = 0.04) and ranking differential diagnoses (r = 0.23, p = 0.04). Finally, the total rubric score of both cohorts was significantly associated with the time spent completing assignment 2 (r = 0.22, p = 0.05).

Discussion

The rubric scores were not significantly different for the control cohort as compared to the intervention cohort. It is possible that the overall structure and inconsistent clinical reasoning approaches followed within the CBS courses contributed to the lack of improvement of clinical reasoning skills for the intervention group. At the time of the study, the CBS courses (I and II) were taught on average by 11 facilitators and we acknowledge the variability in the delivery of their case presentations. Within CBS at RUSVM, there is a need to train faculty to lead structured and deliberate discussions based on developed case materials [20]. Also, the intervention participants completed their role at the end of their third semester at a time when there are more requirements due to final examination preparations and perhaps the students’ motivations surrounding their participation was influenced [21].

The clinical reasoning results for both the intervention and control cohorts surrounding the rubric item scores of prioritizing patient’s problems, identifying differential diagnoses and ranking differential diagnoses were significantly associated with time spent on assignment 2. Even though the total rubric score for both cohorts was not linked with exposure to CBS, it was significantly associated with the time spent completing assignment 2. Perhaps students adequately formalized ranked problems and differential lists which was reflected in their overall score. It can also be considered that 3 of the 7 rubric items were associated with assignment 2, which factored into the total rubric score. There is little supportive evidence proving that time spent improves performance, however it is even more important to note how effectively student time is used [22].

Responses from the question in the professional profile that asked students to rate their understanding of the clinical reasoning process were significant. Students with exposure to CBS I and II reported a higher score for their overall knowledge of the clinical reasoning process. Perhaps attending and participating in CBS I and II for 2 h per week over two semesters allowed a false perception of their true clinical reasoning process abilities [23].

A limitation to the study relates to the grading of the CBS I and II courses as they are pass/fail with no formal examinations and grading is based on solely attendance. As such, there was no identified reason for students to perform other than a genuine desire to learn. Perhaps this supports the outcomes associated with the explanation of the diagnosis for the intervention cohort. Therefore, identification of the students’ true clinical reasoning skills within the realm of the CBS I and II program is challenging. It should also be noted that both cohorts of students processed the same clinical case. A further limitation can relate to the self- reported time spent completing each assignment portion and future studies could be differently designed to improve the usefulness of data.

Overall, the outcomes for the intervention cohort did not reflect the predicted effectiveness of CBS. An explanation of the control cohort’s performance includes the John Henry effect, which proposes that control participants competed to meet the expectations of the intervention cohort [24].

Conclusion

CBL typically includes student-centered and problem-centered learning with development of clinical reasoning skills. Our study results as well as student and instructor feedback led to the transition of all CBS courses to follow a structured and deliberate problem-based student centered approach. We recognized that student-centered learning differs from lecture-centered learning with regard to general teaching and learning methods, student engagement with content and responsibility of learning [25]. Currently, the CBS courses include oral participation, clicker participation, class discussions, student group discussions and review of relevant scientific literature. Additionally, core objectives of CBS courses include horizontal integration, decision making and problem solving with a more consistent clinical reasoning approach.