A Virtual Patient Encounter to Promote Integration in a Preclinical Case-Based Learning Session

Abstract

In medical education, virtual patients increase the realism of learning in a safe environment. We added an integrated learning event using a virtual patient to integrate patient history taking into a preclinical basic science course. Herein, we describe the process and our overall satisfaction with the virtual patient encounter.

Clinical medicine and basic science education are increasingly being integrated in preclinical medical curricula, as integration is firmly established as promoting highly effective learning [1]. The challenge of providing preclinical medical learners with patient interactions during the COVID-19 pandemic led us to explore the use of virtual patients (VPs) to simulate patient encounters [2]. VPs are defined as a computer program that simulates real-life clinical scenarios in which learners emulate the roles of healthcare providers to obtain a history, conduct a physical exam, and make diagnostic and therapeutic decisions [3].

At our institution, preclinical basic science courses utilize weekly case-based learning (CBL) sessions that integrate clinical decision-making and ethics with the basic sciences. However, student practice of taking patient histories and performing physical examinations has been restricted to our “clinical skills” course. We recently added an encounter with a VP to an existing inflammatory bowel disease CBL session in the gastrointestinal system course to both promote integration and increase the realism of the case by simulating a telehealth visit.

The first 5 min of the 2-h CBL involved the faculty demonstrating the mechanics of interacting with the VP. Then, each group of six students was given only the patient’s chief complaint and the link to the website for the VP. Groups were allotted 25 min to interview the VP to determine his symptoms, medical history, and physical exam data and use this information to generate a preliminary differential diagnosis. Groups reassembled after the VP interview for a 15-min class-wide discussion that included significant inter-group debate regarding the optimal differential diagnosis. The CBL proceeded similar to a “standard” case at our institution, without any additional interface with the VP, to discuss the pathophysiology of disease and management/treatment, as the patient’s laboratory results and treatment outcomes were progressively revealed.

To create the artificial intelligence (AI)-driven VP that interacts with learners in a way that simulates actual, non-scripted conversations, we contracted with a company that specializes in AI-powered learning for medicine (www.recourseai.com/). The faculty wrote realistic patient responses to the myriad of patient history questions that might be asked by the learners and determined the appropriate physical exam findings. This was programmed into the VP software, allowing the VP to respond to verbal inquiries about his symptoms and history. At the conclusion of this build phase, the faculty tested the VP and were satisfied with his responses to their normal clinical lines of inquiry.

When the integrated learning session was piloted with a full cohort of 84 learners in 2021, the session went relatively well despite the VP occasionally misunderstanding the question being asked and thus responding inaccurately. Additionally, the faculty realized the importance of emphasizing the integrated aspect of the case when introducing the session and the need to address (unfounded) student concerns about grading.

When the integrated session was run with the subsequent cohort of 84 students in 2022, it included an improved introduction by the faculty, and machine learning had made the VP’s responses noticeably more accurate. Students were surveyed to understand their perception of this learning event and solicit opportunities for continued improvement before potentially adding similar learning events to other preclinical courses. Ninety-one percent of learners agreed/strongly agreed that the VP interview “offered a chance to learn in a safe environment,” and 67% agreed/strongly agreed that the VP “made this CBL more enjoyable than the average.” Student written comments included that the simulated patient encounter increased the realism of the case and their appreciation of the small group aspect of taking a patient history. All participating faculty observed learners were more engaged and seemed to enjoy themselves more than the average CBL.

In summary, the faculty were very satisfied with the educational value of the VP encounter that integrated student taking of a patient history and physical into a basic science course. While others have previously integrated clinical skills into basic science courses and CBLs, we believe our use of a VP to take an extensive patient history in such a setting is novel. The session described herein also partially simulated a telehealth visit and is a “lower resource” method than using standardized patients or high-fidelity simulators. The main limitation of this approach is the VP is a computer-based simulation, and the occasional non-logical responses are off-putting to a subset of learners. Other limitations include learner discontent with the idea of integrating “clinical skills” into preclinical science courses and learners feeling that a “lower resource” solution that partially replicates a standardized patient or high-fidelity simulator with a VP is an unworthy method of learning.

Declarations

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional review board and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study (#2021–0797) was approved as exempt by the Western Michigan University Homer Stryker M.D. School of Medicine institutional review board.

Consent to Participate

Informed consent was obtained from all individual participants included in the study.

Conflict of Interest

The authors have no relevant financial or non-financial interests to disclose.