Are Artificial Intelligence Virtual Simulated Patients (AI-VSP) a Valid Teaching Modality for Health Professional Students?

Abstract

Introduction: Simulation-based learning is robust, but the COVID pandemic created opportunities for novel modalities, including Artificial Intelligence Virtual Simulated Patient (AI-VSP) scenarios. Methods: Between 2019 and 2022, the following health professional students experienced AI-VSP at one US university: (a) “Headache” scenario: 24 Family Nurse Practitioner (FNP) and 48 Physician Assistants (PA) students, (b) “Insomnia” scenario: 64 Bachelor of Science in Nursing (BSN) and 47 Accelerated Bachelor of Science in Nursing (ABSN) students. Each individually conducted a brief strongly encouraged encounter and subsequently optionally participated in our study. Results: When asked about the scenario's realism, positive answers were 50% (FNP), 16% (PA), 63% (BSN), and 87% (ABSN). Also, 41% FNP, 52% PA, 65% BSN, and 82% ABSN felt capable of creating diagnoses and treatment plans as thoroughly as they would with human patients. Regarding improving diagnostic abilities, favorable responses were 73% (FNP), 74% (PA), 72% (BSN), and 90% (ABSN). When asked whether they would recommend AI-VSP encounters to others, 91% (FNP), 84% (PA), 93% (BSN), and 90% (ABSN) agreed. Conclusions: AI-VSP scenarios were well accepted by students and demonstrated significant promise as a complementary simulation-based learning modality.