Blueprint for Constructing an AI-Based Patient Simulation to Enhance the Integration of Foundational and Clinical Sciences in Didactic Immunology in a US Doctor of Pharmacy Program: A Step-by-Step Prompt Engineering and Coding Toolkit
While pharmacy education successfully employs various methodologies including case-based learning and simulated patient interactions, providing consistent, individualized guidance at scale remains challenging in team-based learning environments. Artificial intelligence (AI) offers potential solution...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Pharmacy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2226-4787/13/2/36 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | While pharmacy education successfully employs various methodologies including case-based learning and simulated patient interactions, providing consistent, individualized guidance at scale remains challenging in team-based learning environments. Artificial intelligence (AI) offers potential solutions through automated facilitation, but its possible utility in pharmacy education remains unexplored. We developed and evaluated an AI-guided patient case discussion simulation to enhance learners’ ability to integrate foundational science knowledge with clinical decision-making in a didactic immunology course in a US PharmD program. We utilized a large language model programmed with specific educational protocols and rubrics. Here, we present the step-by-step prompt engineering protocol as a toolkit. The system was structured around three core components in an immunology team-based learning activity: (1) symptomatology analysis, (2) laboratory test interpretation, and (3) pharmacist role definition and PPCP. Performance evaluation was conducted using a comprehensive rubric assessing multiple clinical reasoning and pharmaceutical knowledge domains. The standardized evaluation rubric showed reliable assessment across key competencies including condition identification (30% weighting), laboratory test interpretation (40% weighting), and pharmacist role understanding (30% weighting). Our AI patient simulator offers a scalable solution for standardizing clinical case discussions while maintaining individualized learning experiences. |
|---|---|
| ISSN: | 2226-4787 |