Integrating ChatGPT into Pharmacology Simulation: A Structured Framework for Prescribing and Medication-Safety Education - A Narrative Review

Simulation-based training has become an essential part of modern pharmacology education, going beyond traditional lectures to include prescription-writing tasks, virtual patient interactions, OSCE/OSPE assessments, skills-lab exercises, and technology-driven simulations. These methods aim to enhance key skills such as rational prescribing, dose calculation, therapeutic decision-making, interpreting medication charts, ensuring drug safety, identifying Adverse Drug Reactions (ADRs), and patient counselling. However, providing high-quality simulationbased pharmacology education is often constrained by faculty time, the ongoing need for scenario development, logistical issues, and limited opportunities for students to practice repeatedly. Recent progress in large language models, especially ChatGPT, offers a scalable and flexible way to support pharmacology simulations. With guided structured prompts, ChatGPT can create virtual patient cases, prescription and drug-chart exercises, OSCE/OSPE-style stations, medication- error scenarios, acute toxicology cases, complex multi-step cases, and personalized feedback. This review presents ChatGPT not as a substitute for real or high-fidelity simulations but as a supplementary tool to help educators design, customize, and improve pharmacology simulation activities for teaching, practice, and assessment. The paper outlines a structured framework for simulation design, including aligning with learning outcomes, rational prescribing, medication history-taking, communication and counselling, clinical reasoning, interpretation of clinical and pharmacokinetic data, acute care and toxicology scenarios, workflow-based simulations, and assessment rubrics. Each part is illustrated through representative simulation scenarios and learner tasks to demonstrate potential implementation. While ChatGPT allows for quick case creation and focused practice, expert supervision remains vital to ensure clinical accuracy, adherence to guidelines, and responsible AI use.