Featured e-Learning Project
CustomGPT: AI-Powered Wound Care Discharge Simulation
I developed an AI-powered simulation using ChatGPT to help nurses, physician assistants, and discharge educators practice delivering clear and empathetic post-op wound care instructions. The simulation is guided by a fictional clinical coach, Nurse Elena Parker, and focuses on helping learners communicate critical points patients often misunderstand after surgery.
Audience: Nurses, PAs, and discharge coordinators.
Responsibilities: Instructional Designer, e-Learning Development.
Tools Used: ChatGPT,
Problem:
Many clinicians are confident with medical content but struggle to explain discharge instructions in plain, patient-centered language. This can lead to complications like infection, improper medication use, or missed follow-ups, all factors that increase readmission rates. Existing training rarely offers low-pressure, repeatable opportunities to practice this kind of communication.
Solution:
I designed a guided, AI-based simulation led by a fictional nurse educator, Nurse Elena Parker, who coaches learners through a realistic discharge scenario. The learner selects a case (e.g., appendectomy, C-section or knee surgery), types out how they would explain each discharge instruction, and receives simulated patient responses and real-time coaching from Elena.
Each simulation concludes with a copyable transcript and self-assessment form, allowing for reflection, supervisor review, and integration into clinical training programs.
ChatGPT screen showing conversation.
Process Overview:
I designed a branching experience where learners select from multiple surgical scenarios and then walk through a realistic discharge interaction. To fine-tune the tone and functionality, I tested early versions with both nurses and educators, gathering feedback on realism, coaching style, user functionality, and summary documentation.
Prompt logic was carefully refined to ensure the AI responds appropriately, nudges the learner forward when needed, and provides supportive yet actionable feedback. I also built a copyable summary reflection tool that includes space for self-assessment and optional supervisor review.
ChatGPT screen with prompt option.
Results & Takeaways:
The final product is a reusable, realistic simulation tool that builds discharge communication confidence in a safe environment. Feedback from testers highlighted the usefulness of receiving patient-like responses and comparing their own language to best-practice phrasing.
The project reinforced how much value structured coaching and simple reflection tools can add, especially when paired with low-stakes practice. Adding the option for transcript submission increased learner engagement and accountability.
Anticipated outcomes are that users would have increased confidence and consistent discharge instructions, while reducing readmission rates for patients.