Enhancing Difficult Conversation Skills Using an AI Chatbot

Palliative care delivery can be challenging due to the complexity and diversity patient needs and confidence of interprofessional clinicians in having difficult conversations. Standardized patient (SP) simulation enhances realism and promotes learner confidence but is a costly teaching/learning tool.Goals, objectives, and purpose: The purpose of this project was to pilot use of an AI chatbot as an SP replacement during an end-of-life simulation to enhance competence and self-efficacy among interprofessional pre-licensure students. Our aims included:

1. To assess feasibility of the AI chatbot and student feedback mechanisms using a pre-programmed scenario and evaluation rubric.

2. To compare student self-efficacy and perceptions between SP and AI chatbot simulations.

3. To validate the AI chatbot as an alternative SP approach through faculty focus group and student open-ended survey feedback.Methods/Methodology: Following IRB approval for this pilot study, we utilized a mixed methods approach to evaluate the use of an AI chatbot as an SP alternative for an end-of-life/hospice simulation in an online interprofessional palliative care course.Results/Findings: Paired t-tests and descriptive statistics were used for quantitative analysis, and thematic analysis was used to summarize qualitative data. Though students rated standardized patient simulations higher in realism, value, and preference compared to AI chatbot simulation, statistically significant (pConclusions, implications, and/or curiosities: Use of an AI chatbot, though not an effective replacement for standardized patient simulation, does demonstrate potential as a tiered learning approach to prepare students to engage with patients in difficult conversations. Interprofessional faculty training is needed to guide the development of future instructional strategies for the benefit of students across disciplinary programs.