Educational intervention demonstrates improved counseling confidence in pharmacy students, indicating AI chatbots offer a scalable tool for communication practice.
OBJECTIVE: To (1) improve pharmacy students' perceived confidence in communication skills and (2) identify perceptions of counseling a text-based artificial intelligence (AI)-simulated patient using ChatGPT 4o. METHODS: This study involved first-year pharmacy students in a required communications course at the University of Georgia. Four AI-simulated patients with unique characteristics and medication-related problems were developed in ChatGPT 4o (OpenAI). After didactic instruction, students completed a pre-survey, practiced counseling with the text-based AI patients to prepare for 2 separate graded standardized patient counsling (GSPC) assessments and completed identical post-surveys after each GSPC to assess changes in confidence and overall perceptions. Quantitative data were analyzed using the Wilcoxon signed rank test, and thematic analysis was used for open-ended responses. RESULTS: Of 146 students, 117 (80.1%) completed all 3 surveys, with significant improvements in perceived confidence observed across all 7 communication domains. The greatest gains were noted in initiating counseling sessions, applying prime questions, and using probing techniques in the first post-survey. While significant in the first post-survey, no significant changes were noted in nonverbal communication skills (body language, active listening) in the second post-survey. Thematic analysis of 386 comments identified 5 benefit themes (eg, realistic practice, accessibility, structured counseling), 4 limitation themes (eg, lack of human-like interaction, response limits), and 4 improvement areas (eg, expanded scenarios, enhanced feedback). CONCLUSION: ChatGPT 4o text-based chatbot patients improved student perceived confidence in verbal counseling skills and were valued as a convenient, low-pressure, and accessible training tool. While limitations in realism and nonverbal skill development were noted, text-based AI-simulated patients represent a scalable supplement to traditional communication training in pharmacy education.
No takes yet. Share an insight, caveat, or question.
Thompson et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: