Utilizing AI Chatbots to Improve Clinical Interview and Examination
Utilization of an AI chatbot (AIC) to simulate pediatric physical therapy (PT) interviews and exams, practicing clinical reasoning, multi-informant communication, tailored communication, and engaged learning for students.
Presented by:
Lindsay Marquardt, University of Michigan-Flint
From the author, in their own words...
Lindsay Marquardt, University of Michigan-Flint
Transcript:
Hello, I’m Lindsay Marquardt from the University of Michigan–Flint Physical Therapy program. This project explored using AI chatbots to simulate a physical therapy evaluation with a pediatric patient and their caregivers. Students worked through the interview, interpreted clinical information and subjective history, and completed the necessary evaluation and plan of care documentation. Following the assignment, 16 students consented to complete a post assignment survey. Of those students, 68.8% reported improved engagement with the pediatric content and 75% reported improved information-gathering skills. In addition, students reported that utilizing the AI chatbot compared to traditional methods helped to both improve motivation for participation in the pediatric course as well as improve their history taking skills. These findings suggest AI-supported simulation may offer a low-risk, experiential way to practice pediatric clinical reasoning and communication.
Abstract:
An AIC–based assignment scaffolded a pediatric PT evaluation, documentation, and progression of care through a three-phase case study focused on interviewing/exam, data interpretation, and documentation. Although 64 students completed the assessment, survey responses were limited (n=15). Higher chatbot engagement correlated with increased course motivation (p=0.04, effect size=0.69). Perceived AIC effectiveness also correlated with preference for AIC over traditional methods (p=0.03, effect size=0.716). Overall, students who viewed AIC as beneficial reported greater motivation towards the class and stronger preference to use it—potentially supporting improved motivation in this niche area.
Key Words:
AI, Health Care, Interviewing
Outcomes:
1. Identify ways to use AI chatbots to increase student practice opportunities.
2. Conceptualize 1-2 specific assignments in their class that AI chatbots may be beneficial for student learning.
3. Identify methods to support a multi-informant AI chatbot.
References:
Belda-Medina, J., & Kokošková, V. (2023). Integrating chatbots in education: Insights from the Chatbot-Human Interaction Satisfaction Model (CHISM). International Journal of Educational Technology in Higher Education, 20, Article 62. https://doi.org/10.1186/s41239-023-00432-3
Or, A. J., Sukumar, S., Ritchie, H. E., & Sarrafpour, B. (2024). Using artificial intelligence chatbots to improve patient history taking in dental education: Pilot study. Journal of Dental Education, 88(Suppl. 3), 1988–1990. https://doi.org/10.1002/jdd.13591
Røed, R. K., Baugerud, G. A., Hassan, S. Z., Sabet, S. S., Salehi, P., Powell, M. B., Riegler, M. A., Halvorsen, P., & Johnson, M. S. (2023). Enhancing questioning skills through child avatar chatbot training with feedback. Frontiers in Psychology, 14, Article 1198235. https://doi.org/10.3389/fpsyg.2023.1198235