Integrating a chatbot into simulation-based perfusion training: A pilot randomized controlled trial

Abstract

Purpose

Chatbots have enhanced student engagement and learning outcomes in education. Simulation-based training is critical in perfusion education, providing risk-free environments for skill mastery and clinical readiness. This research aimed to assess whether a chatbot (PerfusionPal) impacted students’ learning outcomes.

Methods

This pilot randomized, controlled trial, conducted in 2024, included 21 university students enrolled in the Cardiovascular Perfusion Master’s Program. Students were randomly assigned to the experimental group, receiving the PerfusionPal intervention, or the control group, without PerfusionPal. Two simulated scenarios were utilized. The primary outcome was pre-simulation and post-simulation assessment scores.

Results

In simulation 1, there were no significant differences between the control and experimental groups in pre- and post-simulation assessment scores (p = 0.539 and p = 0.276, respectively), completion time (p = 0.197), or the number of reservoir checks (p = 0.072). However, the experimental group had significantly longer average times between reservoir checks compared to the control group (p = 0.036). In simulation 2, no significant differences were found in pre- and post-simulation assessment scores (p = 0.832 and p = 0.605, respectively), completion time (p = 0.350), or number of reservoir checks (p = 0.061). The experimental group showed significantly higher average times between reservoir checks (p = 0.043).

Conclusion

This pilot study found that although the PerfusionPal chatbot did not significantly improve assessment scores or simulation performance, it influenced reservoir monitoring. More research is needed to optimize its integration into simulation training and to assess its feasibility and generalizability.