Large Language Model–Based Simulated Patient Training for Heart Failure Palliative Care Communication: A Pilot Study

Aims Heart failure palliative care communication is essential but difficult to train at scale because conventional role-play programs require facilitators and standardized patients. Large language models have emerged as potential tools for scalable communication training. This pilot study aimed to evaluate the feasibility of a web-based large language model-driven communication training application and to explore its early educational signal on physicians’ self-efficacy. Methods and results This single-arm pilot study included physicians who completed one session using a Japanese-language web-based large language model application designed to simulate patients with advanced heart failure and provide automated framework-based feedback. The primary outcome was change in self-efficacy scores assessed by pre- and post-session questionnaires. Ten sessions were analyzed. Physicians engaged in a mean of 7.6 ± 2.0 dialogue turns. Mean response time per model output and feedback generation were approximately 3 and 17 seconds, respectively. Significant improvements were observed in knowledge of palliative care communication (mean difference +1.7, adjusted P < 0.01) and confidence in heart failure palliative care communication (+1.2, adjusted P = 0.03). Other domains showed non-significant changes. Conclusion This pilot study demonstrated the feasibility of a web-based large language model simulated patient system and suggested an early educational signal in physicians’ self-efficacy for heart failure palliative care communication. Our scalable large language model-driven communication training may complement traditional educational approaches with further evaluation in larger controlled studies.