Enhancing patient participation in emergency department through patient-friendly clinical notes generated by large language models

Patient-centered care (PCC) emphasizes providing patients with clear information to support active participation in medical decision-making. However, the fast-paced nature of emergency departments (ED), coupled with communication barriers and varying health literacy, limits effective patient engagement. While large language models (LLMs) have shown potential in generating patient-friendly documents, their use in ED settings remains underexplored. This study aimed to develop LLM-generated patient-friendly clinical notes (PFCNs) that transform clinical notes into plain language, and to evaluate whether PFCNs could enhance patient participation in ED consultations. In this study, a total of 120 PFCNs were generated and evaluated, receiving high understandability ratings from both 10 clinicians and 20 patients (PEMAT score: 87.2%). Patients who used PFCNs during simulated ED consultations reported significantly higher participation levels compared to prior ED experiences (PPQ, P < 0.05). Qualitative data showed that PFCNs supported understanding, question preparation, emotional reassurance and improved relationships with clinicians, though concerns about hallucinations and integration into clinical workflows remained. These findings suggested that PFCNs generated by LLMs show promise for enhancing patient participation in ED consultations. Future work should address accuracy and explore real-world integration to support safe and effective deployment.