A Patient-Focused Comparative Analysis of Large Language Models for Epidural Analgesia Education

Epidural analgesia is a common labor pain management method, yet patients often require clear education on its risks and benefits. This study assessed 2 large language models (LLM), ChatGPT (OpenAI) and MediSearch (Create.ai), as potential tools for educating patients about epidural analgesia. Using 100 patient-focused questions categorized by the Rothwell System, responses were evaluated with DISCERN reliability scores and readability metrics (Flesch-Kincaid Grade Level, Coleman-Liau Index). MediSearch achieved higher reliability (P < .0001), while ChatGPT excelled in readability (P =.0013). Findings highlight an imbalance between reliability and readability between the 2 LLMs.