ChatGPT-4o in Enhancing Informed Consent in Pediatric Surgical Practice

Informed consent (IC), beyond being a foundation of surgical practice, is a fundamental principle of high-quality healthcare. However, its proper implementation presents significant challenges within the medical community. The potential role of artificial intelligence (AI) in medicine has recently attracted considerable attention. The aim of this study was to assess the performance of ChatGPT-4o in generating personalized IC documents using a pre-designed prompt, evaluating the satisfaction of pediatric surgeons and patients/legal representatives with the AI-generated content.

We performed a cross-sectional, observational, single-center analysis to compare the quality of ICs provided by pediatric surgeons (Original-IC) with those generated by the GPT-4o model (AI-IC) for pediatric surgical patients from May to June 2024. The quality of both ICs was evaluated by patients/legal representatives and pediatric surgeons based on eight distinct parameters.

A total of 102 paired responses were collected to compare Original-IC and AI-IC. Legal representatives rated AI-IC higher across all quality parameters, with statistically significant differences (p < 0.05). Pediatric surgeons also scored AI-IC higher on most parameters, except for conciseness (p = 0.06). The preference for AI-generated ICs was 82.6 % among legal representatives and 87.5 % among pediatric surgeons, indicating a strong, statistically significant preference in both groups.

AI enhances the satisfaction of legal representatives and pediatric surgeons with the informed consent process, potentially improving its quality in pediatric care. Therefore, despite the challenges, studies assessing the benefits of AI platforms in medicine are becoming increasingly essential.