ChatGPT in Pediatric Dental Trauma: A Cross-Sectional Assessment of Response Reliability to Parental Queries

Aim and background Parents often turn to the Internet for answers regarding their child's health due to limited access to doctors on-demand and curiosity about medical conditions or treatments. With its human-like interaction and artificial intelligence (AI), ChatGPT (OpenAI, San Francisco, California) is a popular AI-supported tool that is expected to become commonly used. Traumatic dental injuries (TDIs) are common in children, yet parents are often unaware of appropriate aid. Since timely emergency care significantly affects treatment outcomes, many parents may seek advice from AI chatbots during such situations. This study aimed to assess the relevance, correctness, logic, clarity, and completeness of ChatGPT's responses to frequently asked parental questions following dental trauma scenarios. Methods This study was conducted at Goa Dental College and Hospital, Bambolim. A set of 15 questions from a parent-perspective depicting scenarios were drafted and entered into ChatGPT version 4.0. A panel of 29 experts evaluated the responses using a pre-validated questionnaire, rating each answer on a five-point scale across five parameters, namely, relevance, correctness, logic, clarity, and completeness, along with an open-ended question to determine what the experts would have added to the ChatGPT response. Results The mean score across all responses was 4.35, indicating that most ChatGPT answers were of good quality with minimum inaccuracies. Completeness had the lowest score among all parameters, with only 82.76% of respondents rating it ≥ 4 (score 4 = adequate), indicating that there were some missing components in the answers. Conclusion Answers were mostly found to be relevant, logical, correct, complete, and in clear language. Clinical significance ChatGPT can serve as a helpful tool for parents seeking timely, evidence-based information about dental trauma in children. This research is a foundational step for early evidence on the reliability of ChatGPT in the dental domain.