ChatGPT 4.0: accurate, clear, relevant, and readable responses to frequently asked fertility patient questions

Abstract

Objective: To evaluate the accuracy, clarity, relevance and readability of ChatGPT 4.0’s responses to frequently asked infertility questions. Materials and Methods: Cross-sectional study. Fifty frequently asked questions were collected from online patient forums. ChatGPT 4.0 generated responses using a standardized prompt (“as if ChatGPT were an infertility specialist, using the best available evidence”). Ten reproductive medicine specialists (5 senior, 5 junior) evaluated answers for accuracy, clarity, and relevance using Likert scales (1–5). Overall quality was assessed with the Global Quality Scale (GQS). Readability was determined with the Spanish Flesch-Kincaid index. Results: 94% of responses scored ≥3 in GQS (mean 3.6 ± 0.6). 62% were rated “very good” or “excellent”. Highest ratings were observed in emotional support (4.4–4.5), prognosis (4.2) and lifestyle (4.2–4.3). Infertility diagnosis showed lower accuracy (3.7). All specialists (100%) agreed ChatGPT could be a useful complementary tool for patients, when supervised, and 7 of 10 rated its overall performance as “very good” or “excellent”. Readability was 19.6 ± 4.2, equivalent to secondary school level. Conclusion: ChatGPT 4.0 provides clear and relevant responses, with acceptable accuracy, making it a potential complementary tool for initial infertility counseling. It does not replace medical consultation and requires professional oversight.