ChatGPT can generate patient education materials (PEMs) for pediatric eye surgeries, but the quality of its content is currently lower than that of materials from the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) website, particularly in terms of accuracy, up-to-date information, and tone, according to a study.
Researchers asked ChatGPT to generate sixth-grade-level PEMs in English and Spanish for strabismus surgery (with and without adjustable sutures), pediatric cataract surgery, and nasolacrimal duct probing.
The English-language materials from ChatGPT were assessed for quality using the Quality of Generated Language Outputs for Patients scale and compared to AAPOS materials. AAPOS PEMs scored higher in accuracy, currency, and tone. However, there was no significant difference in overall readability between the sources. The study also found no notable differences in quality or readability between ChatGPT’s English and Spanish outputs.
Researchers concluded that while current ChatGPT-generated materials are not yet on par with established professional sources, the technology is evolving and merits re-evaluation as it improves.
Reference
Yang A, Reid M, Nguyen AM, et al. Quality and Readability of Patient Educational Materials Generated by ChatGPT-4o for Pediatric Ophthalmologic Surgeries. J Pediatr Ophthalmol Strabismus. 2025;doi: 10.3928/01913913-20250404-01. Epub ahead of print. PMID: 40423504.
Related Content
