Currently submitted to: JMIR Formative Research
Date Submitted: Aug 29, 2026
Open Peer Review Period: Aug 30, 2026 - Oct 25, 2026
(currently open for review)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Accessibility of Open-Angle Glaucoma Patient Education Materials Written and Simplified by AI Chatbots
ABSTRACT
Background:
Through handouts, physicians can offer patient information that lasts outside of appointments. However, patient education materials (PEMs) must be accessible to a patient audience, including being written at a suitable reading grade level. Organizations like the American Academy of Ophthalmology (AAO) have improved accessibility of high-quality online resources, particularly within the field of glaucoma. However, topics are limited and readability of PEMs varies depending on disease process.
Objective:
Readability of glaucoma patient education materials (PEMs) generated by ChatGPT, Google Gemini, and OpenEvidence was compared to American Academy of Ophthalmology (AAO) and UpToDate resources to determine whether these large language models (LLMs) adjust reading levels to a fifth-grade standard.
Methods:
Three freely available LLMs—ChatGPT (GPT-5.1), Google Gemini (Flash 2.5), and OpenEvidence (2.0)—were used to generate glaucoma PEMs. Readability was scored before and after a standardized language simplification prompt using Flesch Reading Ease (FRE) and Flesch-Kincaid Grade Level (FKGL). Readability was compared across LLMs and against AAO and UpToDate patient materials using paired t-tests and ANOVA with Bonferroni correction.
Results:
LLM-generated PEMs exceeded a fifth-grade level as measured by FKGL (all p ≤ 0.03). Following prompting for language simplification, the LLMs significantly reduced reading grade level (all p < 0.05) and achieved FKGL near the fifth-grade (all p < 0.05), with Google Gemini producing the highest FRE and lowest FKGL. Simplified LLM-generated PEMs were significantly more readable than AAO materials (all p ≤ 0.02) and comparable to UpToDate patient education articles, which were already written around a sixth-grade reading level.
Conclusions:
LLM-based AI tools can be prompted to generate glaucoma PEMs with readability approaching recommended fifth-grade levels. While unprompted outputs remain too complex, targeted language simplification substantially improves accessibility.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.