Accepted for/Published in: JMIR Formative Research
Date Submitted: Jan 16, 2026
Date Accepted: Jul 13, 2026
LLM Simplification of Open-Access Pediatric Strabismus Literature: Cross-Sectional Validation of Readability and Clinical Fidelity
ABSTRACT
Background:
Low health literacy limits caregiver and adolescent access to evidence-based information about pediatric strabismus, a time-sensitive condition requiring informed decisions. Peer-reviewed literature consistently exceeds recommended readability thresholds.
Objective:
To assess whether a large language model (LLM) can generate plain-language summaries of pediatric strabismus articles without compromising clinical fidelity.
Methods:
Design, Setting, and Participants: Cross-sectional analysis of 85 open-access, peer-reviewed pediatric strabismus articles (2022–2024), stratified by strabismus type (esotropia, exotropia, other), surgical relevance (surgery-related vs. non–surgery-related), and article type (case reports, reviews, original research). Intervention: Full texts were simplified using DeepSeek-V3.2 with a prompt mandating ≤7th-grade reading level, ≤800 words, and factual preservation. Main Outcomes and Measures: Readability assessed via Flesch–Kincaid Grade Level (FKGL) and Simple Measure of Gobbledygook(SMOG); fidelity rated by two pediatric strabismus specialists (Good/Moderate/Poor).
Results:
Original articles had mean FKGL 15.79 ± 1.53 and SMOG 14.41 ± 1.09. After LLM processing, FKGL decreased to 7.84 ± 1.30 (P < .0001) and SMOG to 7.68 ± 0.94 (P < .0001). Subgroup FKGLs were highest for case reports (8.35 ± 0.89, P=.0026); no significant between-group differences post-simplification (P > .05). Fidelity was rated Good in 95.3% (81/85) of summaries—including 100% of case reports and original research—and Moderate in 4.7% (all reviews); none were Poor.
Conclusions:
DeepSeek-V3.2 reliably simplifies pediatric strabismus literature to near National Institutes of Health(NIH) targets (≤8th grade) with high clinical fidelity. Slightly elevated FKGLs in case reports reflect inherent linguistic complexity—yet remain within acceptable limits. With clinician review, LLM outputs may support equitable, scalable plain-language communication in clinical practice Clinical Trial: none
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.