Accessibility and readability of Spanish and Portuguese-translated online information about common eye conditions: A...

To evaluate the readability and accountability (reliability, quality, and credibility) of online patient education materials for common eye conditions in English, Portuguese, and Spanish.

Cross-sectional content analysis SUBJECTS: A total of 192 websites across seven eye diseases were analyzed in three languages: English (n = 63), Portuguese (n = 67), and Spanish (n = 62).

First-page Google search websites for macular degeneration, cataract, diabetic retinopathy, glaucoma, conjunctivitis, uveitis, and dry eye were assessed for readability using Flesch-Kincaid Grade Level (FKGL), Gunning Fog Index (GFI), and Flesch Reading Ease (FRE). Accountability was evaluated using JAMA and DISCERN benchmarks. Websites were also categorized by source type into: academic, national, private, or crowdsourced. Statistical analyses were done using MANOVA, ANOVA, and Tukey's HSD post-hoc tests.

Readability (FKGL, GFI, FRE) and accountability (JAMA, DISCERN) indices RESULTS: English-language websites had significantly more accessible readability scores (mean FKGL 8.1) than Spanish (16.6) and Portuguese (16.0) counterparts (p < 0.001). Portuguese websites had the lowest JAMA scores (1.2) and were predominantly authored by private entities. Spanish-language sites showed mixed results, outperforming English in some accountability metrics. Crowdsourced pages like Wikipedia had the highest accountability but poorest readability. Language was a significant predictor for readability and accountability in a multivariate analysis (p < 0.001).

Online ophthalmology patient education materials in Portuguese and Spanish demonstrate significantly poorer readability, and in some cases, lower accountability compared to English-language resources, posing a significant barrier for patients with limited English proficiency. Future efforts should focus on developing standardized, multilingual patient education materials that balance readability with content reliability, potentially incorporating tools such as large language models (LLM's), which have shown potential in improving accessibility of non-English health information.