Abstract: TH-PO0969
Enhancing Health Literacy in CKD: Quality Improvement Initiative Comparing Artificial Intelligence (AI)-Generated Content and Expert Nephrologist Insights
Session Information
- Diversity and Equity in Kidney Health
November 06, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Diversity and Equity in Kidney Health
- 900 Diversity and Equity in Kidney Health
Authors
- Pena Zapata, Oscar Yasser, Quinnipiac University School of Health Sciences, North Haven, Connecticut, United States
- Singh, Shashwat, Quinnipiac University School of Health Sciences, North Haven, Connecticut, United States
- Umar, Sumayya, Quinnipiac University School of Health Sciences, North Haven, Connecticut, United States
- Shearier, Emily R., Hartford HealthCare Corporation, Hartford, Connecticut, United States
Background
Chronic Kidney Disease (CKD) affects more than 37 million Americans, yet only 10% are aware of their diagnosis. Low health literacy, particularly among underserved populations, is a major barrier to self-management. AI tools like ChatGPT offer a promising solution to simplify complex medical information, but their reliability and acceptance in clinical practice remain underexplored. This quality improvement (QI) project assesses whether ChatGPT-3.5 can effectively rewrite CKD educational content to improve readability while maintaining clinical accuracy, as judged by nephrologist.
Methods
Educational material from reputable CKD sources (e.g., National Kidney Foundation) was simplified by ChatGPT to a sixth-grade reading level, validated using the Flesch-Kincaid readability scale. Board-certified nephrologists anonymously reviewed both original and AI-generated versions of content spanning 12 CKD-related topics. Each version was rated using a 5-point Likert scale for accuracy, clarity, and applicability. Responses were analyzed using the Wilcoxon signed-rank test. We also examined whether years in practice influenced acceptance of AI-generated content.
Results
AI-generated content was rated “Accurate” or “Very Accurate” by 82% of nephrologists. Clarity scores were significantly higher in the ChatGPT version compared to the original (p < 0.01). A positive correlation was observed between years of experience and favorable ratings. Furthermore, 66% of participants indicated they would consider using ChatGPT as a patient education tool. Notably, experts highlighted AI’s potential for improving communication with patients from linguistically and educationally diverse backgrounds.
Conclusion
ChatGPT significantly enhances readability of CKD educational materials while maintaining strong perceived clinical accuracy. These findings support the integration of AI into nephrology education strategies, particularly for improving understanding among patients with low health literacy. This approach may reduce disparities in care and promote greater health equity in CKD management.