Readability of AI-Generated Patient Information Leaflets on Alzheimer’s, Vascular Dementia, and Delirium

Topic: Risk, prevention and care in population studies

Background Large language models such as ChatGPT, DeepSeek, and Gemini are increasingly used to generate patient-facing medical content. While their factual accuracy has been explored, the readability of these outputs remains less well understood. Readability is a crucial component of health communication, particularly for older adults and those with lower health literacy. This study aimed to evaluate and compare the readability of patient information leaflets generated by three large language models - ChatGPT, DeepSeek, and Gemini - on the topics of Alzheimer's disease, vascular dementia, and delirium, using five validated readability metrics. Materials and methods We conducted a cross-sectional comparative study of patient information leaflets generated by three large language models on the topics of Alzheimer's disease, vascular dementia, and delirium. Each model was prompted using identical queries, and the resulting texts were evaluated using five established readability metrics: Flesch Reading Ease, Flesch-Kincaid Grade Level, Gunning Fog Index, Simple Measure of Gobbledygook (SMOG) Index, and Automated Readability Index. Readability scores were compared using Kruskal-Wallis tests to identify statistically significant differences between models. Results ChatGPT consistently produced the most readable content, with the highest Flesch Reading Ease scores and the lowest grade-level indices. DeepSeek generated text that was markedly more complex and less accessible. Gemini performed intermediately, sometimes matching ChatGPT in specific indices but not consistently across all metrics. The difference in Flesch Reading Ease scores between models was statistically significant (H = 7.20, p = 0.027). Other metrics showed trends that approached significance. Conclusions There are meaningful differences in the readability of patient information generated by different large language models. ChatGPT appears to produce content that is more suitable for patient understanding, particularly in the context of older adult care. These findings highlight the need for careful evaluation of readability when using generative AI in clinical communication. Future research should incorporate expert review of content accuracy and appropriateness alongside readability.

Key findings

• Readability scores were compared using Kruskal-Wallis tests to identify statistically significant differences between models.
• Results ChatGPT consistently produced the most readable content, with the highest Flesch Reading Ease scores and the lowest grade-level indices.
• The difference in Flesch Reading Ease scores between models was statistically significant (H = 7.20, p = 0.027).
• These findings highlight the need for careful evaluation of readability when using generative AI in clinical communication.

Why it matters

This study compared the readability of patient information leaflets on Alzheimer's disease, vascular dementia and delirium generated by three large language models, finding ChatGPT produced the most readable content. It concerns health communication quality and does not test any treatment, cure or prevention.

Study details

• Study designcross-sectional
• Evidence strengthLow

Read the original

• DOI: 10.7759/cureus.85463
• PubMed: 40621318
• Original source →
• Free full text: Cureus →

Full text: free to read — OpenAlex lists this copy as free to read: the link above (checked 6 Feb 1020).

@article{jido2025-40621318,
  author  = {Jido, Jamila Tukur and Al-Wizni, Ahmed and Aung, Su Nandar},
  title   = {{Readability of AI-Generated Patient Information Leaflets on Alzheimer’s, Vascular Dementia, and Delirium}},
  journal = {Cureus},
  year    = {2025},
  doi     = {10.7759/cureus.85463},
  pmid    = {40621318},
  url     = {https://doi.org/10.7759/cureus.85463}
}

Studies this one cites

Referenced by this study and also in the alzhub library, so you can read them here.

• The inter-relationship between delirium and dementia: the importance of delirium prevention Nature reviews. Neurology · 2022-08-26 · usefulness score 59/100 · Prevention and risk

Related studies in Lewy body dementia

• Vascular dementia Lancet (London, England) · 2015-10-01 · usefulness score 59/100
• The overlap between vascular disease and Alzheimer's disease--lessons from pathology BMC medicine · 2014-11-11 · usefulness score 56/100
• Effect of depression on risks of Alzheimer’s disease and vascular dementia: A real-world longitudinal study Psychiatry Research · 2025-09-02 · usefulness score 55/100
• Relative frequencies of Alzheimer disease, Lewy body, vascular and frontotemporal dementia, and hippocampal sclerosis in the State of Florida Brain Bank Alzheimer disease and associated disorders · 2002-10-01 · usefulness score 54/100
• Clinical Phenotype of Apoe Ɛ4 Carriers In Patients With Vascular Dementia: A Cross-Sectional Study From Indonesia F1000Res · 2026-09-02 · usefulness score 52/100
• Vascular dementia: World Stroke Organization fact sheet 2026 International Journal of Stroke · 2026-01-01 · usefulness score 51/100

Also tagged “classification”

835 studies in the library carry the topic tag “classification”; these are this study's neighbours in a fixed, alphabetical-by-hash tour of them, not the closest in content. All 835 tagged studies.

• Kinetic Modeling of Amyloid Binding in Humans using PET Imaging and Pittsburgh Compound-B Journal of Cerebral Blood Flow & Metabolism · 2005-06-09 · usefulness score 46/100 · Brain scans and imaging
• Longitudinal structural MRI-based deep learning and radiomics features for predicting Alzheimer’s disease progression Alzheimer s Research & Therapy · 2025-08-07 · usefulness score 55/100 · Mild cognitive impairment
• Comparison of 7.0-T T₂*-magnetic resonance imaging of cerebral bleeds in post-mortem brain sections of Alzheimer patients with their neuropathological correlates Cerebrovascular diseases (Basel, Switzerland) · 2011-03-21 · usefulness score 35/100 · Brain scans and imaging

Similar studies

• Artificial intelligence in drug development for delirium and Alzheimer’s disease 2025 · Digital health tools
• Artificial Intelligence and Personhood in Dementia 2025 · Digital health tools
• Challenges and opportunities in understanding dementia and delirium in the acute hospital 2017 · Clinical safety
• Review of Artificial Intelligence for Clinical Use in Alzheimer's Disease and Related Dementias 2025 · Getting a diagnosis
• Artificial intelligence in dementia care: challenges, controversies, and policy implications 2026 · Digital health tools
• Investigating the Utility of Explainable Artificial Intelligence for Neuroimaging‐Based Dementia Diagnosis and Prognosis 2026 · Digital health tools

More on Lewy body dementia

• A review that covers this ground A review of studies on vascular dementia and iron metabolism Science Progress · 2025-10-01 · usefulness score 32/100
• Newest in this subject Frequency-Resolved Connectivity Graphs for EEG-Based Classification of Alzheimer and Frontotemporal Dementia bioRxiv · 2026-10-22 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Dementia and PDE5 inhibitors: A retrospective cohort study of veterans Research Square · 2026-10-13 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Lewy body dementia →