Large Language Models' Simulation of Patient Comprehension of Discharge Summaries: Multimodel Evaluation Study.
5 affiliations
- 1Luddy School of Informatics, Computing, and Engineering, Indiana University Bloomington, 700 N Woodlawn Ave, Bloomington, IN, United States, 1 812-223-3981.
- 2School of Medicine, Yale University, New Haven, CT, United States.
- 3College of Computing and Data Science, Nanyang Technological University, Singapore, Singapore.
- 4Biostatistics, University of California, Davis, Davis, CA, United States.
- 5College of Computing and Informatics, University of North Carolina at Charlotte, Charlotte, NC, United States.
Abstract
Background
Large language models (LLMs) are increasingly used to generate and evaluate patient-facing medical materials. Through persona prompting, LLMs can be instructed to simulate responses from individuals with different demographic and experiential backgrounds, offering a scalable approach for approximating patient feedback. However, it remains unclear how closely these simulated persona-based responses reflect real patient perspectives, and whether this alignment is consistent across demographic subgroups.Objective
This study assessed whether contemporary LLMs can simulate diverse patient perspectives on hospital discharge summaries when prompted with personas defined by education, gender, and health care use. We further examined whether persona-based simulation performance was equitable across subgroups by comparing LLM-generated responses with responses from a general-population human cohort under a standardized evaluation framework.Methods
We evaluated 4 contemporary LLMs, GPT-4.1, GPT-5.2, Claude Sonnet 4.5, and Opus 4.6, against responses from 99 participants recruited through Prolific (Prolific Academic Ltd). Participants answered 10 questions for discharge summaries that varied in length and medical complexity. Questions included information-based items assessing factual comprehension and perception-based items assessing subjective understanding. Model-human alignment was measured using modal mean absolute error and expected absolute ordinal distance. We also conducted a chain-of-thought sensitivity analysis for cases with the greatest divergence between LLM-generated persona and matched human responses.Results
Across LLMs, persona responses aligned more closely with human responses on information-based items than on perception-based items. Education was the most consistent stratification axis: every model aligned more closely with the high-education subgroup than with the low-education subgroup on information items, suggesting that LLMs capture broad factual-comprehension differences associated with educational background. Perception items, in which subjective understanding and uncertainty dominate human responses, were simulated less reliably. Alignment also depended on discharge summary characteristics, with short but clinically complex summaries producing the largest divergence. Gender-related differences were smaller and inconsistent in direction, reflecting item-level modal flips rather than systematic distributional bias. Chain-of-thought improved alignment on some factual items but did not recover the full range of less common human responses. Across models, LLMs underselected "I don't know" and "Not provided" responses relative to humans, indicating a tendency to commit to determinate answers even when human participants expressed uncertainty.Conclusions
Contemporary LLMs can approximate common response patterns within demographic subgroups, particularly for factual comprehension and education-based personas. However, they are less reliable for capturing heterogeneous and perception-based patient responses. Equity performance varied substantially across models, making model selection an important design consideration. Persona-based LLM simulation should be used as an assistive screening tool for patient materials and should complement, not replace, human evaluation with subgroup validation.
Related
- AI-assisted identification of potential treatment injury from inpatient discharge summaries: an early-stage sociotechnical evaluation in a New Zealand district hospital
- Gold-standard-guided direct preference optimisation for ICD-10-AM and ACHI code assignment from discharge summaries.
- Structured extraction from Russian/Kazakh discharge summaries: three model families at the human-agreement ceiling
- Context presentation and native conversation structure in large language model responses to escalating suicide risk: a controlled multimodel evaluation
- A Turkish mobile application for single-word auditory comprehension in aphasia: development and preclinical evaluation
BibTeX
@article{ma2026large,
title = {{Large Language Models' Simulation of Patient Comprehension of Discharge Summaries: Multimodel Evaluation Study.}},
author = {Xinyao Ma and Rui Zhu and Zihao Wang and Jingwei Xiong and Qingyu Chen and L Jean Camp and Haixu Tang},
journal = {Journal of medical Internet research},
year = {2026},
month = oct,
doi = {10.2196/89003},
url = {https://europepmc.org/article/MED/42912964},
note = {CC BY},
}
RIS
TY - JOUR
TI - Large Language Models' Simulation of Patient Comprehension of Discharge Summaries: Multimodel Evaluation Study.
AU - Xinyao Ma
AU - Rui Zhu
AU - Zihao Wang
AU - Jingwei Xiong
AU - Qingyu Chen
AU - L Jean Camp
AU - Haixu Tang
PY - 2026
DA - 2026/10/22
JO - Journal of medical Internet research
DO - 10.2196/89003
UR - https://europepmc.org/article/MED/42912964
AB - Background Large language models (LLMs) are increasingly used to generate and evaluate patient-facing medical materials. Through persona prompting, LLMs can be instructed to simulate responses from individuals with different demographic and experiential backgrounds, offering a scalable approach for approximating patient feedback. However, it remains unclear how closely these simulated persona-based responses reflect real patient perspectives, and whether this alignment is consistent across demographic subgroups.Objective This study assessed whether contemporary LLMs can simulate diverse patient perspectives on hospital discharge summaries when prompted with personas defined by education, gender, and health care use. We further examined whether persona-based simulation performance was equitable across subgroups by comparing LLM-generated responses with responses from a general-population human cohort under a standardized evaluation framework.Methods We evaluated 4 contemporary LLMs, GPT-4.1, GPT-5.2, Claude Sonnet 4.5, and Opus 4.6, against responses from 99 participants recruited through Prolific (Prolific Academic Ltd). Participants answered 10 questions for discharge summaries that varied in length and medical complexity. Questions included information-based items assessing factual comprehension and perception-based items assessing subjective understanding. Model-human alignment was measured using modal mean absolute error and expected absolute ordinal distance. We also conducted a chain-of-thought sensitivity analysis for cases with the greatest divergence between LLM-generated persona and matched human responses.Results Across LLMs, persona responses aligned more closely with human responses on information-based items than on perception-based items. Education was the most consistent stratification axis: every model aligned more closely with the high-education subgroup than with the low-education subgroup on information items, suggesting that LLMs capture broad factual-comprehension differences associated with educational background. Perception items, in which subjective understanding and uncertainty dominate human responses, were simulated less reliably. Alignment also depended on discharge summary characteristics, with short but clinically complex summaries producing the largest divergence. Gender-related differences were smaller and inconsistent in direction, reflecting item-level modal flips rather than systematic distributional bias. Chain-of-thought improved alignment on some factual items but did not recover the full range of less common human responses. Across models, LLMs underselected "I don't know" and "Not provided" responses relative to humans, indicating a tendency to commit to determinate answers even when human participants expressed uncertainty.Conclusions Contemporary LLMs can approximate common response patterns within demographic subgroups, particularly for factual comprehension and education-based personas. However, they are less reliable for capturing heterogeneous and perception-based patient responses. Equity performance varied substantially across models, making model selection an important design consideration. Persona-based LLM simulation should be used as an assistive screening tool for patient materials and should complement, not replace, human evaluation with subgroup validation.
SN - 1439-4456
SN - 1438-8871
ER -