EUVIMEDEUVIMED entdecken
Scoping Review

Large Language Models for Generating Patient Education Materials: Scoping Review

Große Sprachmodelle zur Erstellung von Patient:innen-Informationsmaterialien: Scoping Review

Chengjin Li iD, Juan Li iD, Xuezhu Liu iD, Zefang Gao iD, Yuxiao Xu iD, Xiuze Xu iD, Xiaoyu Liu iD, Tong Zhao iD, Yuanyuan Cheng iD, Ming Li iD

Journal of Medical Internet Research · 06.11.2026 · 28 · e98084

DOI: 10.2196/98084 ↗

Auf einen Blick

30 Studien untersuchten von LLMs erzeugte neue Patient:innen-Informationsmaterialien. Die Materialien waren meist gut in Qualität und Verständlichkeit, erfüllten aber Gesundheitskompetenz-Standards oft nicht.

Fragestellung

Was ist die aktuelle Evidenz zu LLMs bei der Erstellung neuer Patient:innen-Informationsmaterialien hinsichtlich Lesbarkeit, Qualität und Nutzbarkeit?

Methode

Scoping Review nach PRISMA-ScR; Literaturrecherche in PubMed, Embase, Cochrane Library, Web of Science und Scopus vom 1. Januar 2000 bis 1. Januar 2026; englischsprachige Originalstudien.

Ergebnisse

30 Studien mit 11 LLMs und >30 Krankheiten. Nur 3 Datensätze erreichten ein Leselevel der 6. Klasse. Die meisten Materialien wurden als gut oder besser bewertet; bei der Nutzbarkeit war die Verständlichkeit meist gut, die Handlungsfähigkeit aber schwach.

Einschränkungen

Keine.

Schlussfolgerung der Autor:innen

LLMs bieten eine transformative Möglichkeit für effektive und hochwertige Patient:innen-Informationsmaterialien, erfüllen derzeit aber die Standards der Gesundheitskompetenz, besonders bei Lesbarkeit und Handlungsfähigkeit, noch nicht.

Kernaussagen

  • 30 Studien mit 11 LLMs und >30 Krankheiten
  • Nur 3 Datensätze erreichten Leselevel der 6. Klasse
  • Verständlichkeit meist gut, Handlungsfähigkeit meist schwach

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Scoping Review in der Zeitschrift Journal of Medical Internet Research, online erschienen am 06.11.2026. Verfasst von Chengjin Li und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Human Resource, First Affiliated Hospital of Soochow University, Soochow, Jiangsu, China. Hauptthemen laut MEDLINE: Large Language Models, Patient Education as Topic, Teaching Materials. Die Arbeit nennt 100 Literaturangaben.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Patient education materials (PEMs) are essential for improving health literacy and clinical outcomes. Although large language models (LLMs) hold promise for optimizing content, their performance in generating new PEMs remains unclear.

Objective

This study aimed to summarize the current evidence on LLMs in generating new PEMs, evaluate readability, quality, and usability in various health care settings, and guide future research and practice.

Methods

This review followed PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. A comprehensive literature search was conducted in PubMed, Embase, Cochrane Library, Web of Science, and Scopus from January 1, 2000, to January 1, 2026. Eligible studies were English-language original research studies that evaluated new PEMs generated by LLMs in health care using information assessment tools. Two reviewers independently screened records and extracted data. We conducted a quantitative analysis of the characteristics of the included literature. Furthermore, we classified the PEMs based on core educational tasks and clinical scenarios, and compared their readability, quality, and usability.

Results

Thirty studies were included, including 11 LLMs and more than 30 diseases; most compared multiple LLMs and used a single prompt. PEMs were classified into 5 categories: perioperative patient education (n=8), chronic disease management education (n=6), specialty disease diagnostic and therapeutic education (n=9), pharmacotherapy safety and adherence education (n=3), and public health and prevention education (n=4). In terms of readability, only 3 records achieved a sixth-grade reading level. Specialty disease diagnostic and therapeutic education performed relatively well, while chronic disease management education had poor readability. The DeepSeek series demonstrated the most consistent performance. In terms of quality, most records were rated as good or higher. Specialty disease diagnostic and therapeutic education showed a polarized distribution, with both high and low quality ratings. Pharmacotherapy safety and adherence education and public health and prevention education had no poor-quality ratings. The ChatGPT series performed relatively well. In terms of usability, most records performed well in understandability but poor in actionability. Perioperative patient education showed a relatively balanced performance.

Conclusions

LLMs offer a transformative opportunity for developing effective and high-quality PEMs, but they currently fail to meet health literacy standards, particularly in readability and actionability. Future efforts must shift from performance evaluation to implementation strategies. Only through rigorous validation and equitable deployment can LLMs truly improve patient education.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Large Language ModelsPatient Education as TopicTeaching MaterialsHumansComprehensionHealth Literacy

Stichwörter der Autor:innen

generative language modellarge language modelspatient educationqualityreadabilityscoping reviewusability

Autor:innen und Institutionen

  • Department of Human Resource, First Affiliated Hospital of Soochow University, Soochow, Jiangsu, China.
  • Shanghai Engineering Research Center of Intelligence Pediatrics, Shanghai Children's Medical Center, Shanghai Jiao Tong University, Shanghai, Shanghai, China.
  • Clinical Teaching Office, First Affiliated Hospital of Soochow University, Soochow, Jiangsu, China.
  • School of Public Health, Shanghai Jiao Tong University, Shanghai, Shanghai, China.
  • School of Political Science and Public Administration, Wuhan University, Wuhan, Hubei, China.
  • Department of Nephrology, First Affiliated Hospital of Soochow University, Soochow, Jiangsu, China.

ORCID-Profile: Chengjin Li, Juan Li, Xuezhu Liu, Zefang Gao, Yuxiao Xu, Xiuze Xu, Xiaoyu Liu, Tong Zhao, Yuanyuan Cheng, Ming Li

Zitieren

APA

Li, C., Li, J., Liu, X., Gao, Z., Xu, Y., Xu, X., Liu, X., Zhao, T., Cheng, Y., & Li, M. (2026). Large Language Models for Generating Patient Education Materials: Scoping Review. Journal of Medical Internet Research, 28, e98084. https://doi.org/10.2196/98084

Vancouver

Li C, Li J, Liu X, Gao Z, Xu Y, Xu X, et al. Large Language Models for Generating Patient Education Materials: Scoping Review. Journal of Medical Internet Research. 2026;28:e98084. doi:10.2196/98084

BibTeX
@article{Li2026,
  author = {Li, Chengjin and Li, Juan and Liu, Xuezhu and Gao, Zefang and Xu, Yuxiao and Xu, Xiuze and Liu, Xiaoyu and Zhao, Tong and Cheng, Yuanyuan and Li, Ming},
  title = {Large Language Models for Generating Patient Education Materials: Scoping Review},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e98084},
  doi = {10.2196/98084}
}
Kostenlos starten →