Privacy-, Linguistic-, and Information-Preserving Synthesis of Clinical Documentation through Generative Agents

Mark van Velzen^1,2Robert Frans van der Willigen^2,3,4*Vincent J. De Beer^2,4Helen I. de Graaf-Waar^1,2Ester R.C. De Graaf-Waar^5,6,7Esther van Leeuwen⁸Micha F. van der Willigen^2,3Martijn J. Van Der Willigen^2,3,9Gavin Renardus^2,3Rayan El Maaroufi^2,3Sven J Satimin^2,3Larissa M. Hartog^10,2Tim Hulsen^11,2Nico L.U. van Meetreren^12,13Mark C. Scheper^14,15,2,3

• ¹Erasmus MC Afdeling Anesthesiologie, Rotterdam, Netherlands
• ²Data Supported Healthcare: Data‑Science unit, Research Center Innovations in Care, Rotterdam University of Applied Sciences, Rotterdam, Netherlands
• ³School of Communication, Media and Information Technology, Rotterdam University of Applied Sciences, Rotterdam, Netherlands
• ⁴HR Datalab EAS, School of Engineering and Applied Science, Rotterdam University of Applied Sciences, Rotterdam, Netherlands
• ⁵Department of Orthopedic Surgery, VieCuri Medical Centre, Venlo, The Netherlands, Venlo, Netherlands
• ⁶Radboud Institute for Health Sciences, IQ Health, Radboud University Medical Center, Nijmegen, The Netherlands, Nijmegen, Netherlands
• ⁷School of Allied Health, HAN University of Applied Sciences, Nijmegen, Nijmegen, Netherlands
• ⁸Medifit Bewegingscentrum, Oss, Netherlands
• ⁹Hogeschool Rotterdam Instituut voor Engineering en Applied Science, Rotterdam, Netherlands
• ¹⁰Top Sector Life Sciences and Health (Health~Holland),, The Hague, Netherlands
• ¹¹Data Science & AI Engineering, Philips, Eindhoven, Netherlands
• ¹²Department of Anesthesiology, Erasmus Medical Center, Rotterdam, Netherlands
• ¹³Top Sector Life Sciences and Health (Health~Holland), The Hague, Netherlands
• ¹⁴Allied Health Professions, Faculty of Medicine and Acience, Macquarrie University, Sydney, Australia
• ¹⁵Solid Start Coalition, Erasmus Medical Center, Rotterdam, Netherlands

The widespread adoption of generative agents (GAs) is reshaping the healthcare landscape. Nonetheless, broad utilization is impeded by restricted access to high-quality, interoperable clinical documentation from electronic health records (EHRs) due to persistent legal, ethical, and technical barriers. Synthetic health data generation (SHDG), leveraging pre-trained large language models (LLMs) instantiated as GAs, could offer a practical solution by creating synthetic patient information that mimics genuine EHRs. The use of LLMs, however, is not without issues; significant concerns remain regarding privacy, potential bias propagation, the risk of generating inaccurate or misleading content, and the lack of transparency in how these models make decisions. We therefore propose a privacy-, linguistic-, and information-preserving SHDG protocol that employs multiple context-aware, role-specific GAs. Guided by targeted prompting and authentic EHRs—serving as structural and linguistic templates—role-specific GAs can, in principle, operate collaboratively through multi-turn interactions. We theorized that utilizing GAs in this fashion permits LLMs not only to produce synthetic EHRs that are accurate, consistent, and contextually appropriate, but also to expose the underlying decision-making process.To test this hypothesis, we developed a no-code GA-driven SHDG workflow as a proof of concept, which was implemented within a predefined, multi-layered data science infrastructure (DSI) stack—an integrated ensemble of software and hardware designed to support rapid prototyping and deployment. The DSI stack streamlines implementation for healthcare professionals, improving accessibility, usability, and cybersecurity. To deploy and validate GA-assisted workflows, we implemented a fully automated SHDG evaluation framework—co-developed with GenAI technology—which holistically compares the informational and linguistic features of synthetic, anonymized, and real EHRs at both the document and corpus levels. Our findings highlight that SHDG implemented through GAs offers a scalable, transparent, and reproducible methodology for unlocking the potential of clinical documentation to drive innovation, accelerate research, and advance the development of learning health systems. The source code, synthetic datasets, toolchains and prompts created for this study can be accessed at the GitHub repository: RESEARCH_SUPPORT/PROJECTS/Generative_Agent_based_Data-Synthesis at main · HR-DataLab-Healthcare/RESEARCH_SUPPORT