Multi-Agent Retrieval Augmented Generation for Clinical Decision Support: A Systematic Review and Integrative Conceptual Framework
DOI:
https://doi.org/10.30871/jaic.v10i1.11900Keywords:
Retrieval Augmented Generation(RAG),, Multi agent systems, Agentic AI Clinical decision support (CDSS), Transparency, SafetyAbstract
Multi agent retrieval augmented generation (RAG) systems are increasingly explored as advanced architectures for clinical decision support combining information retrieval, reasoning and verification through coordinated agent interactions. This study systematically reviews applications of agentic and multi agent RAG in clinical decision support systems (CDSS) and synthesizes an integrative conceptual framework linking technical design to technology adoption considerations. Following PRISMA guidelines, searches were conducted from PubMed, IEEE Xplore and ScienceDirect using structured Boolean strings combining terms for multi agent architectures, RAG and CDSS.The search yielded 12 studies published between 2020 and 2025 that met the inclusion criteria. The review synthesises evidence on multi agent role configurations retrieval and reasoning strategies, verification mechanisms and reported clinical contexts. Across studies, dominant challenges include data and corpus limitations retrieval quality dependency, limited clinical validation and computational overhead, alongside governance concerns such as privacy, bias and accountability. Building on the synthesis, we propose a four-agent CDSS framework retriever, reasoner, verifier, safety and map its deployment determinants to Technology Acceptance Model constructs perceived usefulness, perceived ease of use, trust and diffusion of Innovations attributes. The review concludes with design-oriented recommendations for safer, explainable, and adoption-ready multi-agent RAG CDSS, particularly for low-resource contexts.
Downloads
References
[1] E. J. Topol, “High-performance medicine: the convergence of human and artificial intelligence,” Nat. Med. 2019 251, vol. 25, no. 1, pp. 44–56, Jan. 2019, doi: 10.1038/s41591-018-0300-7.
[2] R. T. Sutton, D. Pincock, D. C. Baumgart, D. C. Sadowski, R. N. Fedorak, and K. I. Kroeker, “An overview of clinical decision support systems: benefits, risks, and strategies for success,” Dec. 01, 2020, Nature Research. doi: 10.1038/s41746-020-0221-y.
[3] M. Mangundu, L. Roets, and E. J. van Rensberg, “Accessibility of healthcare in rural Zimbabwe: The perspective of nurses and healthcare users,” African J. Prim. Heal. Care Fam. Med., vol. 12, no. 1, pp. 1–7, 2020, doi: 10.4102/PHCFM.V12I1.2245;CTYPE:STRING:JOURNAL.
[4] Y. Kinfu, M. R. Dal Poz, H. Mercer, and D. B. Evans, “The health worker shortage in Africa: Are enough physicians and nurses being trained?,” Bull. World Health Organ., vol. 87, no. 3, pp. 225–230, 2009, doi: 10.2471/BLT.08.051599.
[5] A. B. Mbakwe, I. Lourentzou, L. Anthony, C. Id, J. Mechanic, and A. Dagan, “PLOS DIGITAL HEALTH ChatGPT passing USMLE shines a spotlight on the flaws of medical education,” pp. 9–11, 2023, doi: 10.1371/journal.pdig.0000205.
[6] P. Lewis et al., “Retrieval-augmented generation for knowledge-intensive NLP tasks,” Adv. Neural Inf. Process. Syst., vol. 2020-Decem, 2020.
[7] B. B. Ozmen et al., “Development of a novel artificial intelligence clinical decision support tool for hand surgery: HandRAG.,” J. Hand Microsurg., vol. 17, no. 4, p. 100293, Jul. 2025, doi: 10.1016/j.jham.2025.100293.
[8] X. Zhao, S. Liu, S. Y. Yang, and C. Miao, MedRAG: Enhancing Retrieval-augmented Generation with Knowledge Graph-Elicited Reasoning for Healthcare Copilot, vol. 1, no. 1. arXiv, 2025. doi: 10.1145/3696410.3714782.
[9] “How Multi-Agent RAG Systems Transform Healthcare Decision Support - Techkraft Inc.” Accessed: Sep. 03, 2025. [Online]. Available: https://techkraftinc.com/how-multi-agent-rag-systems-transform-healthcare/
[10] G. de A. e Aquino et al., “From RAG to Multi-Agent Systems: A Survey of Modern Approaches in LLM Development,” Feb. 2025, doi: 10.20944/PREPRINTS202502.0406.V1.
[11] L. Deng, H. Hu, K. Lu, and P. He, “LLM-augmented multi-agent cooperative framework for medical case retrieval in cardiology,” vol. 123, 2025.
[12] C. Chang et al., “MAIN-RAG : Multi-Agent Filtering Retrieval-Augmented Generation”.
[13] T. Nguyen, P. Chin, and Y.-W. Tai, “MA-RAG: Multi-Agent Retrieval-Augmented Generation via Collaborative Chain-of-Thought Reasoning,” vol. 1, pp. 1–27, 2025, [Online]. Available: http://arxiv.org/abs/2505.20096
[14] Y. Miao, Y. Zhao, Y. Luo, H. Wang, and Y. Wu, “Improving Large Language Model Applications in the Medical and Nursing Domains With Retrieval-Augmented Generation: Scoping Review,” J. Med. Internet Res., vol. 27, 2025, doi: https://doi.org/10.2196/80557.
[15] D. Moher et al., “Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement,” PLoS Med., vol. 6, no. 7, 2009, doi: 10.1371/journal.pmed.1000097.
[16] C. Zelin, W. K. Chung, M. Jeanne, G. Zhang, and C. Weng, “Rare disease diagnosis using knowledge guided retrieval augmentation for ChatGPT.,” J. Biomed. Inform., vol. 157, p. 104702, Sep. 2024, doi: 10.1016/j.jbi.2024.104702.
[17] D. Baur, J. Ansorg, C.-E. Heyde, and A. Voelker, “Development and Evaluation of a Retrieval-Augmented Generation Chatbot for Orthopedic and Trauma Surgery Patient Education: Mixed-Methods Study.,” JMIR AI, vol. 4, p. e75262, Oct. 2025, doi: 10.2196/75262.
[18] S. Liu, A. P. Wright, A. B. McCoy, S. S. Huang, B. Steitz, and A. Wright, “Detecting emergencies in patient portal messages using large language models and knowledge graph-based retrieval-augmented generation.,” J. Am. Med. Inform. Assoc., vol. 32, no. 6, pp. 1032–1039, Jun. 2025, doi: 10.1093/jamia/ocaf059.
[19] K. Gilani, M. Verket, C. Peters, M. Dumontier, H.-P. B.-L. Rocca, and V. Urovi, “CDE-Mapper: Using retrieval-augmented language models for linking clinical data elements to controlled vocabularies.,” Comput. Biol. Med., vol. 196, no. Pt B, p. 110745, Sep. 2025, doi: 10.1016/j.compbiomed.2025.110745.
[20] T. Abdullahi, L. Mercurio, R. Singh, and C. Eickhoff, “Retrieval-Based Diagnostic Decision Support: Mixed Methods Study,” JMIR Med. Informatics, vol. 12, p. e50209, Jun. 2024, doi: 10.2196/50209.
[21] I. Emssaad, F.-E. Ben-Bouazza, I. Tafala, M. C. El Mezali, and B. Jioudi, “Leveraging multilingual RAG for breast cancer RCPs: AI-driven speech transcription and compliance in Darija-French clinical discussions,” Comput. Methods Programs Biomed. Updat., vol. 8, p. 100221, 2025, doi: https://doi.org/10.1016/j.cmpbup.2025.100221.
[22] A. Madrid-García, D. Benavent, and B. Merino-Barbancho, “From chat to act: large language model agents and agentic AI as the next frontier of AI in rheumatology,” EULAR Rheumatol. Open, vol. 1, no. 3, pp. 147–156, 2025, doi: https://doi.org/10.1016/j.ero.2025.06.012.
[23] E. Tzanis et al., “Agentic systems in radiology: Principles, opportunities, privacy risks, regulation, and sustainability concerns,” Diagn. Interv. Imaging, 2025, doi: https://doi.org/10.1016/j.diii.2025.10.002.
[24] L. Wang et al., “A multimodal LLM-agent framework for personalized clinical decision-making in hepatocellular carcinoma,” Patterns, p. 101364, 2025, doi: https://doi.org/10.1016/j.patter.2025.101364.
[25] B. Clay, H. I. Bergman, S. Salim, G. Pergola, J. Shalhoub, and A. H. Davies, “Natural language processing techniques applied to the electronic health record in clinical research and practice - an introduction to methodologies,” Comput. Biol. Med., vol. 188, p. 109808, 2025, doi: https://doi.org/10.1016/j.compbiomed.2025.109808.
[26] M. Wang, Y. Shen, B. Zhao, X. Zhou, L. Sun, and X. Liu, “Enhancing LLM-based clinical reasoning in anesthesiology via graph-augmented retrieval and explainable generation.,” Heal. Inf. Sci. Syst., vol. 13, no. 1, p. 62, Dec. 2025, doi: 10.1007/s13755-025-00379-x.
[27] S. A. Gebreab, K. Salah, R. Jayaraman, M. H. ur Rehman, and S. Ellaham, “LLM-Based Framework for Administrative Task Automation in Healthcare,” in 2024 12th International Symposium on Digital Forensics and Security (ISDFS), 2024, pp. 1–7. doi: 10.1109/ISDFS60797.2024.10527275.
[28] T.-C. Ureche, L. Boicescu, M. Raducanu, E.-C. Popovici, S. Halunga, and D. N. Vizireanu, “The Convergence of Emerging Technologies and AI-Driven Autonomous Cybersecurity in Smart Digital Ecosystems,” in 2025 IEEE 2nd International Conference on Blockchain, Smart Healthcare and Emerging Technologies (SmartBlock4Health), 2025, pp. 1–6. doi: 10.1109/SmartBlock4Health64843.2025.11189602.
[29] J. C. L. Ong et al., “Large language model as clinical decision support system augments medication safety in 16 clinical specialties.,” Cell reports. Med., vol. 6, no. 10, p. 102323, Oct. 2025, doi: 10.1016/j.xcrm.2025.102323.
[30] O. K. Gargari and G. Habibi, “Enhancing medical AI with retrieval-augmented generation: A mini narrative review.,” Digit. Heal., vol. 11, p. 20552076251337176, 2025, doi: 10.1177/20552076251337177.
[31] A. Mohammad, A. Bulbanat, and F. Aljassar, “Comparative Evaluation of Diagnostic and Management Capabilities of Infiniti AI and ChatGPT-4o in Corneal Diseases.,” Cureus, vol. 17, no. 10, p. e95163, Oct. 2025, doi: 10.7759/cureus.95163.
[32] F. Gaber et al., “Evaluating large language model workflows in clinical decision support for triage and referral and diagnosis.,” NPJ Digit. Med., vol. 8, no. 1, p. 263, May 2025, doi: 10.1038/s41746-025-01684-1.
[33] G. Shan et al., “Comparing Diagnostic Accuracy of Clinical Professionals and Large Language Models: Systematic Review and Meta-Analysis,” JMIR Med. Informatics, vol. 13, 2025, doi: https://doi.org/10.2196/64963.
[34] K. Balakrishnan et al., “Artificial Intelligence in Rural Healthcare Delivery: Bridging Gaps and Enhancing Equity through Innovation”.
[35] S. Varshney, B. Jain, P. Singh, D. Rani, and S. Mehra, “Med-KGMA: A novel AI-driven medical support system leveraging knowledge graphs and medical advisors,” Comput. Biol. Med., vol. 197, p. 110929, 2025, doi: https://doi.org/10.1016/j.compbiomed.2025.110929.
[36] Y. Habchi et al., “Advanced deep learning and large language models: Comprehensive insights for cancer detection,” Image Vis. Comput., vol. 157, p. 105495, 2025, doi: https://doi.org/10.1016/j.imavis.2025.105495.
[37] Z. Yao and H. Yu, “A Survey on LLM-based Multi-Agent AI Hospital,” 2025, [Online]. Available: https://osf.io/bv5sg_v1
[38] Z. Wang, H. Li, D. Huang, H.-S. Kim, C.-W. Shin, and A. M. Rahmani, “HealthQ: Unveiling questioning capabilities of LLM chains in healthcare conversations,” Smart Heal., vol. 36, p. 100570, 2025, doi: https://doi.org/10.1016/j.smhl.2025.100570.
[39] L. A. Biesheuvel et al., “Large language models in critical care,” J. Intensive Med., vol. 5, no. 2, pp. 113–118, 2025, doi: https://doi.org/10.1016/j.jointm.2024.12.001.
[40] G. Tippani, S. Durusoju, L. Mergoju, and P. V. Reddy, “MedicAI: AI-Powered System for Enhanced Medical Information Access,” in 2025 9th International Conference on Inventive Systems and Control (ICISC), IEEE, Aug. 2025, pp. 804–809. doi: 10.1109/ICISC65841.2025.11188019.
[41] T. Çelikten and A. Onan, “Medcongtm: Interpretable multi-label clinical code prediction with dual-view graph contrastive topic modeling,” Knowledge-Based Syst., vol. 327, p. 114103, 2025, doi: https://doi.org/10.1016/j.knosys.2025.114103.
[42] Y. Hu, C. Xu, B. Lin, W. Yang, and Y. Y. Tang, “Medical multimodal large language models: A systematic review,” Intell. Oncol., vol. 1, no. 4, pp. 308–325, 2025, doi: https://doi.org/10.1016/j.intonc.2025.09.005.
[43] Y. Du, S. Li, A. Torralba, J. B. Tenenbaum, and I. Mordatch, “Improving Factuality and Reasoning in Language Models through Multiagent Debate,” Proc. Mach. Learn. Res., vol. 235, pp. 11733–11763, 2024.
[44] Chakraborty et. al, “Medical Application Using Multi Agent System - A Literature Survey Sougata Chakraborty *, Shibakali Gupta **,” Int. J. Eng. Res. Appl., vol. 4, no. 2, pp. 528–546, 2014.
[45] S. M. Palomares et al., “The Impact of Artificial Intelligence Technologies on Nutritional Care in Patients With Chronic Kidney Disease: A Systematic Review,” J. Ren. Nutr., 2025, doi: https://doi.org/10.1053/j.jrn.2025.06.002.
[46] M. Mangundu, L. Roets, and E. J. van Rensberg, “Accessibility of healthcare in rural Zimbabwe: The perspective of nurses and healthcare users,” African J. Prim. Heal. Care Fam. Med., vol. 12, no. 1, p. 2245, 2020, doi: 10.4102/PHCFM.V12I1.2245.
[47] M. Alves, J. Seringa, T. Silvestre, and T. Magalhães, “Use of Artificial Intelligence tools in supporting decision-making in hospital management,” BMC Health Serv. Res., vol. 24, no. 1, 2024, doi: 10.1186/s12913-024-11602-y.
[48] S. Sun, Z. Zhong, N. Yu, X. Gong, and K. Yang, “HumanMoD: A multi-RAG collaborative LLM for inclusive urban public healthcare services,” Appl. Soft Comput., vol. 184, p. 113684, 2025, doi: https://doi.org/10.1016/j.asoc.2025.113684.
[49] V. Zuhair et al., “Exploring the Impact of Artificial Intelligence on Global Health and Enhancing Healthcare in Developing Nations,” J. Prim. Care Community Heal., vol. 15, 2024, doi: 10.1177/21501319241245847.
[50] K. He et al., “A survey of large language models for healthcare: from data, technology, and applications to accountability and ethics,” Inf. Fusion, vol. 118, p. 102963, 2025, doi: https://doi.org/10.1016/j.inffus.2025.102963.
[51] M. Tarczyńska-łuniewska, “ScienceDirect ScienceDirect The procedure of building stable fundamental database with the use The procedure of building stable fundamental database with the use of Matlab software of Matlab software,” Procedia Comput. Sci., vol. 126, pp. 2163–2172, 2018, doi: 10.1016/j.procs.2018.07.232.
[52] C. Lin and C.-F. Kuo, “Roles and potential of Large language models in healthcare: A comprehensive review,” Biomed. J., vol. 48, no. 5, p. 100868, 2025, doi: https://doi.org/10.1016/j.bj.2025.100868.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Tarisai Mugambiwa , Belinda Ndlovu
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
