cs.HCMar 26, 2026

Rethinking Health Agents: From Siloed AI to Collaborative Decision Mediators

Authors: Ray-Yuan Chung, Xuhai Xu, Ari Pollack

Organizations: University of Washington Seattle, WA, USA · Columbia University New York, NY, USA · Seattle Children’s Hospital Seattle, WA, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Large language model based health agents are increasingly used by health consumers and clinicians to interpret health information and guide health decisions. However, most AI systems in healthcare operate in siloed configurations, supporting individual users rather than the multi-stakeholder relationships central to healthcare. Such use can fragment understanding and exacerbate misalignment among patients, caregivers, and clinicians. We reframe AI not as a standalone assistant, but as a collaborator embedded within multi-party care interactions. Through a clinically validated fictional pediatric chronic kidney disease case study, we show that breakdowns in adherence stem from fragmented situational awareness and misaligned goals, and that siloed use of general-purpose AI tools does little to address these collaboration gaps. We propose a conceptual framework for designing AI collaborators that surface contextual information, reconcile mental models, and scaffold shared understanding while preserving human decision authority.

Figures & tables

Figure 1. Proposed diagram comparing siloed AI use with an embedded AI collaborator in multi-stakeholder care. A side-by-side conceptual diagram of AI-agent collaboration. The left panel shows siloed use of AI chatbots by health consumers outside the clinic, highlighting issues such as over-reliance, distrust, and misaligned understanding between patients and clinicians. The right panel shows an AI collaborator positioned between health consumers and clinicians, with arrows indicating bidirectional communication and support for continuous care outside the clinic and improved collaboration during clinical visits.

Explore similar work

CardsList

  1. A Framework for Longitudinal Health AI Agents

    Apr 13, 2026Georgianna Lin, Rencong Jiang, Noémie Elhadad +1Long-Horizon Agent TasksHealthcare

    Columbia University, Biomedical Informatics, New York, New York USA · Columbia University, Computer Science, New York, New York USA

  2. Why Specialist Models Still Matter: A Heterogeneous Multi-Agent Paradigm for Medical Artificial Intelligence

    May 28, 2026Yanan Wang, Shuaicong Hu, Jian Liu +3Multi-Agent LLM SystemsHealthcare

    Department of Biomedical Engineering, College of Biomedical Engineering, Fudan University, Shanghai 200433, China · Department of Cardiology, Xinghua City People’s Hospital affiliated to Yangzhou University, Jiangsu 225700, China.

  3. MedCollab: IBIS-Guided Multi-Agent Collaboration with Hierarchical Disease Relation Chains for Clinical Diagnosis

    Mar 1, 2026Yuqi Zhan, Xinyue Wu, Tianyu Lin +6Multi-Agent LLM SystemsMedical Report Generation

    Hangzhou Dianzi University · Zhejiang University · Children’s Hospital, Zhejiang University School of Medicine, National Clinical Research Center for Children and Adolescents’ Health and Diseases