Clinical agent protocol improves accuracy by controlling evidence use
EGIP: An Evidence-Grounded Interaction Protocol for Clinical Agents
Computation and Language
Summary
Clinical agent programs help diagnose patients by talking to them and checking their records, but how they get information can affect their accuracy. The authors propose EGIP, a clear way to control what the agent sees and does during diagnosis, separating gathering clues from making a final call. They tested EGIP with different systems and doctors found it helped agents give more accurate, evidence-based answers. This method helps make sure agents' decisions rely on the right clinical information rather than quirks of how they interact.
What this means in practice
- •For clinical ai developers: Build clinical decision support systems that improve diagnosis accuracy by controlling how evidence is acquired and accessed during patient interactions.
- •For healthcare it teams: Integrate structured evidence protocols to ensure more reliable data flow between patient records and diagnostic agents.
Authors
Zecheng Ren, Ruichen Cong, Yi Min Chen, Jing Zhang, Feiyang Chen, Wutong Zhang, Kaixiang Wang, Zihan Wang, Jiong Lou, Jie LI, Qun Jin
Abstract
Interactive clinical agents acquire evidence through patient dialogue and diagnostic examinations, yet their performance also depends on how the interaction environment exposes information, available actions, and examination results. When these conditions are implicit or inconsistently controlled, final-answer accuracy can conflate diagnostic reasoning with the effects of the interaction protocol. We propose the \textbf{E}vidence-\textbf{G}rounded \textbf{I}nteraction \textbf{P}rotocol (EGIP), which enforces role-specific information boundaries, exposes diagnostic actions without revealing their results, retrieves examination results from stored clinical records, and organizes acquired evidence for final decision-making. By separating evidence acquisition from terminal prediction, EGIP enables different decision strategies to operate on the same observations. We evaluate EGIP across two clinical-agent frameworks, five datasets, and six language models. EGIP improves diagnostic accuracy over native baselines across all primary comparisons. A clinician review further shows that the proportion of automatically scored correct responses jointly judged evidence-supported rises from 10.20\% to 36.55\%. These findings support the effectiveness of EGIP as a unified interaction protocol for improving clinical-agent performance through controlled evidence access and acquisition.