Modeling Clinical Concern Trajectories in Language Model Agents

TL;DR

This paper proposes a new LLM agent architecture with explicit state dynamics that reveals pre-escalation signals in clinical settings, improving transparency and monitoring of rising concern.

Contribution

It introduces a lightweight architecture integrating a clinical risk encoder with first- and second-order dynamics to produce continuous concern trajectories.

Findings

Second-order dynamics produce smooth, anticipatory concern trajectories.

Stateless agents exhibit sharp escalation cliffs.

Trajectories reveal sustained unease prior to escalation.

Abstract

Large language model (LLM) agents deployed in clinical settings often exhibit abrupt, threshold-driven behavior, offering little visibility into accumulating risk prior to escalation. In real-world care, however, clinicians act on gradually rising concern rather than instantaneous triggers. We study whether explicit state dynamics can expose such pre-escalation signals without delegating clinical authority to the agent. We introduce a lightweight agent architecture in which a memoryless clinical risk encoder is integrated over time using first- and second-order dynamics to produce a continuous escalation pressure signal. Across synthetic ward scenarios, stateless agents exhibit sharp escalation cliffs, while second-order dynamics produce smooth, anticipatory concern trajectories despite similar escalation timing. These trajectories surface sustained unease prior to escalation, enabling human-in-the-loop monitoring and more informed intervention. Our results suggest that explicit state dynamics can make LLM agents more clinically legible by revealing how long concern has been rising, not just when thresholds are crossed.