Multi-agent system improves drug safety for complex patient care
PRISMATIC : Prescription Risk Inspection System for Multi-Agent Treatment Interaction in Clinical Decision-Making
Computation and LanguageArtificial Intelligence
Summary
Medical errors happen when patients with multiple diseases get complicated treatments from different doctors who don't always coordinate well. The authors created PRISMATIC, a system where different specialized agents work together to make safer and clearer drug plans for these patients. The system learns which specialists to trust more over time and helps them fix disagreements in prescriptions for each patient. Tests on real hospital records showed PRISMATIC reduces drug conflicts and improves treatment quality compared to other methods.
What this means in practice
- •For hospital it teams: Integrate a multi-agent coordination platform to reduce prescribing errors and improve drug regimen safety for patients with multiple conditions.
- •For health software developers: Develop clinical decision support tools that use adaptive multi-agent models to enhance collaborative prescription management across specialties.
Authors
Ziyang Ling, Shuyu Lu, Ziyi Chen
Abstract
Medical prescribing errors for multimorbid patients, who require complex drug regimens, are a major patient safety concern. These errors often stem from fragmented decision-making across medical specialties, a critical coordination failure that leads to adverse drug events and rehospitalizations. To address this coordination challenge, we introduce PRISMATIC, a multi-agent architecture where specialized agents, each representing a workflow role, collaboratively generate and refine treatment plans. Operating on a blackboard architecture, the system first structures patient information from clinical notes, then iteratively develops safe, rational, and interpretable drug regimens. The agents' collaboration is governed by two novel adaptive coordination mechanisms: (1) Dynamic Self-Updating Weight Adjustment (DSWA), a long-term mechanism that adjusts the influence weights of agents based on historical case outcomes, enabling the system to adapt and trust more reliable specialists over time. (2) Discrepancy Feedback Precise Calibration (DFPC), a short-term mechanism where agents provide iterative feedback to resolve disagreements and eliminate prescription conflicts for the current patient. Experiments on patient records from the MIMIC-IV database show that PRISMATIC significantly reduces drug conflicts, improves prescription coverage, and enhances treatment consistency compared to strong LLM baselines. Our work demonstrates that structuring clinical co-treatment as a multi-agent coordination problem offers a promising path toward safer and more coherent multidisease management.