Abstract

Prior Authorization delivers safe, appropriate,and cost-effective care that is medically justified with evidence-based guidelines. However, the process often requires labor-intensivemanual comparisons between patient medicalrecords and clinical guidelines, that is bothrepetitive and time-consuming. Recent developments in Large Language Models (LLMs)have shown potential in addressing complexmedical NLP tasks with minimal supervision.This paper explores the application of MultiAgent System (MAS) that utilize specializedLLM agents to automate Prior Authorizationtask by breaking them down into simpler andmanageable sub-tasks. Our study systematically investigates the effects of various prompting strategies on these agents and benchmarksthe performance of different LLMs. We demonstrate that GPT-4 achieves an accuracy of86.2% in predicting checklist item-level judgments with evidence, and 95.6% in determining overall checklist judgment. Additionally,we explore how these agents can contributeto explainability of steps taken in the process,thereby enhancing trust and transparency in thesystem.

To read the full research paper, click on this link.