Introduction

Purpose and Scope

This Responsible AI content for the Prior Authorization Criteria Matching use case aligns to the CHAI Responsible AI Guide (RAIG) by establishing a Testing and Evaluation (T&E) Framework: a consensus-defined set of methods, metrics, and/or benchmarks for developers and implementers to more concretely evaluate the responsible use of AI solutions for prior authorization criteria matching.

Teams developing, deploying, or monitoring an AI-enabled prior authorization criteria matching solution can use CHAI’s consensus-defined T&E Framework to guide evaluation. Additionally, organizations should review use case-specific T&E Frameworks for recommended CHAI-endorsed methods/metrics when completing the CHAI Applied Model Card.

Audience

This Responsible AI content is for developers and implementers of a Prior Authorization AI-Supported Criteria Matching solution:

  • Developers: individuals involved in the software development process, including requirements gathering, design, coding, testing, and maintenance of software applications (derived from IEEE, 12207:2017)

  • Implementers: individuals responsible for the procurement, deployment, and/or overall realization of a system or component in accordance with a specified design (derived from IEEE 829 and IEEE 730)

Use Case Description

The criteria matching component of an AI-supported prior authorization (PA) system automates the process of assessing whether a healthcare service, procedure, or medication meets the payer’s medical necessity guidelines. This step is critical in reducing administrative burdens on providers, improving turnaround times, and enhancing consistency in decision-making. AI facilitates the automatic extraction and alignment of clinical documentation with established coverage criteria, minimizing manual effort and reducing the potential for errors or variability in approvals. AI-supported solutions cannot be used to automatically deny prior authorization. AI is used as a tool to support, not replace, clinical decision making.

The AI-supported criteria matching component is limited to extracting and aligning documentation with coverage criteria. It does not determine approval or denial outcomes. Organizational routing rules (e.g., auto-approval pathways, triage logic, or manual review escalation) are separate, human-governed processes that may incorporate AI outputs but are not controlled by the AI model itself. All approval or denial decisions remain the responsibility of the implementing organization.

Primary End Users

  • Payers & Health Plans: Organizations ensuring AI-supported PA aligns with coverage policies

  • Utilization Review & Medical Management Teams: Clinicians and administrative staff making final determinations on flagged cases

  • Healthcare Providers: Physicians, nurses, and care teams submitting PA requests and reviewing AI recommendations