Aligning AI Payment Policy With Desired Outcomes Rather Than Inputs May Require Customized Pathways

Area of Science:

• Health economics
• Medical artificial intelligence
• Healthcare policy

Background:

• Artificial intelligence (AI) offers value in healthcare beyond traditional clinician inputs.
• Current Medicare reimbursement models, focused on human labor, may undervalue AI's contributions.
• This creates a misalignment between AI's potential value and its financial recognition.

Purpose of the Study:

• To examine the conflict between input-based pricing and outcome-based care in healthcare.
• To compare and contrast Medicare reimbursement for AI with prescription drug pricing.
• To propose a framework for aligning AI reimbursement with its actual value.

Main Methods:

• Comparative analysis of AI and prescription drug reimbursement models.
• Development of a classification system for AI technologies based on their impact on clinician time and cost.
• Policy analysis of current Medicare reimbursement structures.

Main Results:

• Medicare's input-based reimbursement may inaccurately value AI in clinical settings.
• A proposed classification system can differentiate AI types by their resource implications.
• Aligning reimbursement with outcomes, similar to some drug pricing strategies, is suggested.

Conclusions:

• Revising reimbursement policies to focus on outcomes rather than inputs is essential for AI adoption.
• A structured approach to AI reimbursement can foster innovation and benefit all stakeholders.
• Policy makers should consider outcome-based models to fully realize AI's healthcare potential.