PRICE: a personalized recursive intelligent cost effectiveness analysis framework for rare disease diagnosis

Area of Science:

• Medical Informatics
• Health Economics
• Rare Disease Research

Background:

• Rare disease diagnosis is complex, lengthy, and expensive.
• Current cost-effectiveness analyses use static models for diverse patient groups.
• There's a need for dynamic, personalized diagnostic cost-effectiveness frameworks, especially with AI advancements.

Purpose of the Study:

• Introduce the PRICE (Personalized Recursive Integrated Cost-Effectiveness) analysis framework.
• Develop a dynamic model for evaluating diagnostic strategy cost-effectiveness at the individual patient level.
• Integrate artificial intelligence (AI) into diagnostic decision-making frameworks.

Main Methods:

• Utilized a novel, tree-based model (PRICE) for cost-effectiveness evaluation.
• Incorporated expert-alone and AI-delegated decision-making modes.
• Employed a back-propagation algorithm for expected cost computation and Quality Adjusted Life Years (QALYs) for effectiveness quantification.

Main Results:

• Demonstrated PRICE framework utility in a proof-of-concept study for developmental delay (DD) and multiple congenital anomalies (MCA).
• Showcased PRICE's ability to support personalized decisions by modeling outcomes under varying parameters (cost, prevalence, AI accuracy).
• Developed an interactive web-based tool for real-time PRICE pathway construction and cost-effectiveness analysis.

Conclusions:

• PRICE is a novel framework capturing sequential, recursive diagnostic workflows.
• The framework is extensible for future AI-integrated clinical practice.
• Enables personalized economic and clinical evaluations for improved rare disease diagnosis decision-making.