Designing clinical AI for patient-centered support beyond the visit: the PACT framework for health systems

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Clinical Operations

Background:

• Clinical AI excels at in-visit tasks like prediction and documentation.
• Care failures often stem from breakdowns in post-visit follow-up, handoffs, and communication.
• Current AI solutions inadequately address operational continuity across the patient journey.

Purpose of the Study:

• To propose a framework for advancing clinical AI beyond task performance to operational follow-through.
• To introduce the PACT framework (Patient-centered, AI-enabled Continuity and Timely action) for supporting care continuity.
• To outline the operational elements necessary for accountable AI-driven healthcare actions.

Main Methods:

• Conceptual framework development (PACT).
• Specification of operational elements: ownership, communication, confirmation, escalation, and outcomes.
• Illustration of practical application in post-visit coordination workflows.

Main Results:

• The PACT framework reframes clinical AI as a health system function supporting continuity, completion, escalation, and equity.
• Key operational elements for accountable AI action are defined.
• Post-visit coordination serves as a practical test case for PACT integration.

Conclusions:

• The next frontier for clinical AI is ensuring operational follow-through across the care journey.
• The PACT framework provides a structured approach to implementing AI for enhanced care continuity and equity.
• Health systems can leverage PACT to improve workflow integration, monitoring, and support for AI-driven care processes.