An Evidence-Based Framework for Patient-Facing Artificial Intelligence Integration in the Emergency Department

Area of Science:

• Emergency Medicine
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Emergency departments (EDs) are critical healthcare safety nets and digital entry points for acute care.
• Patient-facing artificial intelligence (PF-AI) tools can directly engage patients for information gathering, communication, and decision support under clinical supervision.

Purpose of the Study:

• To present a framework for the effective and safe implementation of PF-AI in the emergency care journey.
• To identify areas where PF-AI can add value and where strict limitations are necessary.

Main Methods:

• Mapping operational touchpoints in the emergency care pathway from prearrival to postdischarge.
• Defining information-flow and accountability checkpoints.
• Proposing a human oversight model with defined escalation roles.

Main Results:

• PF-AI can enhance structured intake, patient education, multilingual communication, and postdischarge follow-up.
• Risks associated with PF-AI include undertriage, inaccurate content, workflow disruption, privacy violations, inequitable access, increased costs, and legal uncertainties.

Conclusions:

• A constrained framework is proposed to guide PF-AI implementation in EDs, emphasizing equity, interoperability, usability, privacy, safety, and auditability.
• PF-AI should augment, not replace, clinician judgment and existing workflows.