Ardent Health: An AI-Enabled Virtual Care Model, from Pilot to Production

Area of Science:

• Healthcare Management
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Healthcare leaders face persistent challenges including workforce instability, financial constraints, and rapid technological advancements.
• The need for innovative solutions to improve efficiency and patient care in dynamic healthcare environments is critical.

Purpose of the Study:

• To present a case study and playbook detailing the implementation of an AI-enabled virtual care model.
• To demonstrate how integrating virtual nursing, attending physicians, and remote patient monitoring can enhance healthcare delivery.
• To outline strategies for managing risk, scaling innovations, and measuring value in healthcare technology adoption.

Main Methods:

• Deployment of an AI-enabled virtual care model integrating virtual nursing, virtual attending physicians, virtual sitting, and remote patient monitoring (RPM).
• Utilization of artificial intelligence (AI) for analyzing video streams for fall risk detection, monitoring vital sign trends from wearables for early clinical deterioration identification, and supporting ambient documentation via speech recognition and natural language processing (NLP).
• Implementation across multiple markets with a focus on a case study at Ardent's East Texas location.

Main Results:

• At Ardent's East Texas location, virtual nursing led to reduced contract labor, decreased voluntary RN turnover, and improved salaries, wages, and benefits (SWB) per patient day, even with increased volume.
• Virtual attending physicians increased patient consultations, enhancing patient retention and generating bed-day capacity.
• AI-assisted vital signs monitoring showed correlation with lower mortality and shorter length of stay.
• The RPM program improved discharge continuity and reduced readmissions.

Conclusions:

• The AI-enabled virtual care model successfully integrated various virtual services and AI tools to provide earlier risk detection and workflow relief.
• The implementation demonstrated positive impacts on operational efficiency, staff retention, and patient outcomes, including reduced mortality and readmissions.
• Key readiness factors include updating consent protocols, establishing algorithm oversight, and investing in technological infrastructure.