Becker's Hospital Review

26 Executive Briefing related to staffing, inpatient bed utilization, patient safety and other metrics. Beyond anticipating future events, the platform provides simple, personalized recommendations to staff and coordinates an immediate response, leading to the best pos- sible outcomes. AI lets hospitals extract greater meaning out of the millions to billions of dollars they've invested in the EMRs. Most important- ly, AI empowers clinicians to achieve their goals to improve the quality of care and enhance the overall patient experience. Beyond the Hype: How AI Helped Two Hospitals Improve Operations in the ED AI can be deployed to support a variety of initiatives and de- partments in the hospital, such as the emergency department, perioperative care, inpatient units and pharmacy. Many choose to start in the ED, the front door to the hospital. "Because of its high level of interdependencies, the ED is par- ticularly susceptible to capacity issues," says Stephen Traub, MD, chairman of the department of emergency medicine at Phoenix-based Mayo Clinic Arizona and advisor to Qventus. However, it's the setting where physicians are likelier to be pro- cess-oriented and success metrics are clear. In the following two case studies, AI enabled two St. Louis-based Mercy hospitals to better manage capacity and improve patient flow in the ED. Mercy Hospital-Fort Smith (Kan.) The ED at 336-bed Mercy Hospital-Fort Smith reported core performance metrics far below national benchmarks. It strug- gled to reduce ED crowding, coordinate processes and effi- ciently move patients through the department. As the primary access point for community residents seek- ing medical care, Mercy Hospital-Fort Smith's ED generated between 60 percent and 70 percent of inpatient admissions. High patient demand coupled with chronic inefficiencies in ED operations challenged the hospital's ability to treat and move patients through the system. As a result, patients experienced long wait times and frequently left without being seen. Hospital leaders determined many inefficiencies at Mercy Hos- pital-Fort Smith stemmed from the ED's reliance on labor-inten- sive tools to manage key processes and identify issues. Manu- ally identifying process outliers was time consuming and made proactive intervention nearly impossible. Nurses commonly discovered problems that occurred hours before and had al- ready delayed patient care before they had an opportunity to alert appropriate staff. How Mercy Hospital-Fort Smith optimized ED operations with AI Mercy Fort Smith implemented Qventus' "air traffic control" platform as a tool to help guide ED staff to best practices through a collection of "decision recipes." Each unique recipe is designed to monitor metrics, predict bottlenecks and recom- mend countermeasures in real time. It then sends personalized communications, called a "nudge," to frontline team members to enable them to take the right actions and collaborate on a response. This helped improve inconsistent workflow and tack- le the communication problems that led to inefficiencies in the ED. For example, the system was set to address ED surges. Contin- uously monitoring data, it would predict a surge two hours in advance of the situation, then send a specific recommendation to the charge nurse: "Congestion likely, prioritize discharges." The system identified the specific discharges that should be prioritized. This allowed the nurses to refocus resources and coordinate an immediate response. Over time, it also enabled a powerful culture of teamwork and empowerment. Five months after adopting the AI-based software, Mercy Hos- pital-Fort Smith saw significant operational improvements in its ED. The left-without-being-seen rate dropped 30 percent. The average length of stay for discharged patients fell to just 24 minutes, a 13 percent reduction, and its door-to-doctor time dropped by 15 minutes, a 20 percent reduction. This enabled the hospital to serve an additional 2,500 patients. Mercy Hospital-Ardmore (Okla.) The 190-bed Mercy Hospital-Ardmore deployed Qventus' AI software to reduce bottlenecks in the ED caused by backups in ancillary care departments, which ultimately slowed disposition to admit times. Nationally, hospitals in the top 10th percentile report disposition to admit times at 42 minutes, according to CMS. Mercy Hospital-Ardmore, on the other hand, reported an average of 88.6 minutes. "It's a hospital-wide problem, not just an ED problem, to get your patients admitted," says Jennifer Bramlett, RN, director of emergency services, catheterization lab and logistics at Mercy Hospital-Ardmore. The hospital's existing analytics solutions of- fered only a general analysis of department performance. She couldn't discern which specific patient handoffs, processes or individual clinicians were the root cause for patient backups in the ED. As a result, it was tough to quickly target breakdowns between departments. How Mercy Hospital-Ardmore used AI to address throughput issues Mercy Hospital-Ardmore worked with Qventus to help its ED re- AI lets hospitals extract greater meaning out of the millions to billions of dollars they've invested in EMRs. Most importantly, AI empowers clinicians to achieve their goals to improve the quality of care and enhance the overall patient experience.

• Cover