When we encounter an unexpected critical health problem, a hospital’s emergency department (ED) becomes our vital medical resource. Improving an ED’s efficiency and timeliness of care, while reducing avoidable readmissions, is fraught with difficulties arising from complexity and uncertainty. In this paper we describe a new ED decision-support system that couples machine learning, simulation, and optimization to address these improvement goals. The system allows healthcare administrators to optimize workflow globally, taking into account the uncertainties of incoming injuries and diseases and associated care, thereby significantly reducing patient length of stay. This is achieved without changing physical layout, focusing instead on process consolidation, operations tracking, and staffing. First implemented at Grady Memorial Hospital in Atlanta, Georgia, the system helped reduce length of stay by roughly 33%. By re-purposing existing resources, the hospital established a care management observation unit that led to a reduction of 28% in ED readmissions. Insights also led to an investment in an alternative-care facility that removed more than 32% of the non-urgent-care cases from the ED. With these improvements the hospital enhanced its financial standing and achieved its target goal of an average ED length of stay close to 7 hours. ED and trauma efficiency improved throughput by over 16.2% and reduced the number of patients who left without being seen by over 30%. The annual realized revenue and savings amount to approximately $190 million (up 72%), a large amount relative to the hospital's $1.5 billion annual economic impact. The underlying model, which is generalizable, has been tested and implemented successfully at seven other EDs and in 2 other hospital units. The system offers various advantages in that it permits a comprehensive analysis of the entire patient flow from registration to discharge, enables the decision-maker to understand complexities and inter-dependencies of individual steps in the process sequence, and ultimately allows the user to perform system optimization.
Organization overview: Slide # 3
Problem and challenges: Slide # 9
Approach and methodology: Slide # 18
Results, impact, and conclusions: Slide # 27