Queuing theory has become part of the foundation of patient flow theory and application as it is applied to hospital processes. It is understood that as hospital occupancy rates increase, wait times will increase as well. In fact, as occupancy rates surpass 90%, hospital processes actually begin to slow down as more patients "queue up" for resources that do not increase as occupancy increases. For example, despite a high occupancy rate, the hospital retains the same number of CAT scan and MRI machines, the same number of stress testing equipment, and so on. Therefore, more patients are lining up for the same number of resources.
Queuing theory is based on these four premises (Jensen, Mayer, Welch & Haraden, 2007):
- As occupancy increases, wait times and service delays increase exponentially
- Unscheduled or uncontrolled arrivals will behave in characteristic fashion
- A balk is an arriving customer who sees a long line and does not seek service
- Reneging occurs when a customer gets off a line
These theories reinforce commonly seen patterns in our hospitals. As we just discussed, increased occupancy rates will result in delays in patient care processes. But even beyond this are some expected delays that happen almost on a daily basis in our organizations. If asked when your hospital experiences most of its daily delays in the emergency department (ED), most of us would answer by saying mid- to late afternoon. If your hospital has high capacity issues, this ED pattern probably happens almost every day as the patient walk-ins and ambulance traffic to the ED increase. One might also comment that the post-anesthesia care unit (PACU) gets backed up on busy days around the same time as well. These patterns are predictable, but are they preventable? If one considers that they happen in characteristic or predictable fashion, then one might also consider that the ED can adjust resources and other care processes in anticipation of these patterns.
-Toni Cesta, PhD, RN, FAAN and Beverly Cunningham, MS, RN