For all of us who have sat for hours because doctors were running late, this may not fall into the category of being good news.
It turns out that the queuing theories we learned in school, not surprisingly have become quite sophisticated thanks to the powers of modern computing. In fact, there is now a specialty known as Dynamic Process Modeling (DPM). Two practitioners, Raid Al-Aomar of Abu Dhabi University in United Arab Emirates and colleague Mahmoud Awad of ALHOSN University, have developed a computer model that could help healthcare providers’ practice management staffs cope more effectively with "no shows." This sounds benign except for the fact that the output of their DPM solution is the establishment of “an optimal rate of overbooking.”
Image via Shutterstock
Yes, you read the last sentence correctly.
From the healthcare provider perspective, as opposed to that of we the customers, “no shows” are a problem. Making strategic use of overbooking is the recommended solution.
How big a problem is this? As the researchers found it is can be substantial. According to the authors, as a result of inefficiencies due to walk-in and no-show patients:
- The cost of no shows can be up to 14 percent of the income of a health practice
- A high no-show rate of 30 percent can cut almost half of clinical efficiency to around 55 percent
In an era where cost-containment has become a top priority, and where healthcare providers are looking for ways to save every dollar possible in operations expense without sacrificing the quality of care, it is no wonder this became a topic of intense interest.
By the numbers
An easy way to think of this is with the analogous situation of airline travel. Struggling for profits, airlines must use every trick in the book to try to fill every seat possible on every flight. This includes not just adjusting schedules so that poorly used routes are discontinued, but also they now have gone seat pricing by seat site location (even by whether you have a window, middle or aisle seat in economy class), time and day of flight, time of purchase, etc. All of this is to “farm” to maximize yield and is augmented by what is supposed to be the judicious use of overbooking. The same is true for healthcare facilities of all types and sizes.
The two authors presented their solution in a paper, “Dynamic process modelling of patients' no-show rates and overbooking strategies in healthcare clinics.” The abstract says it all:
“…a dynamic discrete event simulation model of a clinic operation is developed to assess current clinic performance under various no-show rates and patterns and test several overbooking strategies to improve clinic's efficiency and enhance performance. The approach significance includes the analysis of variability in scheduling and treatment times, the impact of patients reneging, and walk-in patients. In addition, the paper develops a unified clinic utility function that analyses the impact of overbooking on clinic revenue (patients treated), capacity (utilisation), and customer service (wait time). An example of a general practice (GP) healthcare clinic is used to illustrate the development and application of the proposed approach.”
While this certainly is a nice application for DPM, and possibly a significant cost and time-saving tool for the healthcare providers, it makes you wish the authors had gone a step further.
Wouldn’t it be nice if some big data business intelligence could be placed on this? Healthcare providers could query patients in near real-time as to their intentions in order to cut down on no shows. Patients could be notified of waiting times. In fact, think of this as being told when interacting with an IVR that your hold time to an agent is X minutes and you are Y in line. It certainly would make the waiting room environment less hostile along with improving work flow.
In busy hospital settings there is always going to be the problems associated with walk-ins and emergencies. Coordination between hospitals running a DPM-based system with urgent care facilities in the area could ease the burden somewhat on this front. With other care facilities, they might try what one of my doctors already does which is to set aside various days and hours for patients who do not have scheduled appointments.
The nature of the no-show and walk-in problem should not be scoffed at. This is a very serious problem, and as noted this was an area where DPM can be extremely valuable. However, in the era where focus on “the customer experience” is the mantra of almost every industry, it is hoped that healthcare providers take a holistic view of using such as system. There are multiple audiences whose needs should be considered and actions have consequences. They are in a competitive industry. For other than emergencies, they could “farm” themselves into a problematic situation made worse by the increased use of social media by those whose experiences are less than desirable.
It will be interesting to see if such systems are widely deployed, along with the fine-tuning that is likely to take place. It will also be interesting to see where DPM is applied next.
