What would you think if your process improvement experts said: “we can remove 58% of a type of equipment that is often in shortage, provide better patient care, improve response time and never have a shortage again”? Yeah right! Well, that’s what happened. At The Hospital, PCA pumps were in shortage which created enormous frustration with the nurses and decreased patient care. Nurses couldn’t find equipment when needed to do their prescribed job.
PCA pumps are patient-controlled analgesia (PCA) is a method of pain control that gives patients the power to control their pain. The pump, which contains a syringe of pain medication as prescribed by a doctor, is connected directly to a patient’s intravenous (IV) line. The patient has control over when medication is delivered into the IV (with pre-set limitations of course). The patient pushes the button and gets a dose of pain medication. The advantages of using a PCA unit are:
- Patients having control over their pain medication have lower pain scores.
- Patients have greater satisfaction compared with nurse-administered medications.
- Free nurses from evaluating pain levels and administering pain meds to added availability to perform other work.
As you can image, it has got to be frustrating for the staff when a patient experiences more pain, is less satisfied, and the nurse is busier because of insufficient PCA’s.
Discovering the Root Cause
The Hospital had to perform “fire drills” to maintain an adequate pump supply daily. Adding more pumps improved access but significantly increased expenses.
Acquiring additional pumps to address the persistent shortage moved high on the budgeting priority. Adding pumps increases costs. A project was initiated to see if they could avoid the budget increase. The team needed to answer the following questions:
- When are pumps used during the day and by day of the week?
- What is the “right” number of pumps needed to support The Hospital?
- What is causing the disparity between inventory and availability?
- What process improvements would be needed to reduce inventory and increase availability?
This system was a perfect problem for discrete event simulation software. The Hospitals system records a date and time stamp for every PCA request. “We just had to extract the data and put it into Excel format.” Using the Healthcare Arrivals model object found in ProcessModel, the history of arrivals became inputs to drive the model in just minutes.
The tools supplied with ProcessModel not only analyze the data but also generates the ProcessModel code needed to predict what could happen in future weeks using simulation. Difficult data analysis became a single mouse-click. Pre-analysis of arrival data was a huge help in simplifying the model build.
PCA unit patient usage is recorded in The Hospital’s system so that proper billing is applied. This data could also be extracted from The Hospital’s system and converted into a distribution that mimicked the length of time for PCA usage.
The process improvement team then observed the handling of PCA units when the patient no longer needed them. Most soiled PCA units were moved to a storage closet to be picked up later and cleaned. The observed method was not the approved procedure, and real procedure not enforced. These observations allowed the team to make estimates of how long a pump would stay in storage before cleaning. The Hospital’s system also contained the PCA cleaning times.
A Simple Model
The process simulation used to depict the system was simple to build. Creating the model, formatting the data, and entering the information took less than an hour. This model development time included creating charts and graphs that illustrate the problem. Below is the model:
The model does not include detail for the wasting of medications, dual sign-off, etc. But it didn’t need all the detail, because it is the same in both the As-Is and To-Be. The amount of time used in the “other details” was very small in comparison to the usage and cleaning times. Removing less significant details reduces data collection, simplifies modeling and shortens project time.
The model allowed the reporting of the predicted PCA pumps needed by an hour of day and day of the week. The delays caused by the handing of the pumps after usage helped to identify the disparity between the inventory and the actual quantity needed. The graph below shows one week of demand vs. inventory. Multiple replication analysis with reports collected every hour provides a method of providing upper limit needs by day of week and hour of the day.
The team reviewed the existing procedure and identified additional possible solutions that might provide savings. The main issue was storage of soiled pumps in a multitude of locations. The mismanagement of pumps showed that on average pumps would return to service after more than 12 hours (some requiring much, much longer) had elapsed. One simple solution emerged — “If you are working with a patient when a pump is no longer needed, return the pump to a central cleaning area within 10 minutes.” This solution met with initial resistance until the nurses realized they are currently spending more time finding verses returning PCAs to a central location.
The new model provided a method of evaluating the simple scenario. The model showed the immediate return of the units would provide for a significant drop in the required inventory and reduced capital requests by more than $200,000. Some might label this projects as “common sense,” but that’s how all successful projects should look when completed. You know you have done a good job when you make the solutions look simple. The previous reality was that no one was able to answer the questions “what is the inventory of pumps needed based on demand, recovery, and cleaning.” The simulation answered those questions and provided a lower limit of PCA pumps, that if crossed, would start to cause waiting for PCA units. Furthermore, the simulation acted as a training tool to help staff understand the result of not returning pumps per the procedure.
Process improvement doesn’t always have to be long projects with huge budgets or projects that can only be tested by trial and error. This project brought the best of a Kaizen Blitz with detailed statistical analysis to provide patient care improvements, budget reductions, and job satisfaction increases for The Hospital’s staff.