The Quality-Cost Connection: Before implementing changes . . . simulate!
The Quality-Cost Connection: Before implementing changes . . . simulate!
Helps visualize, analyze, and predict performance
By Patrice Spath,
RHIT
Brown-Spath & Associates
Forest Grove, OR
Implementation of actions that achieve intended goals is the primary purpose of all process improvement projects. Whether your organization is seeking to reduce process inefficiencies or eliminate the chance of unintended patient harm, action taking is a critical step in the improvement cycle. The cycle involves devising a new or improved process, implementing changes, monitoring the effects of changes, making further adjustments where necessary, and continuing to monitor.
Repeating this cycle over and over again drives ever-improving quality and patient safety. Yet the process improvement cycle has some disadvantages. Implementing process changes can disrupt work activities and potentially compromise patient safety and satisfaction. Continual monitoring of the effects of redesigned processes is resource intensive and difficult to accomplish for lengthy time periods.
Simulation modeling, commonly referred to as simulation, offers a way to apply the process improvement cycle without actually implementing the proposed action plans. In using simulation, a model of a real system is designed and experiments are conducted to better understand the behavior of the system and/or evaluate various strategies for improving system operation. Simulation is a technique that helps to visualize, analyze, and predict the performance of a system without the cost and risk of disrupting current work processes.
Because of its great versatility, flexibility, and power, simulation is a valuable way of studying current processes in detail to pinpoint problem areas and identify opportunities for improvement. New policies, operating procedures, organizational structures, and information flows can be explored without affecting ongoing operations.
One of the greatest strengths of simulation lies in the ability to explore "what if?" questions. Theoretical process changes can be quantitatively studied and compared. In addition, simulation can be used to forecast system requirements such as staffing needs, capacity requirements, resource utilization, and expenditures with great precision.
Every health care system is made up of a network of processes and structures. Changes in one activity have a ripple effect throughout the system. For example, a change in the inpatient admission process can affect the way that nurses receive new patients on the units.
Revisions in the medication distribution system impact many clinical processes. In simulation, graphic models are developed to functionally represent the system. For example, a flowchart or model of the hospital accounts payable system describes the series of steps involved and decisions made when a hospital bill is generated and eventually paid.
Simulation is a way to create an imaginary representation of the process so that various redesign choices can be tested to determine the effect, if any, on the process.
How simulation works
Simulation can help health care organizations avoid counterproductive and ineffective process changes, both in strategy and implementation. It is a cost-effective mechanism for quickly exploring many "what-if" scenarios to zero in on an optimum solution to a problem.
To demonstrate how simulation works, let’s look at the process of registering new patients during the day shift in the emergency department (ED). A patient arrives in the ED, enters the registration queue, and registration is completed on a first-come, first-served basis.
For demonstration purposes, let’s say that an average of 10 patients arrive each hour and registrations are completed in an average of 10 minutes. Two people staff the ED registration desk during the day shift. Simple mathematics tells us that the registration staff are kept busy 83% of the time (10 patients per hour "4 [2 staff members x 6 registrations/hour] = 10/12 = 83%). (See baseline performance data for the ED registration process, below.) During a simulation, we’d be asking how changes in this equation would affect various aspects of the process. For example, what if an additional staff person was added? What would be the effect of decreasing the average time for the registration process itself?
| Baseline Measures for ED Registration Times (6 months of data) | |
|
Average number of patients registered per hour |
10 |
| Number of registration staff | 2 |
| Average number of patients waiting to be registered | 1.7 |
| Average patient wait time | 10.4 minutes |
| Maximum wait time | 57 minutes |
| Registration staff utilization | 82.3% |
|
|
|
To answer the "what-if" questions, the numbers are rerun two more times with different parameters. For example, we find that if another staff person is added, the time patients must wait before registering definitely will go down; however, staff utilization drops significantly, to 55.6%. (See hypothetical results from adding staff member, below.) Unless staff can fulfill other job duties during their idle time, adding a third registration clerk may not be a productive solution. If the goal of this inquiry is to determine the best way to reduce registration wait times in the ED, the right solution might be to redesign the registration process so that it takes less time. (See example on what can happen if the time for registration is reduced from 10 minutes to 6 minutes, below.)
| Hypothetical Results from One Additional Registration Staff Member | |
|
Average number of patients registered per hour |
10 |
| Number of registration staff | 3 |
| Average number of patients waiting to be registered | 0.9 |
| Average patient wait time | 5.4 minutes |
| Maximum wait time | 38.5 minutes |
| Registration staff utilization | 55.6% |
|
|
|
| Hypothetical Results from One Additional Registration Staff Member | |
|
Average number of patients registered per hour |
10 |
| Number of registration staff | 2 |
| Average number of patients waiting to be registered | 0.2 |
| Average patient wait time | 1.3 minutes |
| Maximum wait time | 20.1 minutes |
|
|
|
Computer simulations
The mathematics needed to answer "what-if" questions can get complicated, especially for complex processes. The example of the ED registration process was purposely made very simple, and therefore the calculations are relatively easy. However, in many instances, computer simulation models will be needed to thoroughly examine all the factors in complicated and dynamic health care processes.
A set of equations called waiting line or "queuing" theory can answer some of the questions. But the equations have some constraints that often are not met in the real world.
The parameters (average arrival and service rates and staffing levels) must be constant over time. Stable processes are uncommon in the real life of health services.
As the process becomes more complex and less stable, computer simulation models are useful. These models can handle the more complicated or dynamic situations often found in health care.
Computer simulation models allow you to vary both the flow and the inputs of the process so that the "what-if" questions about the record request process can be answered simply by entering a different number into the model and rerunning the simulation.
Computer simulations allow you to evaluate more complex (hence more realistic) processes. For example, how should the ED admitting desk be staffed when the hourly number of patients varies from a low of 10 up to 22? Or what happens if the computer used to register patients is not working properly? In this situation, the process defaults to a troubleshooting mode.
Either the staff person needs to gather the patient’s information manually (thus increasing registration time) or registrations are halted until the computer problem is resolved (thus increasing patient dissatisfaction). Simulation software is designed to consider all of the factors that might impact the process and help you determine the best solutions. And most important, you can do all this without making any actual changes in the process until you are fairly confident the changes will be effective.
A number of simulation tools are offered as statistical add-ins to spreadsheet software. For routine processes and tests of hypotheses, basic graphics, and even regression modeling, a statistical add-in product may be adequate. The functionality of products for use with spreadsheets is growing, particularly for risk analysis. For highly complex processes, a software product designed specifically for simulation modeling may be needed.
Not only can simulation software help you analyze both process inputs and outputs, many are able to exchange information with other software tools in an integrated way. For example, simulation may be integrated with presentation software to document and report on findings or to facilitate analyses on spreadsheet or statistical software. A large number of the vendors now provide families of products or modules rather than single, stand-alone software. SAS, a statistical software package commonly found in health care organizations, has add-on modules that can be used to build simulation models (www.sas.com). Many of the vendors have extensive web sites for further, detailed information, and many provide demo programs that can be downloaded from these sites.
For a list of vendors offering computer simulation software along with a description of the products, go to: www.lionhrtpub.com/orms/surveys/sa/sa-surveymain.html.
Simulation allows the user to visualize alternatives and obtain a fairly accurate analysis of how process changes will affect results. Before you start to use simulation software, it’s important to have a solid grasp of flowcharting techniques to understand your process and create the model.
Knowledge of statistics is necessary to design the simulation experiment and analyze the results. Finally, an understanding of simulation applications is necessary to validate and verify the model.
Most simulation applications provide tools that greatly assist the user. The advice and support of an expert, however, may be needed for creating the model and the framework for analysis. Health care organizations with a management engineering department may have their own in-house simulation experts.
Another place to look for assistance is the Society for Health Systems sponsored by the Institute for Industrial Engineers (web site: http://shs.iienet.org/).
Pick ideal’ improvement actions
Building a model rarely is an end in itself; instead, the goal of most analyses is to make a decision. Simulation can be used to study and compare alternative process designs or to troubleshoot existing systems. With simulation models, you can imagine how an existing system might perform if altered or explicitly visualize how a new system might behave before the process redesign solutions actually are implemented. The ability to easily construct and execute models and to generate statistics and animations about results is one of the primary attractions of simulation software.
Using simulation during the process improvement cycle can help the project team identify and address improvement opportunities and determine the best strategies for achieving improvement goals.
Whether your organization is seeking to reduce process inefficiencies or eliminate the chance of unintended patient harm, action taking is a critical step in the improvement cycle. The cycle involves devising a new or improved process, implementing changes, monitoring the effects of changes, making further adjustments where necessary, and continuing to monitor.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.