QUERI – Quality Enhancement Research Initiative

Go to the HSRD website

Quality Improvement Methods

Simulation Modeling

a. Definition: Discrete event simulation is the use of a computer model to replicate operations in order to gain understanding of a system being modeled. This is done without requiring changes to the real system itself. Computer software is available to facilitate creation of such models which incorporate the random variations and logic of the actual events being modeled. Simulation modeling includes constructing the computer model, confirming its validity in reflecting the real system modeled and experimenting with the computer model so as to forecast likely outcomes in terms of operations or quality. Thus discrete event simulation provides a safe and cost effective way to experiment with proposed improvements. Simulation models are used to investigate functions in all areas of healthcare including patient movement, information flow and disease processes. Specialized software is available for healthcare modeling but generic simulation software is frequently used in health care as well.

b. Literature:

  • Rutberg, Matthew Harris, et al. "Incorporating Discrete Event Simulation Into Quality Improvement Efforts in Health Care Systems." American Journal of Medical Quality (2013): 1062860613512863.
  • Jacobson, Sheldon H., Shane N. Hall, and James R. Swisher. "Discrete-event simulation of health care systems." In Patient flow: reducing delay in healthcare delivery. Springer US, 2006. 211-252. (Provides many references)
  • Jun, J. B., S. H. Jacobson, and J. R. Swisher. "Application of discrete-event simulation in health care clinics: a survey." Journal of the operational research society 50.2 (1999): 109-123.
  • Günal, M. M., and Mike Pidd. "Discrete event simulation for performance modelling in health care: a review of the Literature." Journal of Simulation 4.1 (2010): 42-51.

Several simulation software packages are available some of which were created specifically for healthcare. These include:

  • From ProModel Corp.: MedModel, Clinical Trials Simulator, and Process Simulator
  • From Rockwell Automation: Arena simulation software.
  • From FlexSim Software Products: FlexSim Healthcare Simulation
  • From Lanner Group Ltd. WITNESS simulation software which has a version for pharmaceutical, consumer health and medical products manufacturing.

c. Example: Many functional areas in healthcare have benefited from the use of digital simulation models. The discrete events simulated can be such things as a clinical procedure, an administrative decision or patient attributes. Particularly popular are models of a hospital ED, surgery, outpatient clinics and various ancillary departments to model patient flow and to develop and test improvement alternatives in these areas. The simulation models can be built using available hospital data on the timing of patient movement and processes. Often the programming and resulting model are displayed graphically and animated. The model's intent can be to improve productivity, quality, resource utilization or other attributes.

An Example might be to determine the impact of implementing a new type of equipment where there is a choice in the number of devices to purchase. Various combinations scheduling and staffing the use of the devices could be tried and the simulation used to forecast the effect on patient visit time, costs and patient outcomes.

A department or clinic's patient flow can be modeled. The model can be displayed as a flow diagram on a computer screen with icons of patients and staff moving about a diagram. Shown below is a diagram for a simulation model of patient flow in a colonoscopy clinic that was used to improve patient and staff schedules.

Simulation model of a GI clinic

d. Steps:

1) Define the problem, objectives to be addressed, and the scope of the model

2) Gather data needed to define the model's attributes such as volumes, times and patterns of flow

3) Design and program the simulation model incorporating the descriptive data

4) Test (validate) the model to be sure it reflects the situation being modeled, including its ability to correctly react to changes to the real system modeled.

5) Run potential improvement using the simulation model and evaluate the forecasted simulation outcomes. If the improvements appear worthwhile in the simulation, test the changes in the real life system to assure that these improvements can be achieved.