Imagine, if you will ...

MGMA Connexion, Jul 2006 by Baldwin, Christian R

Using simulation to solve operational problems in medical practices

This article adapted from a professional paper submitted to the American College of Medical Practice Executives (ACMPE) in partial fulfillment of requirements to achieve the certification of Fellow. The topic falls under Planning and Marketing, one of eight performance domains identified in the ACMPE Body of Knowledge for Medical Practice Management.

Simulation modeling allows administrators to see the impact of proposed changes to their organizations. When time and money are limited, simulation allows you to identify capacity-increasing and cost-reducing actions. By identifying the best patient systems, you can improve clinical encounters and increase a practice's profitability.

Use computer simulation - also called computer modeling - to optimize patient flow operations by experimenting with input parameters to maximize profit. Take care in identifying your objectives. Do you want to increase revenues by boosting patient volume? Reduce costs or patient waiting times? A combination of these goals?

After practice leaders set and document goals, conduct a baseline analysis in two steps: Develop a process diagram and create the simulation.

Developing a process diagram

The process diagram is a detailed flow chart of all patient processes. It documents each resource that the patient touches or encounters. Diagram each system independently and then join them to show the relationships among departments. The process diagram can reveal how one minor change in a department can affect another department.

The process diagram provides the foundation for the simulation model. To hring it to life, conduct time and motion studies on each process. l:or example, it you aim to increase patient volume, study patient arrival patterns to create an accurate representation for the simulation.

Depending on your goals, record other elements that affect patient flow, such as:

* The amount of time each physician spends with each patient type;

* Patient waiting times;

* Productive time in exam rooms;

* Lxam-room use levels; and

* Distribution of patients' time in the exam room vs. time during the entire clinic visit.

After studying each such process, use the process diagram to create the simulation model.

Building the model

First, choose a software tool. You can purchase a basic simulation package for around $1,000; a package that will allow you to model the complexities of a physician practice can cost around $20,000. A few consulting companies offer simulation services specifically for medical practices.

Model creation starts by importing the process diagram from the software package used to create the diagram to the simulation package, or looking at it and recreating each process in the simulation. Include all practice resources - e.g., physicians, nurses, receptionists, technicians and equipment and their costs. Assign a resource or set of resources to complete each task.

If the process being defined is the physician's patient exam, the doctor might spend five to 30 minutes, depending on patient type. Data collected from the time study help predict the distribution and variation of times required for each patient type and associated resource. This is important due to variability in time required tor different physicians to see different patients and the inherent unpredictability in patient patterns. Shape the simulation for each physician's real-world time parameters.

You develop each process and patient flow pattern in similar fashion until the model logic is complete. The final step of model development is adding animation, which allows the model to mimic the real system by importing the floor plan for the building and creating animated characters representing staff and patients.

Identifying areas of need

The simulation model should now accurately represent your group's environment. You can define time trials for any period, e.g., one day, one week, a year. Include the opening and closing times of each day so the model starts and stops correctly. Staff and patients enter according to their schedules; real-life variation is represented as delays in the model according to the variability in the live system. (Patients don't always show up on time, and doctors frequently run behind schedule.)

As the model runs, it becomes obvious where bottlenecks occur. Queues build where inefficient processes exist; waiting times appear at certain times of the day. You can prioritize projects from a high-level analysis of the simulation, targeting inadequate resource use, excessive waiting times and high costs.

Analyze areas of need by studying operational components. The simulation model will produce statistics on resource use, process times, queue times, transfer times, busy costs, idle costs and more. For instance, a department might have high patient waiting times and low resource use for the nursing staff. The project team would want to learn the reasons for the inefficiencies.

Documentation of results, model verification and validation