Healthcare Engineering Research

The annual U.S. expenditure on healthcare in 2003 was estimated at $1.5 trillion. This expenditure is expected to almost double and reach $2.8 trillion by the year 2011. Healthcare accounted for 13.2% of the Gross Domestic Product (GDP) in 2000 and may reach 17% of the GDP by the year 2011 (Health Affairs, 2002). Hospitals, which are the single largest item on this budget, are expected to account for 27% of the total projected healthcare expenditure by 2012. This estimation represents a decrease in this expenditure, down from 31.7% in 2001 (Price Waterhouse and Coopers, 2003). Similar results were obtained from the Israeli Central Bureau of Statistics (ICBS). According to the ICBS the expenditure on hospitals accounted for 36% of the total annual healthcare budget which reached 43 billion NIS in 2001 (8.8% of the GDP). These figures are true for all industrialized nations. The conclusion is one; healthcare imposes a great burden on the national economy. Therefore, tools and methods have to be developed to enable better and more efficient control over the process of providing healthcare.

The Emergency Department (ED) is the largest conduit through which patients enter the hospital. The ED basically acts as the hospital's 'gate keeper', determining if a patient needs to be admitted or can be discharged. This makes the ED an important link in the healthcare providing chain. In addition, because of its unique characteristics, the ED is also the most difficult department to manage among all hospital departments. The resources the ED operates have to be highly versatile and flexible to be able to treat a large array of incidents ranging from minor cuts and bruises up to life threatening situations. In addition the ED operates in a highly dynamic environment that is beyond its control and as a result has to have the ability to react quickly to fast unfolding events.

Overcrowding, diminished resources, limited ED bed capacity, prolonged waiting times, low staff morale are only a few of the problems ED managers, and other hospital decision makers are struggling with, on a daily basis. The wide scope of these problems and the mutual relations between the different elements that make up the ED, complicate considerably the operational decisions which may be required to keep the ED operating in a satisfactory manner.

Faced with these circumstances, coupled with an increase in patients' awareness and demand for high quality fast and efficient treatment, hospital managers and other healthcare policy makers are being forced to search for ways to distribute efficiently scarce resources, reduce costs, improve productivity while maintaining the highest treatment standards and quality.

From an engineering point of view it may seem obvious that standard Industrial Engineering design and analysis tools such as mathematical modeling (optimization) and simulation are the solution to the baffled hospital manger. As it turns out this is not the case. Healthcare systems are much more complex; they involve multiple decision-makers and literally deal with life and death situations as opposed to all other service and manufacturing systems.

Making Simulation More Accessible To Emergency Department Decision Maskers

System professionals in general agree that discrete-event simulation tools are particularly suitable for modeling these types of systems. Simulation models can provide management with a reasonable assessment of the ED’s efficiency, resource needs, utilizations and other performance measures in face of dynamic external as well as internal changes. Simulation can assist hospital management in understanding the mutual interactions between the different system parameters and their effects on the system’s performance and as a result, help them to develop and enhance their decision-making skills needed for evaluating different operational alternatives in order to improve EDs operations. However, these facts are not so obvious to physicians and other healthcare policy makers. As a matter of fact there is very little enthusiasm and even willingness, among healthcare professionals, to accept these type of tools as viable modeling and analysis methods. One major stumbling block is the reluctance of hospital management, and the physicians in charge, to accept changes, solutions or even recommendations, that stem from a 'black-box' type of tool. The reason simulation models are not used more often in healthcare settings is management’s lack of incentive to do so. Management often does not realize the benefits to be gained from using simulation-based analysis tools, but on the other hand, management is well aware of the time and cost that have to be invested in building detailed simulation models. Simulation is viewed by some as no more than an elaborate computer game and therefore, not a viable modeling and analysis tool. In a recent article entitled "Hospitals biased against optimization" the author claims that healthcare policy makers feel that spending money to improve systems' operations only diverts funds from patient care.

In spite of the above criticism healthcare professionals are not to blame since industrial engineers, simulation experts and consultants have not made simulation easier for physicians to accept and use. Most simulation projects are conducted by external experts or consultants that come in for a short time don’t always make sure to involve sufficiently hospital personnel in these projects and in some cases do not grasp the magnitude of the problems the physicians are faced with.

In order to accelerate the proliferation and acceptance of simulation modeling by healthcare professionals it is essential to get hospital management directly involved in the development of these projects and if possible to move the development out of the external consultants' hands and trust it in the hands of hospital technical personnel. The problem with this approach is that in most cases hospital personnel lacks the experience and the know-how required for the development of comprehensive simulation models. Therefore, the simulation tools have to be simple and intuitive but at the same time retain enough modeling flexibility and capabilities. By incorporating these principals, management’s involvement in developing simulation models will increase, and as a result, the confidence in the model's ability will grow as well. At the same time, due to a decrease in the effort required to develop new simulation models, management’s incentive to use simulation will hopefully grow.

In order to be able to develop such a tool two basic conditions have to hold: first the process patients go through in the ED has to be similar for different EDs and if differences do exist between EDs, they have to be limited to a few well defined parameters which can be easily estimated and changes. If these conditions hold, all that is needed is to identify the basic general process.

How does this fit in the scheme of things?

1. In general there are different modeling options. On the one end of the modeling spectrum we have simulations tools which are based on generic activities. These tools are highly abstract and as such are flexible to model almost any system or scenario. However, a considerable amount of knowledge and experience is needed to use these tools efficiently.

2. On the opposing end of the modeling spectrum lays the dedicated simulation model which can only be used to model the system it was deigned for. It is very difficult to reuse these models especially if they were not developed with reuse in mind. However, due to the low abstraction skills required from the user, it is quite simple to perform what-if type analysis over a predetermined (by the simulation expert or consultant) set of parameters that belong to the original system.

3. In the middle of that spectrum we have a tool which is based on a generic process. The generic process has to be general enough to be capable of modeling a class of systems each of which uses some derivative of the generic process. People that operate the systems which belong to a certain class have to be familiar with the governing processes and therefore, can adjust the generic process to fit their specific needs.

Based on data from a time and motion study, hospital information systems and interview with senior hospital and ED staff 19 different patient process charts, of eight basic patient types, were developed each describing the process a patient of a specific type goes through in one of the hospitals that participated in the study (6 out of 25 general hospitals in Israel).

As it turns out, patient type has a higher impact in defining the process which patients experience when visiting the ED, than does the specific hospital in which the patients are treated. This also means, a generic process can be defined for each major patient type. This was achieved by combining the individual process charts together.

The next test revealed that aggregating the duration of the elements, that make-up the patient process, over patient type and hospital type, actually improves the duration precision of these elements. This means, default duration values for the different process elements can be suggested.

Consequently, it makes sense to develop a simulation tool that is based on the above generic process: A tool that is simple, intuitive and relatively inexpensive to be used by hospital personnel but yet one that retains its generality and can be used to model different aspects of the ED operations.

During the last four years a simulation tool that follows these guide lines was developed at the Industrial Engineering and Management Faculty at the Technion. The main objectives of the project, which was funded by the Israeli National Institute for Health Policy and Health Services Research, were to develop a general, flexible and reusable tool that can model different EDs while at the same time retain its simplicity. The entire study was summarized in two papers. The first paper, published in 2005 in the IIE Transactions journal laid the foundation needed for to the development of the simulation tool.

The paper claims that hospital ED administrators, like managers in many other types of organizations, think their environments are unique, and that they need customized decision-support tools for operational decision-making. However, the study shows that the various EDs have a great deal in common.

From an extensive study of over 16,000 patient visits to EDs, it is shown that it is possible to classify patients into just eight patient types. More importantly, it is possible to design a single unified patient process flow chart that covers all eight patient types. The process flow chart represents the full collection of examinations and laboratory tests that a typical emergency room patient might need to undergo including the sequence in which they should performed.

Analysis of patient visits also led the conclusion that the time required to perform each of the examinations and tests, for each particular type of patient, were similar enough from one emergency room to the next that one could establish default values for use in the simulation. This reduces or eliminates the need for detailed time and motion studies that would usually be required.

With this understanding and quantification the next step was to develop a method to forecast patient arrivals that could be applied to all five hospitals with different parameters to reflect differences in hospital size and mix of patients. Putting all of these together, results in a generic simulation tool which requires only a modest amount of data input. The simulation tool can now be used in-house by technical hospital personnel to aid hospital management with decisions regarding staffing, equipment utilization, reducing patient waiting times and hospital operating costs.

The second paper that is currently under review describes the simulation tool and analyzes its capabilities using six major emergency departments in Israel. Currently the research team is working on making the simulation tool accessible through the internet to hospitals in Israel and abroad.

This study got the attention of both researchers and healthcare professionals. The study was mentioned in professional magazines such as Industrial Engineer (March 2005) and the Hospitals and Health Networks magazine (September 2005).

The reason for this attention is related to the significance of the proVol. 43, No. 14, pp. 2977-2996blem and its economical potential. According to the American College of Emergency Physicians (June 2003), the cost of Emergency Department (ED) operations amounted to 5% ($75 billion) of the total US healthcare expenditure (This means the ED takes up close to 16% of the total budget allocated to hospital operations). In Israel the cost of ED operations is estimated at 2.5 Billion NIS. This means that even a very moderate improvement in the operational efficiency of the ED can be translated into considerable savings in the healthcare expenditure. Combining these numbers with the fact that most systems especially complicated and sophisticated ones as healthcare systems are, can always be improved, turns this project into a very promising one.

Sinreich, D. and Marmor, Y. (2005), "Emergency Room Simulation" Research Executive Summaries, Edited by Candace Yano, Industrial Engineer, Vol. 37, No. 3, pp. 52-53.

Greene, J. (2005) "Simulation Mode: New patient-tracking technique could be a boon for medium-sized hospitals" (an interview with D. Sinreich on ED simulation), Hospitals & Health Networks Magazine, Vol. 79, No. 9, pp. 24-26

Papers

• Sinreich, D.and Marmor, Y.* (2004) " A Staffing Emergency: Keeping Emergency Departments Mooving Quickly and Eficiently with Simulation", Industrial Engineer, Vol. 36, No. 5, pp. 38-41
  
  
• Sinreich, D. and Marmor, Y*. (2004), "A Simple and Intuitive Simulation Tool for Analyzing Emergency Department Operations", Proceedings of the 2004 Winter Simulation Conference (WSC), Washington, DC, on CD.
  
  
• Sinreich, D. and Marmor. Y.* (2005), "The Operations of Hospital Emergency Departments: The Basis for developing a Simulation Tool", IIE Transactions, Vol. 37, No. 3, pp. 233-245.
  
  
• Sinreich, D. and Marmor, Y.* (2005), " Ways to Reduce Patient Turnaround Time and Improve Service Quality in Emergency Departments", Journal of Health Organization and Management Vol. 19, No. 2, pp. 88-105.
  
  
• Sinreich, D. and Marmor. Y.* (2005), "Emergency Department Operations: A Simple and Intuitive Simulation Tool Based on Generic Processes", working paper.
  
  

Presentations

• The 2004 Winter Simulation Conference, Washington DC, December, 5-8, 2004.
  Paper: Sinreich, D. and Marmor, Y. "A Simple and Intuitive Simulation Tool for Analyzing Emergency Department Operations". Speaker: D. Sinreich.
  


• Research Workshop on Emergency Department Operations: Collaboration between Medicine and Management, Desautels Faculty of management, McGill University, Montreal, Canada, May 11-12, 2006.
  Paper: Sinreich, D. and Jabali, O., "Improving Emergency Department Operations Using a Simulation-Based Decision Support System".

Current Funded Research

• "An Expert System for the Cost Effectiveness Evaluation of Operational and Structural Changes in Emergency Departments", The Israeli National Institute for Health Policy and Health Services Research (NIHP), Investigators: D. Sinreich and P. Halpern

Mental Models as a Practical Tool in the Engineer’s Toolbox

Industrial engineering methods are very successful in coping with well-structured systems and processes. However, when it comes to analyzing, planning and controlling systems which contain unstructured processes, managers and engineers are faced with a much more difficult task. This is especially true in systems where teams and individuals have a significant role in the daily operation, monitoring and decision making. In these situations, the processes may be performed differently by different individuals depending on the their perceptions, concepts, ideas and perceived system status all are denoted as the operators’ Mental Model (MM) of the system. This study develops a similarity measure to quantify the differences between MMs. It’s done by eliciting the operators’ subjective perceptions of system and their role within it (Mental Model), and comparing them to a standard description reference model which represents management’s policy of how should the system be operated. Analyzing the differences between these models may facilitate intervention approaches in closing these gaps and may help in creating better-synchronized and synergistic teamwork. Using this similarity measure is demonstrated in a hospital Emergency Department.

Papers

• Sinreich, D., Gopher, D., Ben-Barak, S.*, Marmor, Y.* and Mentchel, R.* (2005), "Mental Models as a Practical Tool in the Engineer’s Toolbox". International Journal of Production Research. Vol. 43, No. 14, pp. 2977-2996.