Title: MHS 2020 -- Envisioning tomorrow to focus today s resources.

Richard Gromadzki

Andrew Cornell

Session Description:

There is every reason not to study the future, to just deal with the tremendous problems and challenges of today. CEOs sometimes state that there is so much uncertainty, they are not sure how they will deliver care in the next six months. They routinely question why we should consider what might happen 25 years from now. Many of our tools for planing and programming do not provide the ability to prepare for a rapidly changing future. Given the uncertainty, the chaos, the rapidly changing conditions within the global community, why study the future? The future will come whether we attempt to plan for it or not. Why water the seed?

We water the seed because there is almost no one in the United States that medicine has not touched, either directly through the health system or indirectly through research and development.

We water the seed because of the increased complexity of our world, the increased pace of change, and the increasing number of worldwide conflicts due to peace with the former Soviet Union.

Finally, we water the seed because there is tremendous opportunity to create our future, the future of health care in the United States.

Therefore the seed has developed into a health care simulation to help CEOs understand the relationships between initial conditions or investments and end states in the year 2020.

Session Objectives

• How to predict, prevent and manage health
• Cost effectiveness of simulations
• Productivity and performance measurement
• Patient and customer satisfaction
• The future of health plans

Session Outline

• Virtual conference
• Futures Information Relationship Modeling System (FIRMS).
• Global key forces
• Introduction to complex systems
• Simulation normative sources and relationships
• Simulation modules
• Collaborative tool

Session abstract

At the outset of the MHS 2020 project, it was decided that participants in a cyber-conference environment would use a virtual conference Internet site to "argue" different points of view. Users would logon to various workgroups and add comments and ideas from widely dispersed geographical sites. The result was a collection of related thoughts. These thoughts were then to be logically grouped to help solidify the content and context. The knowledge developed was to be incorporated into a document that summarized and presented the various views in order to help achieve consensus on a "fuzzy picture" representing the future of health care.

This process was initially planned to be accomplished manually. It was then proposed that the process of retaining, re-formatting, and displaying the results of the conversations with an automated tool and relational database. This SRA developed automated tool has become known as the Futures Information Relationship Modeling System (FIRMS).

FIRMS is a three-part system that allows conversation pieces to be modeled, stored, and displayed graphically. The input was the on-line text of conversations from the virtual conference Internet site. Data from the virtual conference was reviewed and analyzed in order to populate the FIRMS database. Research analysts dissected the conversations into conversation models consisting of four kinds of nodes: Key Forces, Forecasts, Visions, and Leverage Areas. Once the modeling was complete, the data could be viewed graphically using the FIRMS Graphical User Interface. The graphical representation of the conversation models helped the research analysts in assessing the success of the on-line conference, and helped the conference participants understand the structure and relationships in their conversations.

In the first phase of the project, the participants were asked specifically to generate the nodes mentioned above: key forces, forecasts, visions, and leverage areas within their areas of expertise. This phase was primarily an exercise to assist the participants in thinking about the future of health care. In the second phase, the conversations were more structured. Four potential futures (known as scenarios) were discussed and each workgroup had specific topics, or areas of discussion, to address. The narrowing of the on-line discussions was necessary in order for the final report and graphical output to be successfully completed by the end of the project. Additionally, while the participants may have found the structure of Phase II to be limiting, it allowed them to focus on the issues that were most relevant when considering the future of health care.

The Sim2020 Model is a descriptive, interactive, dynamic simulation model of a health care system. Using a baseline set of conditions and relationships reflecting health care delivery today in the Military Health System (MHS), the model supports analysis and provides insight into the future evolution of the system. The model is designed not to be prescriptive; that is, not to provide an ultimate solution but rather to help the user develop a greater understanding of the relationships between initial conditions and end states.

The purpose for developing the Sim2020 model is twofold: to serve as a learning tool for health policy decision-makers and to aid the MHSS 2020 project. MHSS 2020 has become a multi-year, Tri-Service research project to explore future trends in health care and develop scenarios for thinking about health systems in the year 2020. The Sim2020 model supports this effort by capturing system relationships identified by the MHSS 2020 participants and demonstrating the sensitivity of identified leverage areas in the various scenarios as developed. Because MHSS 2020 is a work in progress, the model is always in flux and has been and continues to be improved as we learn more about the complex system that it represents.

Version 1.3 of the model was refined as new relationships were identified during the MHSS 2020 project. Version 1.3 was developed using iThink simulation v3.06 software from High Performance Systems. The developers selected dynamic simulation over other modeling techniques because it is the best approach for capturing the unique features of a complex system. The developers selected dynamic simulation over other modeling techniques because it is the best approach for capturing the unique features of a complex system. Complex systems are comprised of high-order, multiple-loop, nonlinear feedback structures, and their behavior depends on the actions of many subsystems. They can be characterized by the following features:

• counter intuitive: cause and effect are distant in time and space
• resistant to policy changes: typically a one-step level response
• multiple leverage points: high sensitivity to a few parameters that are often difficult to identify
• system counteraction: compensating counteraction resists applied forces; better results occur by modifying structural incentives
• long-term vs. short-term: trends often become worse before they finally improve in response to corrective action
• tendency to drift to low performance: short-term corrections turn bad, followed by a redoubling effect of the original bad decision.

Another approach taken by the developers was to include all relationships deemed important to the operation of the system, even if the confidence levels of those relationships were low. The inclusion of such relationships tend to provide better results than if they are excluded. For this reason, the model was developed with a deep structure with multiple relationships. SIM 2020 s major advantage is its inherent flexibility. Assumptions and relationships can be tailored to more accurately reflect the unique characteristics of a specific organization. Thus, the model serves as an ideal collaborative learning tool.

Within the interactive environment of the simulation model, users have the opportunity to learn more about the sometimes chaotic and counter-intuitive workings of a complex system. Users can identify where the areas of leverage are located in the system and how they react over time to various policy, technology, or resource changes.

Another approach taken by the SIM 2020 developers was to include all relationships deemed important to the operation of the system, even if the confidence levels of those relationships were low. The inclusion of such relationships tends to provide better results than if they are excluded. For this reason, the model was developed with a deep structure with multiple relationships. Experts from the MHSS 2020 have provided enhancements to these and all relationships currently in version 1.3 of the model. The model s deep structure permits better support sensitivity and scenario analyses.

SIM 2020 is a dynamic simulation tool that enhances management s ability to make the best possible resource and policy decisions. SIM 2020 provides feedback regarding a wide range of strategic decisions as show in the chart below.

SIM 2020 STRATEGIC DECISIONS

The model has been used in a number of interactive workshops within the MHSS. Through these tailored workshops, organizations have been able to gain insights into the complexity of their health system, the relationships at work, and the insights resulting from a shared mental model of a very complex and dynamic military health system.

SIM 2020 WORKSHOP FUNCTIONAL REQUIREMENTS REPORT WILL BE PROVIDED DURING THIS PRESENTATION ALONG WITH A LIVE INTERACTIVE DEMONSTRATION OF THE MODEL.