Emerging Opportunities to Lower
 Transaction Costs through Health Care Data Interchanges

J.B. “Sean” Kenney, Ph.D.
NIHP Senior Fellow

The following paper is distributed by the NIHP to encourage comments, suggestions and dialogue on the respective topic.  At this time, the content reflects only the views of the author.

NIHP and the Four Keys to Health Reform
The American health system is extremely complex and political—and change is difficult.   Hundreds of organizations and thousands of individuals have weighed in on innovative ideas for health care reform. Many initiatives offer new ways to pay for health care for all Americans, while others address the quality and quantity of the product itself. What we need now is a common language—a clear framework—that captures the best of these ideas and offers an achievable vision for change.

Through four years of project work, the National Institute of Health Policy (NIHP), in consultation with its National Leadership Council, has identified Four Keys to Health Care Reform. These keys provide a framework for reform. They require strategic change from inside the health care system. They are readily achievable but depend on the continued and essential involvement of all current stakeholders.

The Four Keys to Health Reform encompass four major themes: 

     Key #1:  The greater use of evidence-based medicine (EBM)

    Key #2:  The application of similar principles for evidence-based operations
    (EBO)

    Key #3:  New methods to lower financing and transaction costs (LTC)

    Key #4:  A redefined and more active role for the consumer

This paper reports on progress toward lowering health care transaction costs - Key #3.

CHINS, CHMIS, MHMIS and other weird acronyms
Those of you working in health care in the late 80’s will recognize these acronyms. For those not familiar, CHINS represented a concept of Community Health Information Network to facilitate electronic information sharing among various stakeholders in health care. Community leaders were seeing this area as an opportunity to save costs and improve functionality as far back as this period. Unfortunately electronic vendors and/or a single health care system most often pushed the CHIN approach. This caused push back from perceived vendor interests and most often additional resistance from other hospital systems who did not want to collaborate with an approach tied to one of their competitors. 

One example was the WHIN, Wisconsin Health Information Network that had as sponsors, Ameritech and Aurora Health Systems of Milwaukee. The combined resistance from other community players, resulted in the system never getting to community-wide scale, even though there were some clear advantages for Aurora Health.

In the early 90’s a related concept was developed.  It was  called CHMIS, Community Health Management Information Systems. A very bright and thoughtful leader in the John A. Hartford Foundation introduced this concept. The broad coalition in Detroit (The Greater Detroit Area Health Council), invited the Hartford Foundation to present the model in 1992 to the community. While the presentation launched a Coalition sponsored program called MHMIS, Michigan Health Management Information Systems, we were never successful in getting one of the Hartford Foundation grants to support our efforts. It may have been one of the best things that happened to the MHMIS effort.

Up The Learning Curve
The basic reason for this failure was that the Hartford Foundation was pushing a model that “wired” all the hospitals, doctors, insurers, purchasers and others into a large community-based Data Repository where claims, enrollment, remittance, encounter, clinical outcomes and other elements would be managed for common community use. The politics of sharing data into “one pot” not surprisingly, doomed the initiative. A further element that defeated the model was that most of the community projects funded by Hartford began in the area of centralized claims processing for a unit fee.  Blue Cross and other insurers offered their services at what appeared to be no costs (even though the costs were built into the premiums and Administrative Service Agreements). This was another blow to CHMIS type projects.

So CHMIS projects funded by the Hartford Foundation folded, one by one, across the country ( e.g. Iowa, Seattle, Memphis and other locations). The Detroit based MHMIS had to learn from these experiences and was able to seek alternate directions.

All Stakeholders Working For Standardization in the Health Care Supply Chain
The Michigan Health Management Information System (MHMIS) was funded by annual assessments from employers (e.g. Auto Companies and others), Health Systems, Health Plans, Insurers and Medical Societies at about $200,000 per year. A loaned executive from Michigan Bell, with the assistance of executives from the Henry Ford Health System and Chrysler Motors developed the early program thrusts at the Detroit Coalition.

A big departure from the Hartford approach was to avoid a Common Data Repository and move towards what is called a “Distributive Data Base”.  This  means the focus is upon creating Common Connectivity among trading partners and using Common Standards and Systems for moving information from one entity to another. However, what data is shared, with whom and when, is left to each pair or group of trading partners. This avoided others having access to data. 

Thus, if Blue Cross wanted to share data with Henry Ford Health System, they could use the common connectivity system, with common technical standards for transmission and common content standards for jointly agreed upon applications/functions. This did occur in the Detroit Coalition with a larger group, where for example the auto companies stimulated a common system for display of health plan information to employees at enrollment. Most major employers and all health plans adopted this common web-based system.

Enter THE Health Care Interchange
Subsequently an executive from Detroit Edison was hired by the Detroit Coalition to further the MHMIS project. He, along with a former CIO from a major health plan (Select Care), and the President of the Detroit Coalition formed a model called THE Health Care Interchange (HCI). This took the previous success of MHMIS to a new level of understanding and opportunity. Figure 1. presents the Model.

Figure 1

The early deliverables of MHMIS emphasized joint community work on transactions such as enrollment, eligibility, referral, remittance and other functions that subsequently became part of HIPAA Administrative Simplification rules. So, Detroit was well on the way to trading partner agreements and standards when HIPAA came along in the late1990s.

About 1996, the MHMIS participants sought a common community-wide system for Connectivity that would replace the many, often duplicate, and costly hard wire (phone line) connections between trading partner entities. While the developing Internet was considered, there was concern about both reliability and confidentiality. So the group adopted, after much study, a highly secure, reliable and confidential Extranet (Private Net System that uses Internet protocol) that was developed by the auto companies and their suppliers. They system was called ANX (Automotive Network Exchange) as was developed with unparalleled reliability (i.e., doesn’t go down) and security. It is a global system, available in all parts of the word as well as Detroit and thus supported with unusual resources. The ANX network was adopted and renamed to HNX (Health Network Exchange) for use in health care.

This HNX@ANX network replaced many duplicative and costly hard wire (phone lines) with a common connectivity. Figures 2 and 3 demonstrate this efficiency.

Figure 2.

Figure 3

In the late 90’s the Detroit Regional Chamber of Commerce commissioned a study by a Healthcare IT company in Dearborn, MI, to determine the value of MHMIS and THE HCI approach.  The study concluded with an estimate that use of the first two elements in THE HCI model (Common Connectivity and Common Technical Standards For Transmission) would save $10 Million per 100,000 residents of a community (see figure 4.)

Figure 4

As shown in Figure 5, about this same time a Statewide study in New Jersey estimated savings of hundreds of millions of dollars for expanded use of EDI and related standards.

Figure 5

Common Applications
These projected savings did not even include opportunities in Common Content Standards (Applications) where further costs can be reduced and quality improved. An example is the “V-8” program initiated in Detroit where purchasers and health plans agreed upon a common RFI process each year in lieu of the differential approach of each employer hiring their separate consulting group and issuing slightly different RFIs to the health plans. Multiple, duplicative costs for consultants were reduced, health plans could reduce FTEs devoted to separate responses and quality of care could be improved with scaleable joint plan and purchaser standardization of specifications.

Another cost saving example is to establish a common credentialing system/application. Why have physicians go through the hoops of so many separate applications, duplicative costs for primary verification and the opportunity for quality degradation, due to the complexity. 

"Blown to Bits”
THE HCI Model was subsequently validated further by the work of Evans and Wurster who as executives with The Boston Consulting group authored their landmark  book: Blown to Bits.   A large section of the book deals with health care and even references the ANX. This book is highly recommended reading for those with more interest in THE HCI type of approach.  See figure 6.

Figure 6 - From "Blown to Bits"

In 2001, the MHMIS effort moved from the Detroit base to Lansing Michigan, the state capital, and became THE Health Care Interchange of Michigan.  It was embraced by the state hospital association, state health plan association and many other groups throughout the state.

Implications For Health  Policy 

·        Substantial costs and unnecessary duplication can be eliminated by a joint planning approach as presented above.

·        In most states there is a clear vacuum for planning direction such an approach.

·        There does not seem to be any other operable base for joint health care stakeholder collaboration on Common Connectivity, Data Standards and Content Standards (Applications).

·        A leadership group of  is required to bring diverse stakeholders( employers, health plans, providers etc.)  to the table for joint planning, priority setting, funding and implementation of  a similar project.