HL7 Development Framework Project

Project Charter

Health Level Seven

3300 Washtenaw, Suite 227, Ann Arbor, MI 48104-4261

February 22, 2002

1

Health Level Seven Development Framework Project Charter

Table of Contents

Introduction

Background

Health Level Seven

HL7 Version 3.0

Unified Modeling Language

Message Development Framework

HL7 Development Framework

Project Scope and Objectives

Project Scope

Project Objectives

Project Assumptions and Constraints

High-Level Project Plan

Project Phases, Activities, and Timeline

Major Project Deliverables

Allocated Project Resources

References

Introduction

The purpose of the Health Level Seven (HL7) Development Framework Project is to research, analyze, design, and document the processes, policies, and artifacts associated with development of HL7 standards specifications. The HL7 Development Framework (HDF) project will enable HL7 to:

  • Maximize the benefits it derives from using the Uniform Modeling Language (UML) as a foundation for its model-based approach to standards development;
  • Expand application of its modeled-based approach for standards development beyond messaging to its other standards such as structured documents, context management, and standards related to electronic health records;
  • Facilitate the participation of HL7 members, subject matter experts, and implementers in the development of HL7 standards.

This document is the HDF project charter. It provides an overview of the HDF project including the project background, scope and objectives, and high-level project plan. The purpose of this document is to serve as a communication tool for the project team, project sponsors, participants, stakeholders, and interested parties. The primary goal of this document is to ensure that there is a common understanding of the scope, objectives, and work plan for the HDF project among all project participants and interested parties. It will be updated as needed throughout the life of the project to reflect any significant changes in the project objectives, scope, or plans.

The HDF project is sponsored byHealth Level Seven, Inc. with partial funding from the Veterans Health Administration (VHA). Project management is provided by Shakir Consulting. For additional information regarding the HDF project contact:

Mark McDougal, Executive Director
Health Level Seven, Inc.
3300 Washtenaw Ave, Suite 227
Ann Arbor, MI 48104-4261

Background

Health Level Seven

Health Level Seven (HL7) is an American National Standards Institute (ANSI) Accredited Standards Developer. The mission of HL7 is to provide a comprehensive framework and related standards for the exchange, integration, storage, and retrieval of health information that support clinical practices and the management, delivery and evaluation of health services. The mission of HL7 encompasses the complete ‘life cycle’ of a standard’s specification – development, adoption, market recognition, utilization and adherence.

HL7 is a not-for-profit volunteer organization. Its 2200 members represent over 500 corporate members, including 90 percent of the largest information system vendors serving healthcare. HL7 is an international organization with affiliates in nearly 20 countries, representing six of the seven continents (Africa, Asia, Australia, Europe, North America, and South America).

HL7 began developing standards in 1987 with the publication of its messaging specification - the Application Protocol for Electronic Data Exchange in Healthcare Environments. The standard provides the layout of messages that are exchanged between two or more applications based upon a particular trigger event. It specifies which data elements are to be sent, the data type and suggested length of each element, and indicates whether the data element is required or optional and whether it may repeat.

In the years since its founding, HL7 has evolved beyond traditional messaging protocols. Today HL7 standards development initiatives include:

  • standardization of knowledge representation (Arden Syntax);
  • specification of components for context management (known as CCOW);
  • support for healthcare data interchange using object request brokers;
  • standardization of XML document structures;
  • specification of robust vocabulary definitions for use in clinical messages and documents; and
  • work in the area of security, privacy, confidentiality, and accountability.

In 2001 HL7 modified it mission statement to specifically include standards, guidelines, methodologies, and related services for the management of electronic health records.

HL7 Version 3.0

In 1992 HL7 made a fundamental shift in the methodology it uses to develop its standards specifications. The new methodology, referred to as HL7 Version 3.0 or V3, is a model-driven methodology based upon modern object-oriented software development practices. HL7 spent four years creating the methodology that adapts modern analysis techniques from system building to message design.

Initially the HL7 Executive Committee chartered an independent task force to establish the broad outline of the approach. In January 1996, the Technical Steering Committee agreed to adopt the main features of the approach and take over its management. Planning work continued in the Modeling and Methodology Technical Committee. In the spring of 1997, all the HL7 Technical Committees began to use the V3 process.

HL7 version 3.0 is the most definitive HL7 standard thus far, incorporating more trigger events and message formats than any previous version. It uses a Reference Information Model (RIM) as a commons source for specification of the information content of message specifications. The RIM is an essential part of the HL7 Version 3.0 development methodology. It provides an explicit representation of the semantic and lexical connections that exist between the information carried in the fields of HL7 messages.

As part of version 3.0, the HL7 Vocabulary Technical Committee is developing methods that allow HL7 messages to draw upon codes and vocabularies from a variety of sources. The V3 vocabulary work assures that the systems sending and receiving V3 HL7 messages have an unambiguous understanding of the code sources and code value domains they are using. HL7’s primary goal for version 3.0 is to offer a standard that is definite and testable, and to provide certification of vendor’s conformance.

Balloting on an initial set of V3 message specifications began in August 2001. This first set of V3 messages provides comprehensive coverage of patient administration, laboratory and pharmacy orders, medical records management, and initial proposals for scheduling. A new set of electronic claims messages support a contingent of HL7 international affiliates now working on creating financial transactions to support an e-claims submission process.

The HL7 Version 3.0 development methodology is a continuously evolving process that seeks to develop specifications that facilitate interoperability between healthcare systems. The HL7 RIM, vocabulary specifications, and model-driven process of analysis and design combine to make HL7 Version 3.0 an exemplary methodology for development of consensus-based standards for healthcare information system interoperability.

Unified Modeling Language

The models used in the HL7 V3 process are based upon the Unified Modeling Language (UML). The UML is a graphical language for visualizing,specifying, constructing, and documenting the artifacts of a software-intensive system. The UML is an Object Management Group standard that represents the unification of best practices in practical object-orientedmodeling. The Object Management Group (OMG) is an open-membership, not-for-profit consortium founded in 1989 that produces and maintains computer industry specifications for interoperable enterprise applications.

Development of the UML began in 1994 when James Rumbaugh and Grady Booch of Rational Software Corporation began combining the concepts from the Object Modeling Technique (OMT) and Booch methods, resulting in a unified specification in 1995. In the Fall of1995, Ivar Jacobson joined Rational and the unification effort, merging in the Object-Oriented Software Engineering method (OOSE). The joint work of Rumbaugh, Booch, and Jacobson was called the Unified Modeling Language (UML).

In 1996, the Object Management Group (OMG) issued a request for proposals for a standard approach to object-oriented modeling. The UML authors collaborated with methodologists and developers from other companies to produce the final UML proposal that was submitted to OMG in September 1997. The final UML specification is a product of collaboration among many people and companies. The Unified Modeling Language was adopted unanimously by the membership of the OMG as a standard in November 1997. The OMG assumed responsibility for the further development of the UML standard.

The current version of the UML specification (v1.14) was published in September 2001. Under the stewardship of the OMG, the UML has emerged as the software industry’s dominantmodeling language. It has been successfully applied to a wide range of domains, ranging fromhealth and finance to aerospace. Its extensive use hasraised numerous application and implementation issues by modelers and vendors. Over 500 formal usage and implementation issues have been submitted to the OMGfor consideration. Many of the issues have been resolved in minor revisions by the OMG UML Revision Task Force. Other issues require major changes to the language;consequently,the OMG has issued RFPs to defineUML 2.0.

The UML is part of OMG’s catalog of modeling specifications. Other OMG modeling specifications with relevance to the HDF project include the XML Meta-Data Interchange (XMI) and the Meta-Object Facility (MOF) specifications.

XMI is a specification enabling easy interchange of metadata between modeling tools and metadata repositories in distributed heterogeneous environments. The OMG XMI specification leverages XML(eXtensible Markup Language) technology to support the interchange of meta-data andmeta-models between MOF-based meta-data repositories.

The MOF is a set of standard interfaces used to define and manipulate a set of interoperable metamodels and their corresponding models. Itprovides a framework that supports the management of metadata. The MOF is based upon a four layer meta-modeling architecture. The key feature ofthis architecture is a meta-meta-modeling layer that provides a common language thatties together the meta-models and models.

The MOF provides a basis for defining future UML metamodel extensions. It allows dynamic extensions of UML models through profiles and facilitates aligning the UML metamodel with other standards based on the same metamodeling principles. The following figure, extracted from the UML 2.0 Proposal, is an example of the four-layer architecture on which the MOF is based:

Message Development Framework

The HL7 Message Development Framework (MDF) defines the HL7 V3 model-based process of message development. It identifies the phases, activities, and models used in the process of developing HL7 message specifications. It also includes a meta-model specification of the Reference Model Repository (RMR) used to store and manage the artifacts produced during the message development process. The following figure, extracted from the MDF, provides a graphical depiction of the message development process and artifacts:

The HL7 MDF was first published in 1997. It has undergone two major revisions since then; once in 1998 and again in 1999. The current version of the MDF (v3.3), published in December 1999, has not been maintained and is consequently out of alignment with current message development practices. Enhancements to the methodology have been captured in HL7 committee meeting minutes, tutorials, and tooling documentation but have yet to be compiled into a new release of the MDF.

An abbreviated rendition of the methodology is included in the V3 messaging ballots under the heading “Version 3.0 Guide”. The V3 Guide is not as extensive in its coverage of the methodology as is the MDF, but it does reflect recent enhancements to the methodology of importance to reviewers of the ballot. The HL7 Message Development Framework will be superceded by the HL7 Development Framework, the HDF.

HL7 Development Framework

The HL7 Development Framework (HDF) is a replacement for and an extension to the HL7 Message Development Framework (MDF). It is the primary deliverable of this project. The HDF differs from the MDF in terms of:

  • Scope of coverage;
  • Alignment with UML; and
  • Maintenance/versioning procedures.
Scope of Coverage

The HDF documents the processes, tools, actors, rules, and artifacts relevant to development of all HL7 standard specifications, not just messaging. Initially, the HDF will address messaging, structured documents, and context management. Eventually, the HDF will encompass all of the HL7 standard specifications, including any new standards resulting from analysis of electronic health record architectures and requirements.

Alignment with UML

The metamodel of the artifacts to be produced as part of the development processes documented in the HDF will be UML conformant. The MDF contains exceptions to the UML specification. Most of the exceptions are minor; however a few are major exceptions. Expediency and the lack of adequate commercially available tools have given rise to HL7 adopting modeling conventions that are somewhat askew of UML.

The inconsistencies between the HL7 metamodel and the UML metamodel will be analyzed and resolved. Resolution of the inconsistencies may result in updates to the HL7 metamodel, formal submission of issues from HL7 to OMG regarding the UML specification, and the development of an HL7 UML Profile that leverages the UML extension capabilities. The final UML-compliant HL7 metamodel will be documented in the HDF.

Maintenance/versioning Procedures

In addition to documenting the processes for development of HL7 standard specifications, the HDF will also document the policies and procedures that govern the enhancement and version management of those processes. The HDF is to be a “living document”, capable of keeping abreast of continuous improvements to the methodology in a systematicand disciplined fashion. Proven techniques in document management, configuration management, and version control will be applied to the HDF.

Project Scope and Objectives

Project Scope

The scope of the Health Level Seven Development Framework project is to:

  • Develop and publish the HDF Metamodel;

This activity includes the analysis and comparison of the HL7 HDF metamodel to the OMG UML metamodel and the resolution of any and all discrepancies between the two metamodels. Resolution of the inconsistencies may result in updates to the HL7 metamodel, formal submission of issues from HL7 to OMG regarding the UML specification, and the development of an HL7 UML Profile that leverages the UML extension capabilities.

  • Develop a final draft of the initial version of the HDF;

This activity includes the research, analysis, design, and documentation of processes, policies, and artifacts associated with development of HL7 standards specifications for messaging, structured documents, and context management. This includes the publication of a first and final draft of this initial version of the HDF.

  • Develop a first draft of the HDF Practical Guide.

This activity produces the first draft of a practical guide to the HL7 development methodology documented in the HDF. The HDF Practical Guide differs from the HDF in that it is focused less on the methodology and more on providing practical guidance related to applying the methodology to the day-to-day activities of the standards development within HL7.

Project Objectives

The objectives of the HDF project are to maximize the benefits HL7 derives from using UML as a foundation for its model-based approach to standards development; expand the modeled-based approach for standards development beyond the HL7 messaging standard; and facilitate the participation of HL7 members, subject matter experts, and implementers in the development of HL7 standards. This project enables HL7 to remain the industry leader in model-driven development of comprehensive standards for application interoperability in the Health industry.

Project Assumptions and Constraints

The success of this project is dependent upon the existence of critical success factors, the validity of planning assumptions, and an awareness of project constraints. The following is a list of the critical success factors, planning assumptions, and constraints for this project.

  1. The HL7 MDF metamodel v1.14 and the OMG UML metamodel v1.4 are to be used as the starting point for metamodel comparison and development of the HDF metamodel.
  2. Input from preliminary drafts of the OMG UML metamodel v2.0, the HL7 V3 Guide, and proposed enhancements to the vocabulary portion of the HL7 MDF metamodel will be used as input to the metamodel comparison and development of the HDF metamodel.
  3. The OMG MOF will be used as the architectural framework for expressing the relationships between models, meta-models, meta-meta-models, and model instances.
  4. Specific areas of the OMG UML metamodel to be considered in the HDF metamodel include the UML semantics for Packaging, Stereotypes, Association roles, and Object Constraint Language.
  5. An HL7 UML Profile will be developed to document extensions to the OMG UML metamodel as one means of resolving discrepancies between the final HL7 HDF Metamodel and the OMG UML Metamodel.
  6. Some discrepancies between the final HL7 HDF metamodel and the OMG UML metamodel will remained unresolved by the HL7 UML Profile. These discrepancies may be submitted by HL7 to OMG as issues related to the OMG UML specification.
  7. HL7 will need to weigh the costs and benefits associated with modifications to the existing HL7 MDF metamodel in order to align it with the OMG UML metamodel. HL7 may choose to retain a discrepancy from the OMG UML metamodel in the final HDF metamodel if in the view of HL7 the cost of resolving the discrepancy outweighs the benefits.
  8. Positioning itself to use XMI as a means of interchanging metamodels between modeling tools and metadata repositories is viewed by HL7 as a major benefit, possibly providing sufficient motivation for making modifications to the HL7 MDF metamodel, even if the modificationscarrysignificant costs.
  9. HL7 will seek to protect and leverage its current investment in development and publishing tools, skilled personnel, and intellectual capital.
  10. Rationale will be documented for each unresolved discrepancy between the final HL7 HDF Metamodel and the OMG UML Metamodel.
  11. Representatives from the Structured Documents and CCOW Technical Committees will be made available to function as subject matter experts in the development of the HDF.
  12. A sufficient number of qualified HL7 volunteers will be available to contribute material, participate in discussions, and review and approve document drafts.
  13. An expert in the OMG UML specification and metamodel will be available as a resource to the project team.
  14. The HDF will address weaknesses in the current message development methodology such as the specification of constraints, documentation of requirements, and reuse of specification patterns.
  15. The HDF will define processes for maintenance and use of shared work products such as the RIM, CMETS, and Vocabularies domains across committee boundaries.
  16. The HDF will clarify the methodological implications related to the roles and interrelationships among different categories of information models such as the Reference Information Model (RIM), Domain Information Model (DIM), and Refined Message Information Model (R-MIM).
  17. The HDF will define an approach for differentiating among the multiple levels of abstractions within the structural, semantic, and behavioral aspects of HL7 modeling activities.
  18. The development of the HDF will involve multiple contributors, thought leaders, and content editors. To expedite the process it will be necessary for those involved to meet outside the three scheduled HL7 working group meetings.
  19. One or more face to face meetings of all project participants will be required. Face to face meetings shall be scheduled to coincide with interim meetings of the MnM TC, such as the RIM harmonization meetings.
  20. The MnM Technical Committee shall have primary responsibility and accountability for the HDF project. The HL7 Technical Steering Committee shall be the primary governing body providing oversight and issue resolution for the HDF project.
  21. The time period for conducting the HDF project shall be from January 2001 through September 2001. Additional work will be required beyond September, however, that work will be conducted under the auspices of a separate project.

High-Level Project Plan