Resources

The openEHR advantage

Health care informatics is ripe for change

The open Electronic Health Record initiative of the openEHR Foundation - established by University College London and Ocean Informatics - has become an active agent of change in the critical domain of electronic health records. The Foundation now has growing influence in international EHR standards development and is generating enormous business opportunities by challenging traditional models of development and empowering patients and health-care professionals. openEHR attacks the heart of the semantic interoperability problem - the clinical model of modern health-care.

What is openEHR?

openEHR is a new way of structuring, storing and managing patient data so that it can be shared and exchanged between different healthcare providers and other stakeholders in a safe and secure manner. The goals governing the design of openEHR specifications are a comprehensive clinical model, maximal interoperability of maximal data sets to meet the needs of patients and their carers. The architecture of openEHR is derived from advances in two closely related areas: knowledge modelling of the clinical domain and information modelling designed to achieve openness and interoperability of resulting IT systems and applications. These two technological breakthroughs are a giant step towards providing a level-playing field for all participants, but first and foremost, it gives the patient a platform for controlling his or her own destiny through full access and control over personal medical data.

The idea of an electronic health record is one of the oldest in health informatics - so old and yet so elusive as a goal within the framework of traditional and still dominant approaches to software construction. Quite often, the flexibility, interoperability and openness are sacrificed to the rigid formats of proprietary solutions, databases, electronic messages - in the United States alone at one point there existed 440 data exchange formats in the field of health insurance and health claims processing.

For a number of years, international organisations such as the OMG, HL7 and CORBAmed in particular have supported the development of open standards and architectures. Interoperability is the key concern of adopted standards such as PIDS (Person Identification Service), COAS (Clinical Observation Access Service), LQS (Lexicon Query Service). Despite significant advances, the health IT landscape consists of unrelated islands separated by vast spaces of thematic 'wilderness'. We are still far from a "unified theory" of the medical field. openEHR provides the foundations for a general approach with regard to standardisation of personal health information.

The original version of openEHR was developed in Europe nearly a decade ago. The Good European Health Record project, supported through the AIM program of the EU, resulted in the formulation of key requirements for an electronic health record: clinical comprehensiveness, semantic sophistication, sharing, computability and implementability, open standards. The public policy, security, medico-legal (privacy, accountability), research and educational aspects formed its broader context. Dr. Sam Heard, one of the founders of Ocean Informatics, was a co-author of this European project. Its software architecture was systematised by Thomas Beale, now CTO of Ocean Informatics.

There have been a number of projects and organisations addressing the issues of the electronic health record. However, the revolutionary breakthrough, which enables openEHR to be fully implemented and considered as a future world standard for health records, has been made in Australia in 1998. This leading idea of Ocean Informatics is the adoption of a rigorous knowledge modelling approach, founded on a basic tenet: the separation of domain and technical concerns in information systems. In real terms this translates into:
  • The removal of domain concepts from concrete software and database models, into independently managed and standardised vocabularies and libraries of domain concept models.
  • Re-engineering the software and database using a generic reference model system architecture, designed to process information by using externally supplied domain definitions.
The new approach to building software is one in which the analysis and design models express not domain concepts, but informational concepts which are capable of describing any domain concept of a certain category. Standardised vocabularies, or structured dictionaries of standard term definitions, are a key element of a knowledge-based approach. Vocabulary development has existed in clinical medicine for years, and is finally beginning in other domains, as part of the general drive for B2B communications. Some vocabularies encode meaningful semantic relationships, such as classifications and associations between concepts, and it is preferable that such semantics are not replicated in domain models, but used by them.

openEHR Archetypes - a European and ISO standard

The knowledge modelling approach based on the systematic use of archetypes is the most revolutionary change in traditional software devel;opment methodology. The ability to structure a particular business domain through the use of an evolving set of archetypes, defined as constraint-based definitions of business concepts particular to that domain, has provided the Australian openEHR group with a powerful technology for designing working health information systems and applications. It makes full use of already established standards such as XML (XML-schema, XSLT and other W3C recommendations) to represent definitions and constraints that are both formalised and computable, understandable to humans as well as machine-readable.

The term archetype is used here for definitions of domain concepts, since it connotes "an original model, prototype or typical specimen." Archetypes serve various purposes:

  • To enable users in a domain such as healthcare to formally express their concepts, independent of IT experts. Archetypes are developed and change-managed by representatives of the health professions, user groups, conferences, standards working groups and any other forums available to practitioners, independent of and without reference to systems that will process them.
  • To enable information systems to guide and validate user input during the creation and modification of information at runtime, guaranteeing that all information "instances" conform to domain requirements.
  • To guarantee interoperability at the knowledge level, not just the data structure level.
  • To provide a well-defined basis for efficient querying of complex data.
  • To supply a powerful tool for data cleansing and data validation thus opening up massive legacy information resources for systematic processing, research and modelling.

A re-conception of information system engineering based on archetypes is the key to achieving widespread, knowledge-level interoperability. Under a methodology oriented toward interoperable knowledge, system developers/vendors need only agree on:

  • Much smaller software models, which describe generic domain concepts, rather than actual knowledge.
  • A technical means of interoperability, such as CORBA, Microsoft.NET, J2EE, or web-based distributed architectures.


Compared to the classical methodology, the size of models agreed by software developers in this approach is much smaller. The main difference is that it is the system users who define and agree on domain knowledge models consisting of archetypes built from standardised vocabularies and rules.

The result is systems built from small reference models, which process archetypes in a way semantically equivalent to hard-modelled systems. All information created not only conforms to the concrete models (enabling database storage and intersystem transmission), but also to the domain models described in archetypes, thus ensuring that all co-operating systems can understand the information at the highest possible level. In particular, this allows effective automated processing to occur, such as by decision support systems (DSS), since these systems can now make assumptions about the structure and content of the information in underlying systems, by inspecting the archetypes used to create it.

The methodology consisting of a technical and managerial approach to archetypes has been implemented and tested in the health domain by the Good Electronic Health Record project using object-oriented technologies in underlying systems and XML-schema documents and tools for archetypes.

Thus openEHR archetypes:

  • Define the structure and business rules of information used in health computing systems, i.e. they define concepts ranging from the simple, such as "blood pressure", to the most complex, such as "MRI scan results";
  • Are created and modified by clinicians rather than IT specialists;
  • Contain the validation rules for all data entered into the computing system, therefore surpassing the ability of current systems to prevent errors in health records;
  • Can be shared by multiple health systems and health authorities, enabling health record information to be shared between different systems, different types of healthcare professionals and even different countries;
  • Form a basis for intelligent automatic processing, such as decision support as well as epidemiological and other public health research functions;
  • Can be used as a roadmap for data, enabling faster querying;
  • Allow changes to be made to the system without touching the software or database hence eliminating a major cost in all current computer systems;
  • Ensure the system learns over time, increasingly satisfying user needs, rather than degrading into obsolescence, as invariably happens with today's health systems.

The openEHR Advantage

openEHR has strong advantages over current electronic and paper-based systems. It:
  • Improves the quality of clinical care, while significantly reducing IT costs of providers;
  • Integrates contributions by all clinicians, including GPs, specialists, hospital staff, allied health workers, pathology and radiology services, as well as patients themselves;
  • Provides a complete picture of a patient's health status enabling better care to be delivered;
  • Dramatically reduces the level of data errors compared to paper records and current openEHR systems;
  • Gives patients complete control over access and distribution of their health records;
  • Enables health records to be safely and securely communicated between healthcare organizations such as hospitals, GP clinics and community health providers;
  • Is closely linked to Australian and international health informatics standards;
  • Ensures that electronic health records are "future proof" through the use of a revolutionary new approach called "archetypes".
Other openEHR innovations include:
  • Version-controlled health records, enabling all past states of the health record to be investigated in the event of clinical or medico-legal investigations;
  • "Persistent" clinical information such as family history, current medications, vaccination record, and therapeutic precautions (allergies etc) is stored separately in the record, enabling fast lookup and querying of the most frequently used information.
Ognian Pishev
2006
OpenEHR logo