Since Version 1 was published in 1995, the National Health Information Model (NHIM) has been adopted as the underlying architecture for the National Health Data Dictionary (NHDD) and the Knowledgebase. The Knowledgebase is the electronic registry of national health and welfare metadata standards maintained by the Australian Institute of Health and Welfare (the Institute) on behalf of the National Health Information Management Group (NHIMG) and the National Community Services Information Management Group (NCSIMG).
The development of Version 2 of the Model commencing in 1997 marked a change from the entity-relationship model presented in Version 1 to a high-level, relationship-free, multi- business framework. From 1998 the NHIMG approved the use of the Model version 2, in draft form, as the organising structure for the NHDD and Knowledgebase. This paper is a formalisation of that approval.
This change recognised the NHIM's general acceptance as a high-level framework and the need for multi-layering of the modelling process. That is, it reflects the importance of consistent identification of entities at the national level, and the greater importance of relationships or business rules at lower levels.
The NHIM is an 'information model'-it is independent of process. In other words, it is not concerned with 'how' something happens, but rather with the information structure underlying the diverse processes and policies of healthcare delivery in Australia. By understanding the structure of health information resources, we are better able to exploit the information these resources contain. It is a 'conceptual model' aimed at establishing an agreed high-level structure. It thereby enables broad entities to be identified and described and provides a framework to develop more detailed subordinate models.
In 1998, the NCSIMG adopted a National Community Services Information Model (NCSIM) as the organising framework for the initial edition of the National Community Services Data Dictionary (NCSDD). This Model was based on Version 2 of the NHIM.
As a national framework, the NHIM enables related data elements from the NHDD to be grouped under a single entity rather than organised alphabetically. Entities are the things about which we need to know or hold data on. They may be people, places, objects, events or concepts. The Knowledgebase uses this aspect of the NHIM as the conceptual gateway to locate, identify and download data elements.
The 12 major super-entities of the NHIM can be loosely organised into four categories- Parties and states, State changing events, Environmental factors and Classifying systems. The coverage and importance of particular entities can be assessed by the population of an entity with data elements. For example, if there are few or no data elements for some model entities, this may be helpful in identifying areas for further development or in reassessing the structure of the NHIM. Version 2 of the NHIM contains more entities than Version 1, largely because of the development of several sector-specific contextual models such as the NCSIM, the Disability and Aged Care Model, the Primary and Community Health Services National Information Model, the National Institutional-based Ambulatory Care Model and the Community Health Information Model.
There is increasing interest in the use of the NHIM as the main tool for standardising health and welfare information in Australia. This includes its potential use in developing electronic data, designing information systems and as a framework for the consistent collection, storage and transmission of data. The next 3 to 5 years are expected to prove a watershed for the NHIM, with significant national health information development projects (including electronic health records) being actively pursued. There are a number of current developments that could see the NHIM applied as an overall model for context-specific models-the HL7 Reference Information Model and the Good Electronic Health Record (GEHR). It could also form the basis of other health information developments, such as Health Online and HealthConnect, and of the work of the National Electronic Health Records Task Force. The Model will have to continue to prove its worth and utility in these projects, and will need to continue to learn from and develop with them in order to remain at the forefront of this work.
The development of Version 2 represents a significant period of consolidation and maturity for the NHIM. It reflects the Model's progression from an initial concept and design to a more robust architecture. The likelihood is that pressure for its enhancement and development will continue at a more rapid pace in the near future. Greater alignment between the projects under the Health and Community Services Information Management Groups will highlight the need for common information structures. The Model could act as a tool for building consensus, assisting business planning, providing logical frameworks and influencing application development across human service sectors.
Although models can improve information resource use and management in many ways, they are not substitutes for sound data development practice and management. Equally, there is no single best model for health or for any business activity. The best conceptual models continue to be challenged and supported by contextual level models, while accommodating the technical and semantic diversity that generates them.