Report on ISO/TC215, Health Informatics Technical Committee of the International Organisation of Standards (ISO).

Orlando, Florida, USA, 25-26 August 1998

Michael Mair FRACO, FRCOphth, New Zealand Delegate

Contents:


Explanation of my involvement

I originally volunteered for the Standards New Zealand WG3 on Electronic Medical Records, which has met intermittently since 1995 (see appendix two for a summary of this committee's work), representing the New Zealand Association of Salaried Medical Specialists. I was active in getting the final 'data model' structure which is displayed on the NZHIS Web site, and which is based on the deliberations of the group and discussion with Ashwin Patel of the New Zealand College of General Practitioners.

Working group members were invited to take part in a teleconference with their Australian counterparts in Wellington in July, in preparation for ISO/TC215, the International Organization for Standard's Technical Committee on interoperability for Electronic Medical Records. I assisted in formulating the New Zealand submission to this group. New Zealand is one of 19 nations that have 'P' class or voting membership of this technical committee. I volunteered to attend the meeting in Florida when it became clear that there might otherwise not be representation.

I have been instrumental in developing an Ophthalmic Management System with a New Zealand-based software company, and have experience in the theory and design of electronic medical records. The 'meta model' of the 'Medical Process' which is stated on the NZHIS Web site is important to me, particularly the idea of the 'surface structure' of actual applications being at a different conceptual level from the 'deep structure' of the items (or objects) from a shared data dictionary, and shared protocols for defining and exchanging them. It is important to me for philosophical and practical reasons as an independent software designer that these notions 'get through' both nationally and internationally. I shall discuss at the end of this report the extent to which I think they have done so.


Top of page

Introduction

ISO TC/215 had a leisurely genesis. Originally an initiative of CEN (the European Standards Organization) to build bridges in the health informatics area with North America, the matter was considered at a meeting in London, UK, in June 1997. A new ISO technical committee to address healthcare standards was set up, and the secretariat given to the US organisation, the American National Standards Institute (ANSI). There was dispute between representatives of the Centre Europeanne de Normalisations (The European Standards group 'CEN') and the Americans about where the Chairman should be from (US or Europe). Subsequently, provisional chairmanship of this committee was given to Australia, a compromise solution that both sides could live with. Peter Tresseder's chairmanship was confirmed at the meeting. The brief 'mission statement' of the scope of the technical committee was:

Standardisation in the field of information for health, and health information and Information Communications Technology (ICT) to achieve compatibility and interoperability between independent systems.

Among riders to this statement proposed by contributors were 'to optimize the use of limited resources available worldwide for standards development through collaboration' (Canada) and 'to improve the healthcare of the public' (from the Finnish delegate).

A further meeting was arranged in Bermuda in September 1997, and at this there was extensive discussion of the areas where it was felt that standards should be developed. The virtues of few rather than many topic areas were recognised, and in the end just four Working Groups have been set up, with two additional 'ad hoc' working groups. There was an additional meeting in Kyoto, Japan, in December, all preparatory to the first meeting in Orlando which I report on here. All this meant that by the time of the first meeting there was a well worked out agenda, which just rolled through the one and one half days of meetings. The real work had gone on before, and would go on after, but 'being there' was still important.


Top of page

The Meeting Itself

General Remarks:

In his opening address, Peter Treseder, the acting Chairman of the committee mentioned the sense of urgency with which he considered should accompany our deliberations. He considered that our activities would 'facilitate the global society', and 'integration and harmonisation of healthcare resources between nations'. As he put it, 'if we do not get those standards out soon, someone else will do it for us', a reference to burgeoning private sector de facto activity.

He suggested that we should be aiming for a 'single object terminology' for healthcare communications, the integration of messaging standards (eg, HL7 and Edifact), and the standardised usage of 'mark up' languages (eg, SGML, XML). The goal was interoperability of computer systems globally, with a unique person identifier also applied globally. He even talked about a global 'business plan' for health. He stressed consumer access to high quality knowledge bases as a desirable byproduct of a global standard for health informatics, and hinted that 'financial entitlement to services' was something that might be established. His vision was quite utopian, appropriate perhaps in such a group.

In the opening address from the ISO representative from Geneva, Keith Brannon, gave the time scale for the full implementation for the new Global Standard, of 36 months.

The tone of the conference was remarkable for its informality. This was a group that had met before, and would meet again on the conference circuit. They related like old friends (or combatants), and all 'decisions' were taken by consensus. The actual technique by which the meeting progressed was more by an absence of marked dissent rather than proof of consensus. There was only one issue where there was overt conflict, and even then a vote was avoided. The implication was that real voting might be divisive and therefore shouldn't happen.

The mood was a little like that of a senior staff meeting at a large provincial hospital. There was quite a lot of fraternisation, yet contradictory sentiments could be expressed.



Peter Treseder - Chairman, Theresa Cendrowska - Secretary, Keith Brannon - ISO representive, and Gary Kushnier - ANSI representive, at the closing session (R to L)
Top of page

The Resolutions

I will summarise the proceedings by first listing the Resolutions which came out of them.
Resolution 1.
Was an administrative 'thank you' to the drafting committee
Resolution 2.
Repeated the 'scope' statement:
"Standardisation in the field of information for health and health Information and Communications Technology (ICT) to achieve compatibility and interoperability between independent systems. Also, to ensure compatibility of data for comparative statistical purposes (eg, classifications), and to reduce duplication of effort and redundancies."
Resolution 3.
Resolved that ISO/TC215 agrees that it is not the intent of the scope to:
  • Standardise the clinical practice of medicine
  • Define a standardised health care delivery service structure
  • Standardise medical knowledge, although the representation and exchange of knowledge is within the scope of ISO/TC215
  • Standardise the performance of healthcare, although the definition of standardised comparative performance data is within the scope of ISO/TC215
  • Standardise the internal operation of systems and devices, although the standardisation of data structure and the data output from the systems and devices is within the scope of ISO/TC215.
This resolution was mainly proposed by the Japanese delegation, and is a necessary disclaimer for a 'would be' standards group. In firmly dissociating ISO/TC215 from all such prescriptive activities this resolution will greatly reassure sensibilities both inside and outside the healthcare industry.
Resolution 4.
Resolved that ISO/TC215 establish an ad hoc group chaired by Brazil to identify the need for a biomedical imaging working group; and be it further resolved that the ad hoc group contact DICOM regarding their participation and report the outcome at the next meeting of ISO/TC215.
'DICOM' is the acronym for 'Digital Imaging and Communications in Medicine', an imaging standard evolved by the American College of radiology and the national Electrical Manufacturers Association. This organisation is apparently unwilling to take on the formal mantle of an international standard, and the purpose of the ad hoc committee is to persuade them to coordinate their activities with ISO/TC215.
Resolution 5.
Resolved that ISO/TC215 establish an ad hoc working group chaired by Germany to identify if there is a need for international standardisation of health cards. This resolution evoked little discussion.
Resolution 6.
Resolved that ISO/TC215 establish a working group to address modelling coordination and health records, and this was to be chaired by Australia.
Resolutions 7.
Resolved that ISO/TC215 establish a working group to address messaging and communication, and this was to be chaired by the USA.
Resolution 8.
Resolved that ISO/TC215 establish a working group on health concept representation, and this was to be chaired by the UK.
Resolution 9.
Resolved that ISO/TC215 establish a working group on security, and this was to be chaired by Sweden, but with an 'appointed secretary' from Japan.

As each working group was announced, volunteers were asked to indicate their willingness to nominate their member countries for the committees. The first discussions took place on that afternoon. I volunteered New Zealand for the modelling coordination / health record group. Although we might have expertise in the other areas, they all appeared to me to have heavy technical investments by larger nations. Modelling and Health Record architecture, however, were matters we had concerned ourselves with in Working Group three, and it was this area that the New Zealand Submission to ISO/TC215 had addressed. We have strong representation in medical informatics in actual software design, and we have had some success in the international market place. Design is a modelling exercise. Finally, the architecture of the record is the 'main part' of it, the kernel which all other aspects facilitate.

Each member nation must nominate candidates for the work group for which it has volunteered. It was expected that these members would be confirmed within 6 weeks, and the working groups to be fully established by 7 November, 1998.


Top of page

Modelling Working Group Meeting

I attended this discussion. Englishman Stephen Kay originally chaired this group. He has a position within the CEN 251 committee on health records. Discussion was very active, with several participants doubting whether any generic Health Record model was possible at all. The Japanese members of the group stressed how each institution and even each clinic within an institution might have its own record conventions. Christopher Chute from the Mayo Clinic pointed out that their concept of 'Encounter' already comprised a substantial volume. I argued for 'bottom up' generic record structure for the standard, with shared objects from a shared data dictionary being 'harvested' in the clinical encounter. These could be exchanged 'as is', and reassembled into the protocols that the end user was familiar with. The Mayo Clinic representative and the delegate from Germany felt this was impractical, and that some specialty specific information or structures needed to be 'in the standard'. This point was not resolved, and indeed the meeting was too brief to resolve any issues, being rather an open-ended exploration of the scope of the work before it.

Whereas some of the American delegates appeared to think that one valid task of the working group on modelling might be to establish a 'taxonomy' of models, some of us argued against this, suggesting that such a taxonomy could never be complete, since the limit to the models that could be made was the limit to ingenuity itself. The electronic medium itself hugely increases the modelling options in a way which is only just becoming visible. The head of the American delegation, Peter Wagerman, appeared to support this position. Some delegates, including myself, emphasised the primacy of the 'encounter' or 'transaction' as a unit of healthcare which was universal. This is encapsulated in the New Zealand model. Peter Williams from Australia usefully provided some Australian sourced definitions about 'the encounter' which were included in the summary of the first meeting of the working group.

The conclusions as summarised by Stephen Kay included statements about the fusion of health record architecture and modelling and coordination concerns. We were to model the 'total health process applied to one subject of care'.'Our conclusions', he said, 'said everything and said nothing'. One had rather the feeling that he thought the discussion had been trivial compared with the level he was working at with CEN 251.

We had, he said, agreed to use existing modelling ventures rather than develop our own. All record components should be testable. Part of the brief of the working group would be to facilitate and enable decision support. Subsequently there was a rebellion against his chairman role on this group, initiated by Brazil and South Africa and Canada. I suspect it was felt that his association with CEN 251 would mean that the ISO/TC215 activity would be a subset of the CEN group rather than the other way round. In his place was substituted Peter Williams of Australia, and again one felt that an Australasian presence in this crucial working group was acceptable partly because they/we are 'compromise candidates'. Culturally and geographically we are neither European or North American. Southern Hemisphere leadership of a working group was also felt desirable, and the UK already had the convener role in the working group on 'content'.

The deliberations of the other working groups were briefly summarised, but I can only pick out a few points in this account.

On messaging: it appears that the US-based HL7 organisation might work more closely with the European (and UN) Edifact communication standard. The 'mark up language' XML is widely assumed to be going to replace earlier versions, and XML can apparently be used by both HL7 and Edifact as the 'interchange format'.

On content: Standardised terms and their synonym are needed which record patient findings, circumstances, events and interventions with sufficient detail to support clinical care, decision support, outcomes, research and quality improvement and can be efficiently mapped. There are huge volumes of work done on this, from coding systems such as the ICD-9 and ICD-10, Read codes, higher-level interoperability projects such as Galen, and more as discussed in detail in the International Journal of Medical Informatics.

On security: there was significant conflict between the Japanese and Gunnar Stein from Sweden, who insisted on convening the Security group, and tried to have a formal vote to choose between himself and the Japanese-designated working group leader. It was only resolved by the separation of the role of Convener and 'Rapporteur', which went to the Japanese.

It was emphasised that each working group would need to clearly identify its relationship with 'health records' which is perhaps the most central shared concept, arguably the central subject matter of the ISO/TC215.


Top of page

Discussion

I have to admit that I felt only semi-informed at this conference. There were a lot of issues to be known about, and I was joining an argument that was already halfway through. I asked a lot questions at the interpersonal level, which must have considerably betrayed my ignorance, and one of the German delegates advised me to read a publication which would, he said, generally introduce me to the issues and where the arguments were up to. This turned out to be the special issue of the 'International Journal of Medical Informatics' devoted to Standardisation in Health Informatics. It was the conference report on the Bermuda conference, and I requested it from the Editor, G. De Moor of Belgium, who was also at the conference, and who promptly sent it. It has been most helpful.

I would like to focus the discussion on two topic areas:

  1. Our relationship to the Australian effort.
  2. What lessons I took home about our national effort, our data model, and what we should do now. I will also comment on the Good European Health Record Initiative (GEHR).

I use the terms 'Australasian' and 'our' with some audacity because there is not at present established an Australasian position. There are several indicators however that one could develop.

  1. There already is considerable interchange of personnel and ideas between the two countries. Paul Cohen regularly visits the Australian committee meetings, especially TC 14. Peter Williams repeatedly said that many of the Australian ideas were in fact Kiwi sourced (although I have been unable to find out which ones he specifically means).
  2. In my discussions at the conference, particularly with Robert Mayes, of the US Health Funding Authority (HCFA) it became clear that there was a role for New Zealand as a 'test bed' for implementation of ideas, but only in so far as were part of an Australasian scenario. He asked about our relationship with the Australians. I said we would go our own way if we had to, but if we don't, we won't. He remarked that reasons of 'hard cash' would also persuade us to work closely with the Australians. It is of interest that some contributors to the Bermuda meeting which consider that the evolution of a global health record architecture too big a task for any one nation alone. We could connect in to serious funding which would help the development of our health sector, but only if we were Australasian.
  3. The Australian view may not be compatible to our own in all respects, at this time of rapid evolution of views and opinions, we could make it compatible if there was the commitment on both sides. They already say that we seed them with ideas, and it is true that at the application level we have products that are as good or better than anything they have 'over there'. We might influence their position, and in turn be compliant with theirs in some respects. This is indeed a time for cooperation, for 'going beyond' where we are to where we might be. In fact I think our 'Conceptual Data Model' is simpler and better than the 'Information Model' displayed on their Web site.
  4. When the Americans look to the Australians for an 'economy of scale', so the Australians might look here. 'Test bed New Zealand' might benefit from this, but it would need to be carefully monitored. In particular:

Robert Mayes said 'we do not want to re-invent the wheel'. The scale and complexity of the US health sector makes it very difficult for them to innovate or implement any programme. Social Laboratory Australia (and New Zealand) might be a 'test bed' for new healthcare information technologies, and his agency might be able to contribute funding to such an enterprise for that reason. Mr. Hayes was very interested in 'meta models', and I felt there was the potential there for interest in the New Zealand model.

Support for our model comes from the GEHR project (The Good European Health Record). Widely thought of as the 'active arm' of CEN 251, GEHR is an actual application using CEN 251-derived definitions. It has been actively taken up by a UK company called Trident/Beacon, which has recently had a 50 percent capital injection from a consortium of pharmaceutical companies with a programme to develop a 'decision support' knowledge base for medical practice. The 'Paradigm' group of Hawke's Bay GPs (IPA) has employed them, and Trident plans to use that base as a springboard for the South Pacific Market. They claim to be implementing concepts similar to the New Zealand model already, including an explicit disavowal of structure other than their hierarchy of 'transaction', 'item'. The transaction is a dated 'item cluster', and there would appear to be a conceptual identity between the concept of 'transaction' and that of 'encounter'. Peter Schloeffel from Trident Australia expressed that view in a recent presentation (Christchurch, 28 August, 'Preparing for the Next Millennium') He agreed that the higher-level concepts 'Episode', 'Disease Course' and 'Medical Process' might well complement the GEHR concepts.

Trident/Beacon may go ahead rapidly in the Australasian market. We do not yet know if they are delivering more than promises but will assuredly soon find out. They now have linking to significant funding from the pharmaceutical industry, and this says oceans about the need for interoperability and decision support. It is not a question of whether it is a good idea or not, the reality is here and being marketed with some urgency. Hopefully, if standards developed through ISO/TC215 do supervene internationally, the experience with GEHR of organisations such as Trident can be used, and an all-embracing standard is something that even GEHR can co-evolve into. The future is a big enough place to allow such accommodation. The GEHR/Trident experience should be seen as an interesting exercise in implementation of a generic health record architecture, but it may not be the 'end of the story'. In fact, they may not be using GEHR architecture at all at present in the interoperability between practices that they are supplying in Hawke's Bay, but perhaps this area is commercially sensitive...


Top of page

Where to from now?

In many ways New Zealand could play a pivotal role in the global evolution of healthcare standards and their implementation. Fortunately we happened on a 'right tack' in our 'conceptual data model', and we go beyond both the Australian model and GEHR in our concepts of the Episode, the Disease Course, and the Medical Process as being hierarchically above the 'Encounter' or 'Transaction'. If we go for an identity with the Australian effort, we could have significant input into TC 215, which with Australian convenership of the Health Records/Monitoring sub-committee is open to our influence. Meanwhile it seems our own health sector is more ready for the implementation of a standard than most countries, and our smallness of scale and relative simplicity makes us a natural site for such a 'pilot' implementation. We might attract resources from larger nations, which would then become 'stakeholders' in our healthcare program.

There can be little doubt as a Japanese document has said that 'the healthcare information system is the key to improving the quality and efficiency of healthcare' (referenced in Toyoda, p 57; Ref 1). More than that, the structure of the information system mirrors and facilitates 'what is done' in the health sector, and stands in a dialectical relationship with it. We have a sense now that if we attempt to implement such a global system for New Zealand, we would by that act be contributing very significantly to the evolution of global standards. Ultimately the kind of 'business plan' for global health that Peter Treseder mentioned in his opening address might be the scenario for which we might strive. However, that is a distant goal, and in the foreground is the quest for shared standards and interoperability, which might be achieved first in New Zealand with the help of investment by overseas concerns such as the Australian effort, and the US UCFA.

New Zealand needs to participate in this process because otherwise we will be forced simply to adapt and interpret whatever standards are evolved. We have a contribution to make because our health sector is technically advanced, our 'best of breed' applications are definitely viable products in the international arena, and the future of our medical software industry will depend on building in ISO standards as they emerge. We are committed to membership of the 'Modelling' Working Group. Our status as 'P' (voting) members of the ISO technical committee lapses if we do not in fact vote on every document produced, so we must be represented.

The candidates that New Zealand puts forward must be identified promptly, and that one or more members might also expect to travel to Berlin in April 1999, and so plans must be made and commitments given.


Top of page

Reference and appendices


Top of page

Appendix 1

ISO/TC215 Resolutions:

  1. RESOLUTION
    Resolved that ISO/TC215 thank the Drafting Committee, Christine Michel, Sharon Stanford, and Patricia Village.
  2. RESOLUTION
    Resolved that the title and scope of ISO/TC215 shall be:
    ISO/TC215 for Health Informatics Scope:
    Standardisation in the field of information for health, and Health Information and Communications Technology (ICT) to achieve compatibility and interoperability between independent systems. Also, to ensure compatibility of data for comparative statistical purposes (eg, classifications), and to reduce duplication of effort and redundancies.
    And recommends that the Secretariat submits to the ISO Technical Management Board for approval.
  3. RESOLUTION
    Resolved that ISO/TC215 agrees that it is not the intent of the ISO/TC215 scope to:
    • Standardise the clinical practice of medicine
    • Define a standardised health care delivery service structure
    • Standardise medical knowledge, although the representation and exchange of knowledge is within the scope of ISO/TC 215
    • Standardise the performance of healthcare, although the definition of standardised comparative performance data is within the scope of ISO/TC215
    • Standardise the internal operation of systems and devices, although the standardisation of data structure and the data output from systems and devices is within the scope of ISO/TC215.
  4. RESOLUTION
    Resolved that ISO/TC215 establish an ad hoc group chaired by Brazil to identify the need for a biomedical imaging working group; and be it further Resolved that the ad hoc group contact DICOM regarding their participation and report the outcome at the next meeting of ISO/TC215.
  5. RESOLUTION
    Resolved that ISO/TC215 establish an ad hoc working group chaired by Germany to identify if there is a need for international standardisation of health cards and be it further Resolved that the ad hoc group report the outcome at the next meeting of ISO/TC215.
  6. RESOLUTION (?)
    Resolved that ISO/TC215 establish a working group to address modelling coordination and health records.

    Title
    Working Group 1 - Health Records and Modelling Coordination

    Scope
    To develop standards to facilitate the capturing, safe communications and trusted management of information concerning the total health process applied to one subject of care for individual and public health purpose and, furthermore, to coordinate modelling of other relevant standards efforts such as those regarding terminology, messaging and security.

    Terms of Reference
    • It was agreed that existing modelling conventions be used rather than new ones developed.
    • That a modelling convention be adopted across all relevant standardisation activities.
    • That terms such as "structure", "content", whether "electronic patient record" or "electronic health record" needed defining carefully.
    • That all record components be attestable, ie that this requirement recognises that it be possible to uniquely identify the patient, provider, service, product, time and location pertaining to specific encounters in the healthcare system.
    • To facilitate and enable decision support for clinical management.
    • To faithfully preserve clinical context where whole or part of the electronic healthcare record is communicated.
    • To provide an inventory of known on-going work of record systems perhaps by developing a taxonomy of "electronic patient record" or "electronic health record", examining models using top-down and bottom-up approaches.
    • To standardise a domain model within, and where possible across disciplines permitting views according to defined criteria.
    • To constrict mechanisms to extract portions of the record in a consistent and reproducible way from the "pattern" or "architecture" for authorised purposes, eg, minimum basic data sets, interchange formats, and clinical alerts/cohorts.
    • Projected audience for the standards are expected to be primarily vendors and suppliers, with the expectation that the products will ultimately benefit the clinicians and patients.
  7. RESOLUTION
    Resolved that ISO/TC215 establish a working group to address messaging and communication.

    Title
    Working Group 2 - Messaging and Communication

    Scope
    The scope of effort of the Working Group on Messaging and Communications is defined in three sections - a general statement of the types of standards to be considered, the specific domains of activity, and a set of additional tasks or working principles that will guide the effort.

    General Scope of Activity
    Any messaging and communication standard, or data interchange standard offered for consideration by the Working Group should include:

    • The functional purpose of the interchange and the circumstances in which the interchange should occur;
    • an abstract definition of the interchange or message that identifies the concepts to be communicated and the data elements necessary to represent those concepts in the interchange;
    • a means of implementing that interchange in one or more syntaxes or communication modalities.

    These standards will consider and seek to adhere to a concept model of health information and health records as defined in collaboration with the Working Group on Modelling Coordination and Health Records.

    Specific Domains
    The Working Group will undertake tasks and work items in three general areas:

    • Clinical messaging - Advance standards for data interchanges within and between health services entities that support the delivery and management of healthcare and wellness services to an international population.
    • Medical device communication - Advance standards for data interchange between medical devices and instruments, and between those devices and other health information systems to support the exchange of health-related data.
    • Business and financial messaging - Advance standards for data interchange within and between health services entities and between those entities and their financial or business partners, including sponsoring authorities, clients, consumers and subjects of care.

    Additional tasks and notes For the Working Group
    The following additional tasks or challenges should receive careful attention as the Working Group undertakes its responsibilities:

    • Establish a means of collaboration with the other TC 215 Working Groups, in particular the Working Group on Modelling Coordination and Health Records.
    • Minimise or eliminate constraints on the communications technologies that can be used with these standards.
    • Recognise and address the overlap between the clinical messaging and the business and financial messaging domains.
    • Avoid establishing standards, where acceptable transaction standards exist in other venues, and where there are not unique requirements for health information systems.
  8. RESOLUTION
    Resolved that ISO/TC215 establish a working group on health concept representation.

    Title
    Working Group 3 - Health Concept Representation

    Working Group tasks (preliminary)

    • Multi-lingual glossary of language about terminology for development purposes.
    • Good terminology development practices, leveraging best practices from other disciplines.
    • Specifications for IT-enabled terminology.
    • Structures and mechanisms for terminology access and distribution.
    • Harmonisation of a consistent set of terminology models to practically represent healthcare.
    • Enabling links/interface to other classification systems and terminologies.
    • Engage the stakeholders.
    • Proactive involvement of vendors, providers, payers, government, and private sectors to ensure broad-ased support and the emergence of implemented standards.

    Deferred or de-emphasised tasks

    • Specifications of tools for terminology use.
    • Functional specification of tools for terminology development and maintenance.
    Tasks explicitly not in present scope
    • Content; enumeration of value domains
      - lists of anatomy sites or disease entities.
    • Data models and structures.
    ISO Terminology Task Force
    Possible early tasks
    • ISO criteria for maintainable terminology
      - Distribute as database
      - Context-free identifiers
      - Concept-based orientations
    • Metaterminology about terminology
      - Word, Concept, Identifier, Code, Term, etc.
  9. RESOLUTION
    Resolved that ISO/TC215 establish a working group on security.
    Title
    Working Group 4 - Security
    Scope
    Defining standards for technical countermeasures to ensure confidentiality, availability, integrity and accountability as well as guidelines for security management in healthcare. This working group will, at this time, exclude specific standardisation in the area of quality and safety but the TC215 must be aware of the fact that it will probably become necessary to define guidelines also in these areas in the future. The working group will generally not invent new healthcare specific technologies but seek to define profiles of inter-sector security standards supporting requirements on the health sector.
  10. RESOLUTION
    Resolved that ISO/TC215 appoints the following for the Working Groups:
    Working Group 1: Appointed convenor - Australia
    Working Group 2: Appointed convenor - USA
    Working Group 3: Appointed convenor - UK
    Working Group 4: Appointed convenor - Sweden
    Working Group 4: Appointed secretary - Japan

    National bodies shall inform the Secretariat of their appointed individual and the date of the first meeting by 25 September 1998.
  11. RESOLUTION
    Resolved that ISO/TC215 instructs the Secretariat to circulate to P members a call for experts to the established working groups by 7 November 1998.
  12. RESOLUTION (?)
    Resolved that ISO/TC215 instruct the Secretariat to contact ISO and IEC Technical Committees and the suggested external organisations with the view to establishing liaisons with ISO/TC215.
  13. RESOLUTION
    Resolved that the ISO/TC215 Secretariat investigate a means of expediting communication through use of information technology tools.
  14. RESOLUTION (?)
    Resolved that it is highly desirable that new work item proposals be submitted to the Secretariat and appropriate Working Groups for discussion prior to submission for formal ballot amongst member countries.
  15. RESOLUTION (?)
    Resolved that ISO/TC215 supports the Secretariat's nomination of Mr Peter Treseder as Chairman of ISO/TC215 and requests the Secretariat to forward this decision to the ISO Technical Management Board for approval.
  16. RESOLUTION (?)
    ISO/TC215 thanks our host for their excellent meeting arrangements and their assistance throughout the meeting which contributed to a successful and productive meeting.
  17. RESOLUTION
    ISO/TC215 thanks the chairman, secretariat, and ISO representatives for guiding the meeting to a successful conclusion.
  18. RESOLUTION
    Resolved that ISO/TC215 shall hold its next meetings in:
    April 1999 in Germany
    November 1999 in Japan
    June 2000 in Canada
    March 2001 in France
    November 2001 in Brazil

Top of page

Appendix 2

Standards NZ SC606 WG3 Electronic Medical Records Progress Report - June 1998

This document the progress that has been made towards the development of a New Zealand Standard for EMRs and the future plans. It concludes with some recommendations and acknowledgements.

Background

The Committee has been established for approximately two years now and is currently composed of 26 people. It has representatives from many of the professional associations, some health service providers, the Ministry of Health, Standards New Zealand and some IT service providers and equipment manufacturers. Dick Whiddett of Massey University has been the chair of the committee since 19 March 1997. Inga Hunter of the Medical Women's Association is the Secretary, and for much of 1997 Jose Pastor took on much of the administration of the development work.

Most of the business of the committee is conducted using telecommunications, most papers are circulated via e-mail and meetings are held by audioconferencing. Full committee meetings were held on the following dates:

19 March 1997
15 May 1997
21 May 1997
4 June 1997
1 August 1997
8 June 1998

There were also some informal meetings of subsets of the committee and discussions by e-mail.

Activities and progress

Data model

The major focus of the year has been to develop a generic data model for the EMR. This model Identifies the general structure that we feel is capable of being used to store most kinds of patient information.

The model was developed by integrating the structures used by some existing record structures, and it encompasses and integrates a GP model and the NZHIS Minimum Dataset and Medical Warnings System. The model also includes the draft model of the NPI.

The model identifies some general classes of information, some attributes of the class, the format of the stored attributes and in some cases it also links to a data dictionary that defines the possible values of the data.

The model has been set up as a hypertext document which was initially available at Massey University and it is now available on the NZHIS website for general viewing and comment; it contains links to the Web site of the Term Dictionary Project. I recently arranged for a thorough review of the Web document, which identified a few minor or typographic errors, but the model as it stands is relatively coherent and consistent.

The communications issues

The Standards New Zealand committee SC606 WG4 Health Communications has been re-established and is working on the communications issues relating to EMRs. There are links between the two committees and an exchange of information in an attempt to coordinate their work.

Privacy, confidentiality, accessibility and security

No work has been undertaken in these areas by the committee. However, recent events have demonstrated the importance of these issues and it may be appropriate for a separate committee to be established to focus on these issues.

Conformance issues

These issues have yet to be considered in any detail.

Future activities

In summary, the committee is now reasonably well established, and over the last year it has begun to make significant progress towards its goals. The next tasks of the committee should be to:

  1. Consolidate current the model, particularly in the area of pharmaceuticals and prescriptions.
  2. Develop strategies to obtain wider review and feedback on the model.
  3. Integrate tests and investigations into the model.
  4. Map the requirements of specialist consultants.
Dick Whiddett, Committee Chair
Department of Information Systems
Massey University, Palmerston North.
17 July 1998
Top of page

Appendix 3

Useful Web sites:

Australian National Health Information Model

CEN 251

Corba home page

CHIME (Centre for Health Informatics and Multiprofessional Education)

Health Informatics Standards Guide, Dukes University

ISO home page

TC215 Health Informatics

The New Zealand Electronic Medical Record Standard

The New Zealand Electronic Medical Record Data Model SC606.WG3


Top of page

Appendix 4

Submission for Consideration at ISO/TC 215 Health Informatics Conference 25-26 August 1998, Orlando Florida

We would like to tell you of some work done by Standards New Zealand SC 606/Health Information Standards Committee WG3 on a generic model for Electronic Patient Records

The Model is to be found under 'Standards' on the NZHIS Web site (www.nzhis.govt.nz/model/sld01.html).

It was achieved by an evolutionary process, and by more detailed discussion between representatives of the Royal New Zealand College of General Practitioners, and the New Zealand Association of Salaried Medical Specialists. In response to your particular points:

1. What regional, national or de facto standards are candidates for ISO adoption?

We are proposing the conceptual data model for electronic patient records detailed in the Web site above.

For each proposed standard, the following supporting information is required:

  1. The purpose of the standard?
    To facilitate communication within the health sector, and to facilitate the process of healthcare generally in the information age.
  2. Why was the standard produced?
    In response to an initiative by Standards New Zealand, to fill a perceived need for a common currency of medical transactions, in creation, management, tracking, and archiving of medical records.
  3. Why was the standard produced the way it was?
    To produce a language for medical communications that is specific enough to be useful, but generic enough not to impose standardised methods of work, or restrict or limit future developments.
  4. What problems did the standard attempt to overcome and was it successful?
    The proposed standard suggests that software developers conform to a shared format for the exchange of 'objects', and makes some minimal definitions about the medical process which would need to be accepted by participating developers and users. We believe the Standard would permit untrammelled evolution in software design, while ensuring interoperability between diverse systems.
  5. Did the standard overcome unforseen barriers?
    The Standard is not yet applied; it is a prototype concept. The major achievement is in consensus between participants abut the generic content of the Standard.
2. What are the types of ISO standards that would benefit most?

The solution we propose provides a 'currency' for medical transactions that would benefit all players in the healthcare arena.

Consumers/patients would benefit from an efficient healthcare process bringing all the benefits of EPR.
 
Healthcare practitioners would preserve freedom of choice in their software platform, which would nevertheless allow communication with possibly disparate systems run by colleagues, and allow them to tap into communal healthcare resources and audit and peer review processes, again without the imposition of a single way of work or provider.
 
Software developers would at last know what standards they would need to build in and conform to, yet be liberated into a universe of unlimited possibility and innovation.
 
Academic institutions would have access to all the data-gathering efforts of ordinary practitioners. The possibilities for research of all sorts on this vast database would be unequalled. Each medical institution, even each practitioner, could become a research institute.
 
Funders would at last have a common currency for tracking, auditing, and costing the medical process.
 
Regulators similarly would have the means to put in place peer review and audit processes that could happen automatically, and would be cheap to administer.

3. What are the issues, problems and opportunities faced by these types of individuals and organisations? We consider this question too general for a useful response.
We consider that the major problem both nationally and internationally in the establishment of Health Informatics Standards has been the identification of a paradigm sufficiently specific to be useful, while being sufficiently generic to permit diversity. We consider it particularly important that any standard should not limit future potential. At this time of dramatic development in information technology, no rigid standard would endure anyway. We believe that we have found a model that fulfils these criteria, and would like to have the opportunity to present it at ISO/TC215 N4.
Top of page