CRM Taskforce Meeting Minutes - 3 May 2012

Jump to: navigation, search


  • Linda Bird (Ministry of Health Holdings, Singapore)
  • Galen Mulrooney (ONC, U.S.A.)
  • Michael van der Zel (Results4Care, Lifelines)
  • Josh Mandel (SMArt)
  • Gerard Freriks (EN13606 Association)


  • Richard Kavanagh (NHS Connecting for Health)
  • Thomas Beale (Ocean Informatics)
  • Stan Huff (Intermountain Healthcare)
  • Grahame Grieve (Health Intersections)
  • Cecil Lynch (Accenture) - CIMI Clinical Modelling Taskforce


  • Review Report
  • Breakout session in Pleasanton
  • Plans/Actions

Brief Summary

  • Review of draft CIMI meeting agenda
  • Discussion of breakout session
  • Review Report – Update to Recommendations
    • Recommendation 1: The proposed reference model is thoroughly tested using a representative set of clinical models that together demonstrate all aspects of the reference model;
    • Recommendation 2: A set of stable clinical patterns (e.g. OBSERVATION, EVALUATION, INSTRUCTION, ACTION, TIME_SERIES, SCHEDULE) are defined as archetypes, upon which more specific clinical models are based, within an agreed framework of ‘clinical model specialisation layers’.
    • Recommendation 3: The CIMI reference model is fully documented in a format that promotes effective understanding by both clinical modellers and implementers. This documentation should include both example models and example instance data;
    • Recommendation 4: A coherent modelling methodology is documented to ensure consistent models, including modelling style guides and best practice guidelines;
    • Recommendation 5: The Archetype Object Model (and associated serialisations) is reviewed against requirements, and extended to fully express the semantic meaning of archetypes, (including the meaning of the relationship between nodes);
    • Recommendation 6: Further investigation is planned to consider issues, such as the following:
      • The syntax and use of terminology bindings for both the meaning and valid values of nodes in the clinical models;
      • The semantic meaning between all structural nodes;
      • The use of CLUSTER.value (including defining the meaning)
      • Whether null_flavour (or similar) should be allowed on ENTRY or SECTION
      • Whether further simplifications can be made to the reference model (e.g. demographics);
      • Modelling methodology to ensure consistency of models, including the development of modelling style guides and best practices;
    • Recommendation 7: The requirements of iso-semantic clinical models are explored further, in terms of both (a) the ability to transform CIMI models to/from iso-semantic representations in other languages/standards (e.g. CDA, openEHR, ISO13606, DCM, CEM), and (b) the ability to transform CIMI models between iso-semantic representations that use a different split between terminology pre-coordination and structure (Please refer to Appendix B).
    • Recommendation 8: Appropriate tooling is developed and/or adopted, which allows clinical models to be created and is capable of generating ADL 1.5.
    • Recommendation 9: Any relevant Intellectual Property issues relating to the CIMI reference model are addressed.
    • Recommendation 10: A UML Profile is developed, in collaboration with the OMG, which allows CIMI models to be created in UML.
    • Recommendation 11: The taskforces be reorganised to support the above activities
  • Review of proposed reference model


  • Distribute minutes, report and model for final comment: Linda
  • Review report: All