# Doubt about the link between archetypes' ontology and reference model **Category:** [Implementers (archive)](https://discourse.openehr.org/c/implementers-archive/158) **Created:** 2011-09-21 08:58 UTC **Views:** 3 **Replies:** 8 **URL:** https://discourse.openehr.org/t/doubt-about-the-link-between-archetypes-ontology-and-reference-model/15088 --- ## Post #1 by @Jose_Antonio_Lorenzo I have recently started working with archetypes and medical terminologies, and I have a doubt regarding how the terms in the ontology section of archetypes are related to the underlying reference model. For example, if I am not wrong, the "meaning" attribute of the RECORD_COMPONENT class of the reference model of the CEN 13606 standard is the place where the ontology term will be stored for that record component. An example will make this more clear. Suppose we have an archetype like this: definition Composition[at0000] ... Section[at0001] ... Entry[at0002] ... ... ... ontology terminologies_available = <"SNOMED-CT", ...> term_definitions = < ["en"] = < items = < ["at0000"] = < text = <"Body figures"> description = <"Main body figures of the subject of care"> > ["at0001"] = < text = <"Weight figures"> description = <"Figures related to weight"> > ["at0002"] = < text = <"Body mass index"> description = <"The body mass index of the subject of care"> > ... ... term_binding = < ["SNOMED-CT"] = < items = < ["at0000"] = <[SNOMED-CT::60621009]> Now suppose we have an instance of the EHR extract built with this archetype. We have an Entry object that has an attribute "meaning" (because Entry inherits from Record_Component). The question is, which is the value of the attribute "meaning" for this Entry object? at0002? "Body mass index"? or SNOMED-CT::60621009? Thank you very much. --- ## Post #2 by @Heath_Frankel3 In openEHR it would be at0002. Regards Heath --- ## Post #3 by @system Hello José Antonio, CEN/ISO 13606 is designed to work with and without archetypes. In other words, you can have an instance (EHR_EXTRACT) created following an archetype definition or not. In the first case, the meaning attribute can be used to link the appropriate node of the archetype that defines the instance. "meaning" is a CV, so the CV.codeSystem should be the archetype identifier and CV.code should be "atXXXX". In this case the archetype is used as a mediator or interface between the data and the real terminologies. In the second case, when an instance is created without following any archetype, you can use the meaning attribute to directly map to a terminology. In this case, the CV.codeSystem should be a terminology and CV.code a code from that terminology. David 2011/9/21 José Antonio Lorenzo <[jalorenzo@bdigital.org](mailto:jalorenzo@bdigital.org)> --- ## Post #4 by @thomas.beale You can see from the below the differences between the 13606 RECORD_COMPONENT class on the left and the openEHR LOCATABLE class on the right. These provide the attributes in which archetype meta-data are recorded in the data - specifically archetype ids and node ids. The LOCATABLE class was co-designed for use with archetypes. The 13606 class is a partial copy of an earlier (6 or 7 years older) idea of how LOCATABLE should look, modified by the standards process. The 'meaning' attribute should correspond to a term from the ontology part of the archetype, e.g. local::at0002 or similar. You can see that it is much simpler in openEHR - the 'archetype_node_id' field is just a String, containing e.g. 'at0002'. I have to admit I am not 100% clear on whether people using 13606 know what to do. Fr the 13606 revision in ISO next year, if it is to be continued, it would be highly desirable to align these attributes to openEHR.

- thomas beale --- ## Post #5 by @Ognian_Pishev José, The Spanish word ‘duda’ is translated as ‘query’, ‘question’. Ogi Pishev --- ## Post #6 by @Jose_Antonio_Lorenzo Thank you very much for your answers, they have been very helpful. --- ## Post #7 by @thomas.beale I probably should clarify a bit - in 13606, the 'meaning' object is a CD, i.e. same idea as an openEHR DV_CODED_TEXT. So it has the various fields including - terminology id / code system id - in openEHR we put 'local' to mean 'from the archetype' - code / code string - i.e. 'at0002' - code text - the actual text of at0002. This is, in natural language terms, the 'meaning' - thomas --- ## Post #8 by @system Thomas, it seems that there is an error in the imported model for 13606 that you have. The meaning attribute is a CV, not a CD. David 2011/9/23 Thomas Beale <[thomas.beale@oceaninformatics.com](mailto:thomas.beale@oceaninformatics.com)> --- ## Post #9 by @thomas.beale David, thanks, I fixed it. There are undoubtedly other errors - if anyone wants to review, that would be useful. You can do it pretty conveniently in the AWB. But I don't think this error changes the basic argument... - thomas --- **Canonical:** https://discourse.openehr.org/t/doubt-about-the-link-between-archetypes-ontology-and-reference-model/15088 **Original content:** https://discourse.openehr.org/t/doubt-about-the-link-between-archetypes-ontology-and-reference-model/15088