# difference and relationship between openEHR and EN13606 **Category:** [Technical (archive)](https://discourse.openehr.org/c/technical-archive/156) **Created:** 2015-08-26 09:14 UTC **Views:** 3 **Replies:** 46 **URL:** https://discourse.openehr.org/t/difference-and-relationship-between-openehr-and-en13606/15388 --- ## Post #1 by @edwin_uestc dear all , how could i explain to someone difference and relationship between openEHR and EN13606 thx --- ## Post #2 by @Seref Maybe this would help: [http://search.informit.com.au/documentSummary;dn=950616334398351;res=IELHEA](http://search.informit.com.au/documentSummary;dn=950616334398351;res=IELHEA) --- ## Post #3 by @edwin_uestc it says 2006 .as days go on ,if there is any offical statements ,that will be great help [details="(attachments)"] [ATT00003.txt|attachment](upload://vDWbASYj4Q8jl3VnoCTmsO05gLv.txt) (192 Bytes) [/details] --- ## Post #4 by @system This might help a little [http://www.slideshare.net/atalagk/implementation-and-use-of-iso-en-13606-and-openehr](http://www.slideshare.net/atalagk/implementation-and-use-of-iso-en-13606-and-openehr) Similarities: Both use archetypes and ADL and two-level information modelling. Both share the EHR, FOLDERS,COMPOSITIONS, ENTRY, ELEMENT classes. Some archetype tools can work with both styles of archetype e.g LinkEHR and Archetype Workbench. The just announced ADL2 Archetype editor/ template designer tools (beware!!! Early developer versions!!) [http://ehrscape.marand.si/designer/template-editor.html](http://ehrscape.marand.si/designer/template-editor.html) [http://ehrscape.marand.si/designer/archetype-editor.html](http://ehrscape.marand.si/designer/archetype-editor.html) should be relatively easy to adapt to 13606 or other archetype-based reference models such as CIMI. They will be open sourced very soon. Differences: The EHR reference models are different In spite of sharing the classes above, the attributes within those classes differ openEHR sub-classes ENTRY into ADMIN_ENTRY, OBSERVATION, EVALUATION, INSTRUCTION and ACTION The datatypes are different The demographic models are different The EHR Extract formats are different 13606 is intended primarily for the communication of EHR extracts across systems but some persistence repositories exist. openEHR is intended primarily for data persistence and querying within systems but it is possible to message openEHR data. 13606 does not (currently) support templates but ADL/AOM2 is being considered 13606 does not support AQL Archetype Query Language 13606 is formal ISO standard but is closed source i.e. behind a paywall, as in normal for ISO published material openEHR is open source and freely available There is a great deal of cross-communication between the two communities and a number of people work with both formalisms. It is possible to transform data between the two formalisms but they are not directly compatible. I hope that is accurate and non-contentious! Ian --- ## Post #5 by @system See my earlier response. I think that is probably as official as you can expect for now!! 13606 and openEHR share some aspects of their design, there are many formal and informal links between the two communities e.g. Thomas Berale and I are both invited experts to the 13606 group, and there are opportunities for shared development especially around tooling but 13606 and openEHR do need to be regarded as two different solutions to different problems, with different licensing and development/maintenance models. Ian --- ## Post #6 by @pablo Hi, I would say that the main difference is that 13606 is for data communication and openEHR is for EHR architecture, both based on archerypes. For detailed differences just look at both information models, you will see that 13606 IM is much simple. About the specs, 13606 has 5 "chapters", including communication and security, and openEHR specs don't have those. The best way of knowing the differences is just to download the specs of both and compare them. Hope that helps, Cheers, Pablo. --- ## Post #7 by @system Dear Pablo, According to the scope statement: the 13606 is for the creation of the EHR-EXtract for communication between IT-systems and for the definition of the Information Viewpoint in Interfaces with system services. Gerard Gerard Freriks +31 620347088 [gfrer@luna.nl](mailto:gfrer@luna.nl) --- ## Post #8 by @yampeku I agree with most of the points, but I'm curious why you say that 13606 does not support AQL (and in any case wouldn't be "AQL does not support 13606"?) --- ## Post #9 by @system Yes, that is a good question, I did not know that AQL was considered to be OpenEHR specific\. In my opinion it was a bound to the archetype model, not to the reference model\. --- ## Post #10 by @system Hi Diego, I was not aware of any 13606 implementations that support AQL , although I am sure there is some sort of path-based querying. AFAIK AQL is not part of the 13606 scope. Happy to be corrected. Ian --- ## Post #11 by @Seref Well, technically, it is not part of the openEHR scope either. Happy to be corrected :) --- ## Post #12 by @system It is definitely on the openEHR Specifications Roadmap. That was a clear decision at the Oslo meeting a year ago. There are at least 3 implementations that I know of and more back-end vendors are intending to implement but I know what you mean 'technically';) I agree that AQL is RM agnostic but am not aware of any non-openEHR implementations @Diego/ Gerard?? Ian --- ## Post #13 by @system Is there a Xhosa implementation of 13606 or OpenEHR? Does that mean OpenEHR or 13606 are not able to support Xhosa? I would leave it with: AQL is an archetype bound query language, and every system which is build on archetypes is able to implement AQL\. --- ## Post #14 by @system Hi Bert, "I would leave it with: AQL is an archetype bound query language, and every system which is build on archetypes is able to implement AQL." That is fair enough but we were asked to characterise the differences between 13606 and openEHR and I am comfortable that the actual and formal adoption of AQL is one of those differences. AQL is on the openEHR specifications roadmap but AFAIK this is not the case for 13606. Of course that does not stop 13606 vendors implementing AQL but in terms of actual differences between the 2 communities the adoption, or intention to adopt AQL seems (from the outside) somewhat different both at a practical and formal level. Although AQL adoption in the openEHR community is far from universal, most of the vendors/developers that I have spoken to see it as something they want to implement, particularly as GDL is somewhat dependent on AQL. I am just trying to ascertain if there is similar enthusiasm/intention amongst 13606 vendors, or if AQL forms part of the current 13606 refresh discussions. Ian --- ## Post #15 by @system I'd agree with Ian here. While both could possibly support AQL, the difference I see is in intent, scope and actual implementation. As Gerard says, 13606's main aim is to communicate between IT-systems and for this, AQL may not be quite as fundamental as it is to openEHR. Sebastian --- ## Post #16 by @system If you see it from the formal point of specifications, you are right, but as you say, ISO13606 is in a renewal process, and it is hard to foresee what will come out of that\. You write yourself that support for ADL/AOM 2\.0 is being considered, I am very happy to read that\. Then it is a small step to AQL based on AOM 2\.0 adoption\. I couldn't think of a strong reason why they should not adopt AQL when they adopt AOM2\.0\. Untill now, there has only been AOM 1\.4, also for OpenEHR, and the AQL for AOM 1\.4 has always been a moving target\. That is why I never implemented it, but also because I did not really need it\. So we could say, that there is no formal specification for AQL based on AOM 1\.4 and that OpenEHR, technically said, like ISO13606, does not support AQL\. --- ## Post #17 by @system How about filtering messages? Messaging is a process, in the cloud it will become very important, it will become more then two system interchanging information\. It doesn't matter were your medical data are, but how they come to you matters, and I can imagine usecases for filtering\. \(just a quick example\) --- ## Post #18 by @system Hi, I must repeat the scope of 13606 verbatim once more. It is NOT only for messaging but also for **Interfaces** Gerard Freriks +31 620347088 [gfrer@luna.nl](mailto:gfrer@luna.nl) 1. _**Scope**_ _This standard is for the communication of part or all of the electronic health record (EHR) of a single identified subject of care **between EHR systems, or between EHR systems and a centralised EHR data repository.**_ _**It may also be used for EHR communication between an EHR system or repository and clinical applications or middleware components (such as decision support components) that need to access or provide EHR data**._ *This standard will predominantly be used to support the direct care given to identifiable individuals, or to support population monitoring systems such as disease registries and public health surveillance. Uses of health records for other purposes such as teaching, clinical audit, administration and reporting, service management, research and epidemiology, which often require anonymisation or aggregation of individual records, are not the focus of this standard but such secondary uses might also find the standard useful.* *This Part 1 of the multipart standard is an Information Viewpoint specification as defined by the Open Distributed Processing – Reference model (ISO/IEC 10746). This standard is not intended to specify the internal architecture or database design of EHR systems.* --- ## Post #19 by @Kalra_Dipak Dear All, This is an interesting discussion, and I would like to stress the complementarity of the two. openEHR is, as others have said, an important consolidator of the state-of-the-art in best practices for the design of an electronic health record architecture, repositories and the underpinning of EHR systems. An important advantage is that it specifications are publicly accessible, and of course it has a vibrant community and a large number of tools to support its use. 13606 has always had a good relationship with openEHR, but is primarily intended to be an interface standard between heterogeneous EHR systems, and is therefore optimised for that purpose (e.g. for mappings), which means its reference model is definitely simpler. There are many countries and situations where it is essential to have a formal international standard in order for it to be acceptable as part of a national strategy. Some vendors have also indicated that they like the inevitable stability of a standard, which changes infrequently. 13606 also has a community and tools, and of course many of its community are also part of openEHR, and vice versa. If one takes a high-level look at the many different globally-used representations of health data, it is easy to see that these two reference models are indeed very similar. Whilst near to the ground we can easily be tempted to focus on their minor differences, I believe it is of greater value to society and to our field if we can regard them - and champion them - as a mutually reinforcing pair of models. The specification of archetypes is very mature, and during the revision we expect to upgrade to the latest AOM (which is 2.0). This part of the standard will also remain focused on a logical representation supporting archetype interchange. As has been pointed out, AQL could in theory have been added to the standard, since it could “work" with 13606. However, another important imperative for a standard is that it has reached a sufficient level of maturity and stability. It was also felt important by the working groups of CEN and ISO that we do not introduce something very novel into this revision process. I did suggest that we consider adding a sixth part to the standard to support the distributed analysis of electronic health records (such as communicating queries). It was felt wiser, and I support this view, not to introduce something new to these five parts of the standard, but once it has finished its revision to propose a new work item to CEN and ISO on the querying of EHRs. AQL will inevitably be an important contribution to that new work item, and hopefully by the time we are ready for it the AQL specification will be very mature and there will be much more experience of its use, making it an ideal specification to standardise. Thank you all for your excellent contributions in different areas of EHR representation, communication and implementation - to keep advancing our field and the quality of EHRs world wide. With best wishes, Dipak --- ## Post #20 by @pablo Dear Gerard, IMO "communication" includes the interfaces, I didn't excluded them :D --- ## Post #21 by @system Hi Gerard, Agreed - I was using messaging loosely - 'interfacing between systems' is better. --- ## Post #22 by @system Thanks Dipak, A very clear and helpful statement of current and future intent. I too agree that we should not focus negatively on the differences and that they are mutually reinforcing but people do ask and it's important that we are clear that while 13606 and openEHR share a number of tools, technologies, philosophies and even people + good relationships), they are not currently interchangeable or directly interoperable. From a high-level perspective they are indeed very similar but the detailed differences do matter to implementers, and I think we need to be clear to the market about these differences. Thanks too for the perspective on AQL adoption - makes complete sense to me in the 13606 context. Ian --- ## Post #23 by @Kalra_Dipak Dear Ian, Thanks also for your helpful reflections. I agree that once the standard is close to final we should perform and publish a detailed comparison and cross mapping between the reference models, as an aid to system implementers and tool makers. With best wishes, Dipak Kalra --- ## Post #24 by @system That is good to know. Gerard Gerard Freriks +31 620347088 [gfrer@luna.nl](mailto:gfrer@luna.nl) --- ## Post #25 by @system We are in agreement, then. :-) Gerard Freriks +31 620347088 [gfrer@luna.nl](mailto:gfrer@luna.nl) --- ## Post #26 by @system Hi! Where can one find proposals/diagrams describing the refreshed RM (reference model) in the new 13606 revision? Will 13606 keep using the old data types or harmonize more with CIMI or OpenEHR? Is there now consensus/majority regarding using ADL/AOM 2.0 for 13606? If so, great! When it comes to "simplifying" the RM (or perhaps moving complexity to another meta/design-pattern layer) I think CIMI has gone further than 13606. Are there any plans of aligning 13606 with CIMI? //Erik Sundvall onsdag 26 augusti 2015 skrev Kalra, Dipak <[d.kalra@ucl.ac.uk](mailto:d.kalra@ucl.ac.uk)>: --- ## Post #27 by @system By the way feel free to add some of the onsdag 26 augusti 2015 skrev Erik Sundvall <[erik.sundvall@liu.se](mailto:erik.sundvall@liu.se)>: --- ## Post #28 by @system Thanks Dipak, for announcing this, it is great news\. And also thanks for explaining the current position of AQL in relation to 13606 and the way it is planned to integrate in the standard\. Best regards Bert --- ## Post #29 by @system Oh, that got sent too early, sorry. I meant to say: Feel free to add some of these descriptions to the stack overflow question: [http://stackoverflow.com/questions/32010122/are-the-hl7-fhir-hl7-cda-cimi-openehr-and-iso13606-approaches-aiming-to-solve](http://stackoverflow.com/questions/32010122/are-the-hl7-fhir-hl7-cda-cimi-openehr-and-iso13606-approaches-aiming-to-solve) Two people thought the question was bad enough to down-vote it, but I think this discussion shows it to be useful, so maybe that can change. //Erik onsdag 26 augusti 2015 skrev Erik Sundvall <[erik.sundvall@liu.se](mailto:erik.sundvall@liu.se)>: --- ## Post #30 by @Heath_Frankel3 Technical, the original grammar for AQL was bound to openEHR RM classes, composition, version, observation, etc. theoretically it could be generalised to be a RM agnostic and should be the goal of the current AQL specification work if it hasn't already been done in the antlr grammar. Regards Heath --- ## Post #31 by @system Hi, Next week we will meet in Brussels and discuss the proposals, discussion papers by the various working parties. I think that the RM and data types will be simplified. leaving semantics to be dealt with at the archetype level using standardised archetype patterns. (participations, demographics, and things like ranges and more) On behalf of the EN13606 Association I take part in the CIMI working group. CIMI will help create archetype patterns. CIMI models will be able to be converted to EN13606 artefacts. And all in spite of the fact that CIMI has a very simple and strange RM derived from 13606-1. (At least that is the way I look at it) The strange thing being the fact that they have defined a ‘Super ENTRY’ class that can contain the ‘normal’ ENTRY class. They designed this because of the need to model for instance panels as one entity and each of its components. (I’m of the opinion that the present 13606 RM can deal with all the CIMI requirements. This is how I create panels usually.) Gerard Freriks +31 620347088 [gfrer@luna.nl](mailto:gfrer@luna.nl) --- ## Post #32 by @system Like the Ocean Archetype editor. It only supports the OpenEhr RM only. That is understandable and no problem. The market will fill in that gap. --- ## Post #33 by @system Presumably the outline syntax, SELECT, CONTAINS etc is generalisable? Ian --- ## Post #34 by @Seref All of it is very, very generalizable :) --- ## Post #35 by @thomas.beale openEHR has an EHR Extract specification <http://www.openehr.org/releases/RM/latest/ehr_extract.html>as well, which is more flexible than the 13606 one e\.g\. it can include information from more than one patient, and accommodates both openEHR and non\-openEHR content\. \- thomas --- ## Post #36 by @thomas.beale It's not\. It's exactly the same no matter what the reference model\. Of course, to properly check AQL queries and execute them in a specific environment, access to a representation of the RM being used will be needed\. \- thomas --- ## Post #37 by @thomas.beale I would suggest that CIMI has been simiplified to the point of not being directly usable as an RM by openEHR or 13606 - most of the needed context information is gone in CIMI, and it doesn't distinguish any kind of 'Entry' or clinical statement. This was a conscious choice in the CIMI community, designed to get buy-in from a much wider range of stakeholders than openEHR or 13606 deals with. Technically, the CIMI approach is to soft-model nearly everything in 'reference archetypes'. - thomas --- ## Post #38 by @edwin_uestc could we just add a page on openEHR website to illustrate these points thx --- ## Post #39 by @thomas.beale if you search on the , with either '13606' or 'CIMI' you will find a lot of material. - thomas --- ## Post #40 by @Seref Sorry, but I have to ask: are you doing a homework? --- ## Post #41 by @system > I would suggest that CIMI has been simiplified to the point of not being directly usable as an RM by openEHR or 13606 \- most of the needed context information is gone in CIMI, and it doesn't distinguish any kind of 'Entry' or clinical statement\. Are you saying, that the context information from the reference model is not used? > This was a conscious choice in the CIMI community, designed to get buy\-in from a much wider range of stakeholders than openEHR or 13606 deals with\. Technically, the CIMI approach is to soft\-model nearly everything in 'reference archetypes'\. and the archetypes fill in the missing reference model context parts? If so, then this makes the two level modeling approach, of course, much more flexible, a kind of new database approach/technique, usable for virtual anything\. Sorry if I misunderstand you\. Bert --- ## Post #42 by @system This sounds like an accusation \(the "Sorry, I have to ask"\-part\)\. Then I don't understand the point\. How many in this community, in the past, were on this mailing\-list for their school or university\.? For me, I don't care, homework or other work\. We are all learning and work in the school of life\. --- ## Post #43 by @system > How many in this community, in the past, were on this mailing\-list for their school or university\.? > For me, I don't care, homework or other work\. We are all learning and work in the school of life\. > See the very interesting discussion he/she initiated\. --- ## Post #44 by @system Edwin has been around here for a while. I think the suggestion of creating a wiki page based on this discussion (and updating Erik's StackOverflow page) is worthwhile. It is question that is commonly asked and a single summary would be helpful. Any volunteers? Ian --- ## Post #45 by @system If no one else is volunteering, I can do it over a month or so\. I am really quite busy and under time\-pressure at this moment Bert --- ## Post #46 by @thomas.beale the has no context information in it. that's the idea. it makes it more flexible in one sense, but also harder for implementers - now they cannot know where even basic context like subject, times, locations etc are - all that has to be obtained from archetypes. The 'flexibility' comes with a price... What goes in any particular RM for some particular domain or industry needs to be the result of careful analysis of - thomas --- ## Post #47 by @system I agree it is a balancing act in how far the semantics should be in the RM or in the archetypes\. Both ways have their pro and contra\. Thanks for explaining it Bert --- **Canonical:** https://discourse.openehr.org/t/difference-and-relationship-between-openehr-and-en13606/15388 **Original content:** https://discourse.openehr.org/t/difference-and-relationship-between-openehr-and-en13606/15388