# greetings and 2 questions **Category:** [Implementers (archive)](https://discourse.openehr.org/c/implementers-archive/158) **Created:** 2015-04-20 01:10 UTC **Views:** 2 **Replies:** 28 **URL:** https://discourse.openehr.org/t/greetings-and-2-questions/13474 --- ## Post #1 by @Danny_Nguyen Hi, The following questions may be US centric but my general intent is the general healthcare topic of interoperability: 1. What is the difference between WorldVistA and OpenEHR? I know VistA is written in Mumps which claims a superior organization of vast numbers of dependencies. 2. If HL7 ([FHIR being developed: http://www.hl7.org/FHIR)](http://www.hl7.org/FHIR/) is the standard in data protocol between all healthcare software vendors, doesn't that mean that any system that can use that protocol solve the interoperability problem? I'd really appreciate hearing the facts and then separating those from the different point of views. Best, --- ## Post #2 by @yampeku Hello Danny, 1) I didn't know about the existence of WordVistA, but by the looks of it seems to be a software, is it right? openEHR is a set of open specifications to build future-proof health information systems. There are some reference implementations in different programing languages and technologies. 2) I don't think anyone would say that a single standard solves interoperability problem. I assume that FHIR would be perfect if your requirements align with the original purpose of each Resource. However, I think there is still work to do with the profiling of FHIR resources. I believe that just to know if the profile of the server and the client are "compatible" is still an open problem, and without that you end with one-to-one agreements which I find difficult to call interoperability. I think this part is better solved with the dual model approach. To make an analogy, the archetype approach is similar to FHIR profiling, but everything would come as profiles of FHIR Composition Resource. Instead of a 80-20 philosophy, archetypes follow a maximal approach, which can be specialized or templated for your use case. This assures that specialized archetypes follow both the original archetype and the reference model. You can see examples of archetypes in [http://www.openehr.org/ckm](http://www.openehr.org/ckm) PS: ISO13606 is an archetype based ISO standard for the semantic interoperability of EHR data, and has been a standard for some years now. FHIR is still a DSTU. I think that your second question could be rewritten with any other healthcare standard ;) Regards --- ## Post #3 by @Danny_Nguyen Hi Diego, Thank you for the response. 1. WorldVista is the group but yes it seems to be a software too. There is also a software called openVista. The origination of Vista came from the Veterans affairs in the US. It was written in MUMPS programming language with is also the database. From my understanding, GT.M is the mumps database but mumps is also the language. The database is indexed so I think that is noSQL. How is openEHR different? Which implementations have been successful and do they talk to each other? 2. Wow, this is confusing to me. Can you please elaborate? Please explain like I am a two year old. I am a clinician by training and new to programming. Archetypes? I found this: [http://www.cise.ufl.edu/research/ParallelPatterns/CITarchetypes/archetypes.html](http://www.cise.ufl.edu/research/ParallelPatterns/CITarchetypes/archetypes.html). Also can you explain how interoperability is still a problem if there is a standard like HL7/ISO13606? ISO13606 mentions something about identifiable information. Best, danny --- ## Post #4 by @Danny_Nguyen also, I just found this article to shed some light on what is happening in the United States: [http://www.openhealthnews.com/hotnews/google-joins-vista-team-proposing-open-source-ehr-department-defense](http://www.openhealthnews.com/hotnews/google-joins-vista-team-proposing-open-source-ehr-department-defense) I'd like to hear some of the reasons both technical and not why other options instead of Vista are being used. I do not sell any of this stuff, I just want to learn about the differences and thought process of for all the systems. At the end of the day, I'm interested in how this will be received by users, specifically patient consumers. --- ## Post #5 by @yampeku Sorry for the confusion, I assumed you had a background in FHIR :) I think what archetypes are is well explained here: [https://openehr.atlassian.net/wiki/display/healthmod/Introduction+to+Archetypes+and+Archetype+classes](https://openehr.atlassian.net/wiki/display/healthmod/Introduction+to+Archetypes+and+Archetype+classes) I encourage you to visit the wiki. Archetypes are aimed to clinicians so you will have no problems with that. I'll also send this mail to the clinical discussions list, from which you will get responses suited for you. The use of standards in healthcare is not the as widespread as it is in other domains, and not all factors are 'solvable' (i.e. political factors) --- ## Post #6 by @tonyshannon Danny Not sure what you're background is in informatics, but sense you could fairly easily get confused here. There may be something helpful in these articles. [frectal.com/2013/09/18/transatlantic_thoughts_onvista_nhs/](http://frectal.com/2013/09/18/transatlantic_thoughts_onvista_nhs/) [frectal.com/2014/06/30/21stc-healthcare-open-platform/](http://frectal.com/2014/06/30/21stc-healthcare-open-platform/) Essentially; #VistA is a successful EHR from the VA, which needs refactoring to bring it into the 21st Century. Its architecture is complex + and it could do with improvements along the lines of openEHR imho. #openEHR offers the technical specifications & architecture that many of us feel is well suited to 21st Century healthcare platform, esp the two level modelling elements (aka archetypes and templates) to build scalable healthcare applications. openEHR can be implemented in several flavours with languages from .net to java and databases from SQL to NoSQL in active use. That first article was to suggest these 2 camps could learn from each other.. ie VistA could learn from openEHR and openEHR could learn from the NoSQL properties of M. Much of the standards efforts in healthcare to date has been about standardising the messages between system eg HL7 , FHIR etc, which doesn't get to the heart of the 21st C challenge. openEHR goes deeper than that to standardise the architecture of healthcare applications via a platform approach, (which by the way can help with message standardisation as a by-product). Further background reading here; [http://frectal.com/book/](http://frectal.com/book/) regards Tony Re: greetings and 2 questions openEHR-implementers [[openehr-implementers-bounces@lists.openehr.org](mailto:openehr-implementers-bounces@lists.openehr.org)] on behalf of Diego Boscá [[yampeku@gmail.com](mailto:yampeku@gmail.com)] Outlook Web Access has blocked access to attachments. Blocked attachments: ATT00001.txt. To help protect your privacy, some content in this message has been blocked. If you are sure that this message is from a trusted sender and you want to re-enable the blocked features, click here. To: For openEHR implementation discussions [[openehr-implementers@lists.openehr.org](mailto:openehr-implementers@lists.openehr.org)]; For openEHR clinical discussions [[openehr-clinical@lists.openehr.org](mailto:openehr-clinical@lists.openehr.org)] Sorry for the confusion, I assumed you had a background in FHIR :) I think what archetypes are is well explained here: [https://openehr.atlassian.net/wiki/display/healthmod/Introduction+to+Archetypes+and+Archetype+classes](https://openehr.atlassian.net/wiki/display/healthmod/Introduction+to+Archetypes+and+Archetype+classes) I encourage you to visit the wiki. Archetypes are aimed to clinicians so you will have no problems with that. I'll also send this mail to the clinical discussions list, from which you will get responses suited for you. The use of standards in healthcare is not the as widespread as it is in other domains, and not all factors are 'solvable' (i.e. political factors) 2015-04-20 12:48 GMT+02:00 Danny Nguyen <[dannyn08@gmail.com](mailto:dannyn08@gmail.com)>: Hi Diego, Thank you for the response. 1. WorldVista is the group but yes it seems to be a software too. There is also a software called openVista. The origination of Vista came from the Veterans affairs in the US. It was written in MUMPS programming language with is also the database. >From my understanding, GT.M is the mumps database but mumps is also the language. The database is indexed so I think that is noSQL. How is openEHR different? Which implementations have been successful and do they talk to each other? 2. Wow, this is confusing to me. Can you please elaborate? Please explain like I am a two year old. I am a clinician by training and new to programming. Archetypes? I found this: [http://www.cise.ufl.edu/research/ParallelPatterns/CITarchetypes/archetypes.html](http://www.cise.ufl.edu/research/ParallelPatterns/CITarchetypes/archetypes.html). Also can you explain how interoperability is still a problem if there is a standard like HL7/ISO13606? ISO13606 mentions something about identifiable information. Best, danny --- ## Post #7 by @heather.leslie Hi Danny, I’ll add to Tony’s lists with another couple of blog references: · http://www.woodcote-consulting.com/openehr-a-game-changer-comes-of-age/ and · [https://omowizard.wordpress.com/2014/07/19/clinical-modelling-openehr-style/](https://omowizard.wordpress.com/2014/07/19/clinical-modelling-openehr-style/) regards Heather --- ## Post #8 by @Danny_Nguyen **Diego:** Thank you for that. Are you saying that health is on the same political level as war? The link you sent about archtypes is great! However, the Demo of observation has holes in the sense that it doesn't link clinical data with engineering data frameworks clearly. I think an approach that would both serve clinicians and engineers simultaneously would be to have a complete SOAP(subjective, objective, assessment, and plan) in a "fake" visit that encompasses all types and subtype of archetypes in a typical visit/encounter. The visit would need to be exaggerated where the patient consumer generates all of the types of data then circle the types of "data" be generated and label it like so in the demo document. That would be a great starting point to either refute or accept what kind of "archetype" of data both from a clinical perspective as well as a engineering perspective. **Heather:** I think my approach to combine the clinical and engineering data mentioned above begins to address the archetype editor and template design you mention in your blog. After the relationships become more solidified, then there can be a more intelligent assignment of how the data can be dropped in their respective buckets: xml, json, or images. I think it basically breaks down to text (including alpha numeric) or images (MRI, CT scans, x-rays blah blah). **Tony:** I've heard good things about VistA. Here is the relationship tree that I thought was interested but didn't know how it related to interoperability: [http://code.osehra.org/vivian/](http://code.osehra.org/vivian/). Tom Munnecke and company who built the original kernel for Vista had the vision of creating a program that would allow users (clinicians) to modify it. I think the unintended consequence was that there was no standard on how data would be put in buckets so other computers could talk to it. I want to know how the "vivian" vista framework this relates to openEHR in regards to archetype and template modeling? Best, danny --- ## Post #9 by @tonyshannon Danny The vivian tree doesnt relate to interoperability, more a mindmap of the current VistA architecture. The vivian tree doesnt directly relate to openEHR archetypes or templates either, as VistA essentially misses/lacks the 2 level modelling openEHR offers. Suggest you need to read more about the 2 level modelling that openEHR offers.. then come back with further informed questions. Heathers has sent 2 useful related articles Tony --- ## Post #10 by @Danny_Nguyen I'm looking at the clinical knowledge manager and I just get more confused. An EHR should not be owned by anyone and money should be made on the services (care) rendered to the patient consumer with them choosing, not on how the data is stored. This is in dire need to be simplified. What are the shortcomings of archetypes and why does there need to be templates? I understand that clinicians need an agile platform but why would you need to to have such a vast library. At the end of the day, humans have genetic code that gets translated to proteins then to there the "human organism" then there are events by the environment, including other organisms. Vitals are important but I don't see why you need so many ways to store a vital whether it's by a clinician, patient, or medical device/wearable. Regulations on what is a "legitimate" vital is arguable. A vital sign is a vital sign whether you measure it with a sphygmomanometer or your two fingers on your wrist. Clinicians already do this. Trauma is trauma, whether it's by a turtle or a pencil. Now onto billing, value based cased is important but having 18,000 ICD-10 codes starting October 1, 2015 is absurd. Even when you're looking at the electronic BP monitors in hospitals, each one has different algorithm determined by the vendor. The point is standardization starting at the fundamentals, working from principles. In medicine and healthcare, there are people, products, and services. Healthcare should not be any different unless we are selling immortality. That will eventually be a service if super AI can produce it with genetic modification and regeneration of telomeres and recomposition of atoms. But I digress. Why is healthcare so complicated that new names for things constantly need to be made and then accounted for in data classification??? Back to the fundamentals. People, product, services. Is there something inherently different about health as a service? I'm really curious on what others in the group think. Best, danny --- ## Post #11 by @tonyshannon Danny You need to do some more reading. Healthcare is not just complicated, its complex. Complex Adaptive Systems such as healthcare exhibit vast diversity, yet common patterns are seen within, which is where you are trying to navigate. Read up more here [http://frectal.com/book/chaos-complex-complicated-simple-and-cynefin/](http://frectal.com/book/chaos-complex-complicated-simple-and-cynefin/) Tony --- ## Post #12 by @Danny_Nguyen Okay. You tell me what your approach and how you would solve the problem besides open sourcing everything? Looking at your Leeds initiative, I don't see how increasing the barrier in a day an age where you can easily click a button to opt out of sharing your information (taking into account back doors) makes any sense:[ http://www.leedscarerecord.org/faqs/](http://www.leedscarerecord.org/faqs/). In addition, patients have the right to their own data and privacy which means they can do whatever they want with it. Just because clinicians have taken the hippocratic oath to save the world doesn't mean people can't do whatever they want with their data. Elon musk can break down rocket science and solar energy in a pretty simple way. Why can't the same be done with healthcare? Current laws in all 50 US states only mandate organizations to hold onto health care records for 3-7 years. Explain that because it makes no sense to me. Labs can only be ordered after a diagnosis code to bill to insurance. What about preventative medicine and getting work on your blood whenever you want? [http://truecostofhealthcare.org/](http://truecostofhealthcare.org/). Emergency hospital visits occur and patients have no idea how much it costs until a bill is sent to their house which is nothing near what hospitals are reimbursed for (around 25%). There is the classical paternalistic culture in medicine because of all this "formal schooling". The asymmetry in information between clinicians, insurance, pharmaceutical, hospitals and the patient consumers are vast. That's when checks and balances break down. Medicine is not that difficult. Complicated and complex because of the science or the other moving parts? Elon Musk is working on a spaceship to Mars. I'd really like to hear "non-clinical" people chime in because they have more perspective on the problem since they are the end users/ "patient consumers" in my eyes. I still haven't heard many pros or cons about the necessity for the archetype AND the template methodology. --- ## Post #13 by @tonyshannon Best of luck Danny --- ## Post #14 by @Ignacio_Valdes Summary of the VistA's: 1) FOIA VistA Freedom of Information VistA very raw from the government. Should probably avoid.2) WorldVistA sponsored by 501c3 organization WorldVistA derivative of FOIA that has been worked on for many years to make it usable in the private sector. 3) Astronaut VistA (my company Astronaut, LLC) worked on for many years easy to install yum and apt-get based WorldVistA with lots of necessary clinical space refactors, modifications and enhancements that WorldVistA doesn't have. 4) OpenVistA sponsored by Medsphere corporation, derivative of FOIA 5) vxVistA sponsored by DSS corporation derivative of FOIA. 6) OSEHRA VistA public-private VistA corporation derivative of FOIA supposed to be education only oriented. -- IV --- ## Post #15 by @Joseph_Dal_Molin Danny, I echo most of Tony's comments\.\.\. to me openEHR is focused on getting the DNA right so that interoperability comes naturally\.\.\. while VistA is a public domain and open source EHR solution that could benefit from some degree of genetic re\-engineering\. So could every other EHR and health care solution for that matter\. To Tony's point on refactoring etc\. the VA is investing close to $200 M upgrading VistA levarging technology like java script, node\.js etc\. WorldVistA is a charitable, non\-profit, incorporated in 2002, which established the open source community for VistA, ported VistA to a full open source stack, and is the steward of WorldVistA EHR which is the dominant version of VistA used outside the US \(e\.g\. Jordan's national health system, Mexico IMSS and 12\+ hospitals in India\) The Hardhats forum is the place to explore VistA if anyone interested in learning more about VistA: http://groups.google.com/group/Hardhats The takeaway in this thread for me is what and how can communities learn from each other without requiring complete religious conversion? I know we are all extremely busy, but it's remarkable how little cross\-pollination is taking place the open source world\. Cheers, Joseph Joseph Dal Molin President, E\-cology Corp\. Chairman, WorldVistA Tel: \+1\.416\.232\.1206 Skype: dalmolin --- ## Post #16 by @system 200 Million dollar for a software upgrade? Let's say, 2000 high quality programmer years? It does not occur to me as a healthy situation. Is the system so much behind? What is the vision, can't just be implementing some new technologies. It must be a whole new vision, and a new product coming out of that which has nothing much more then backwards compatibility to remind of the old system. I think with 400 high quality programmers in 5 years they should be able to wipe out all the closed source vendors, especially if the product will be developed on non profit base. It looks like a war against the healthcare software industries. --- ## Post #17 by @Joseph_Dal_Molin I agree it's not a healthy situation\.\.\.the amount of money has little to do with the software\.\.\.VistA is consistently rated by doctors and nurses as better than Epic, Cerner etc\. The cost has more to do with procurement and software development practices in the US federal system and EU, Canada, Australia etc\. for that matter\.\.\. plus the size and complexity of the VA itself\. There are many examples of how improvement can be made faster and at a reasonable cost\.\.\. Tony Shannon and Rob Tweed contributed good examples recently\. Having said that \.\.\.\. this is way off topic for this list\.\.\. happy to discuss offline or suggest going to HardHats\. Joseph --- ## Post #18 by @system I see that Veteran Affairs is a government organization\. In Europe it would not be allowed that a government organization would undertake a 200 million dollar attack paid by tax\-money on a vivid market\. One would not expect such an attack on legal and healthy market partners in the USA\. This action must scare away investors and slowdown innovation I am very interested in seeing the source of this rumor\. I just find it hard to believe\. Thanks Bert --- ## Post #19 by @thomas.beale it depends on where you think innovation comes from. One place it rarely comes from is huge corporations (not never, just rarely). - thomas --- ## Post #20 by @system I wouldn't say that generally (many innovations can only be done by huge companies), but besides that. If the government wipes out the market by dumping a massive amount of tax-money to a single party, innovation on the market will be dead, because the market will be gone. These are very basic economics. It would kill competition, and what innovation is, it would also foster corruption, because suddenly you would need some very specific friends. Anyway, I was wondering about the source of this rumor, because I cannot believe that parties on the market find this acceptable. In Europe (maybe except NHS) this would nowhere be possible. In the USA I believe they have similar regulations. Thanks Bert --- ## Post #21 by @system Hey all In a world where a hospital can pay $250 million for a EHR upgrade, 200 million to upgrade all the VA is chump change that will make no difference to the market. And almost all the money will go to legal fees, consultants, requirements analysis, project plans, change control studies, hardware upgrades, security assessments. For 200million, I'd be surprised if there's any money for programming. No Wonder they're going to open source... Grahame --- ## Post #22 by @Philippe_AMELINE > In Europe \(maybe except NHS\) this would nowhere be possible\. In the USA I believe they have similar regulations\. Bert, You can count France as another counter\-example\. The DMP \(Dossier Médical Personnel aka Personal Medical Record or Shared Medical Record or Public Medical Record\) was heavily founded \(something around 1B€\.\.\. and counting\)\. The interesting effect was to shift concerns from "reinventing health care" \(as was in the early 2000s\) to "interfacing the government platform" \(that, by the way, never delivered\)\. Such gov behavior doesn't only kills the market, it profoundly destroys the very nature of innovation\. Best, Philippe --- ## Post #23 by @system In the Netherlands we had a "kind of" experience too\. But it was different, because, the Dutch idea was the government to facilitate the market by rules and standards, and facilitate by an index\-service to find patients in systems\. Under that umbrella \(called LSP\), the market partners of all kind could keep on innovating and selling their products\. So, in my opinion, that is how far a government can go in Healthcare ICT\. Seen from this context, it was a good project\. Facilitating, not taking over\. --- ## Post #24 by @Jan_Talmon Bert What do you mean by your last sentence? I think the situation in the Netherlands was different\. The majority, who had no intention to use the system \(or had no need\) was concerned about the privacy issues, while those who needed it didn’t care as long as their data was available at the right place at the right time\. And we had an opt\-in situation\. So those who had privacy concerns could have stayed out\. Unfortunately, the government had some bad PR around the LSP\. They used the term “National EPD”, which was and still is a misnomer, since there is no such a thing, only on index\-service to find where a specific patient had had medical service\. I would say the current situation where the insurance companies play a role in the ICT infrastructure is more threatening than any privacy issues that may have played a role\. Jan Talmon --- ## Post #25 by @Philippe_AMELINE IMHO, the interesting question to be asked is "what is the reference frame you are working in?" The usual reference frame \(Cartesian\) is centered on the organization, say the hospital, and is fit to tell the story of people passing through \(say "the patients" from in to out\)\. Another reference frame \(Polar\) is centered on the person, and considers the world as "what surrounds Mrs Smith"\. It is fit to tell the life long story of an individual and to organize the work of her \(health\) providers\. The Cartesian reference frame is local and based on a hierarchy of local roles \(homogeneous to a "domain"\)\. The polar one, on the contrary, travels with the person it is centered on\. When creating a Public Health Information System \(PHIS\), the usual \(wrong\) way is to "extend the walls" of a Cartesian reference frame in order to try to cover the whole country\. In my opinion, it is like trying to manage a lake in the same way a fish tank would be kept safe, through a giant air pump, a huge filter and a building sized electric heater\. It seems to me that open systems and closed ones must not use the same reference frame\. In an open world the person herself must be in charge, and privacy issues are more than mandatory, it should be the very starting point of any project\.\.\. at least any project that makes sense to me, both as a developper and a citizen ;\-\) Best, Philippe --- ## Post #26 by @system > Bert > > What do you mean by your last sentence? I think the situation in the Netherlands was different\. The majority, who had no intention to use the system \(or had no need\) was concerned about the privacy issues, while those who needed it didn’t care as long as their data was available at the right place at the right time\. And we had an opt\-in situation\. So those who had privacy concerns could have stayed out\. > Unfortunately, the government had some bad PR around the LSP\. They used the term “National EPD”, which was and still is a misnomer, since there is no such a thing, only on index\-service to find where a specific patient had had medical service\. > > I would say the current situation where the insurance companies play a role in the ICT infrastructure is more threatening than any privacy issues that may have played a role\. > > Jan Talmon Regarding the opt\-in, at the time the LSP was rejected by the senate in 2012 it was defined as an opt\-out system, that was one of the reasons to reject it\. This is one of the things that is improved in the new situation, but as I understand, the opt\-in is very generic formulated, so that there are still privacy concerns\. The LSP was more then an index\-server, it was also an authorization\-service, and that was a problem\. There was also no patient\-accessible logging and there was no fine\-tuning so that patients could authorize who to access their data\. Together with the changing from opt\-in \(which was the original situation\) to opt\-out, people needed to take action to avoid 500\.000 people in the Netherlands having potentially access to their medical data\. I think it is not right to state that people who needed the LSP because of their medical conditions did not care about privacy issues\. They did not have much choice, then to accept it, but that does not mean they agreed with the situation\. They were talked into it\. Although, there are alternatives, the old SOS\-bracelet\. I think it is someway not right to blame the governement bad PR around the LSP as a reason for rejecting it\. This because the senate had very good reasons to reject it, I mentioned some of them above\. It had nothing to do with PR, it was a bad system\. You say that it is threatening that the insurance\-companies now own the LSP, and that is partially true\. But on one hand, the insurance companies already have a lot of access to medical data, they know which medication we take, and they pay our medical specialist bills\. And the insurance\-companies have repaired some serious flaws in the LSP\. You are right, on medical data, there is no escaping from Big Brother\. But on the other hand, the threatening of500\.000 care\-professionals \(your employer via his company\-GP\) having potential access to medical data is gone\. It was always prohibited, it was easy to do when the patient had no logging and no fine\-tuned authorization\-capability\. Now the authorization better refined, a care\-professional can only see data which have relevance to his work, and there is logging which the patient can see\. Also it is again set back to opt\-in\. There are really important improvements, and it remains very strange, that Nictiz, which was a governmental organization, very well paid by the tax\-payer, did not have these quality\-standards, which insurance\-companies now have introduced\. I wonder why Nictiz was so sloppy with our privacy\-issues\. I never understood that\. --- ## Post #27 by @system You are right, Philippe, well expressed\. I think the polar frame is the patient mandate, it means that the patient is responsible for his medical data\. Others write them, but it should not be regarded as poetry, so there is no copyright in the sense of art, but it are just data, like bank\-account\-data\. The writer of the data cannot claim ownership\. The problem is that attribute based access is not very usual, and it is also hard to define generic\. It gets complicated if the spouse or the mother also has access but the father has not because he mistreated the child, and the aunt has, because she is the legal guardian, and the patient changed from GP or hospital, and when they go to a foreign country everything can change, etc, etc\. The problem is when you define a polar access, then you carry around so many changing particularities of a person on which it all depends\. It is not easy to implement\. The only way is to let the patient him/herself make it up\. Give him all authorization\-power, make him/her owner of the data, and let him/her be responsible\. Role based access is easier in algorithms, easier to define and speeds up the system, but role\-based access is always cartesian \(using your words\)\. But role based access gives the power to the institutions, not to the patients\. Bert --- ## Post #28 by @Philippe_AMELINE Bert, I fully get your point\.\.\. but I would argue that, as it goes in maths and physics, choosing a Cartesian or Polar reference frame plainly depends on the kind of issue to be addressed\. As you very well said, role\-based access is always Cartesian since it defines the roles "inside the organization"\.\.\. usually as a matrix\. It is actually the easy way inside an hospital or any care place\. But it becomes messy when multiple unrelated domains are artificially grouped in a large virtual box, as it is done in regional or national wide systems\.\.\. since the role of a given practitioner inside the virtual box cannot be clearly defined\. As an example, as a radiologist, is she allowed to access any image form any care place? My opinion is that the polar reference frame seamlessly "scales in complexity": it is more complicated to operate inside a "hierarchical ordered box" \(as is a single organization\), but becomes natural in a network of actors when considered from one of its nodes \(ie the patient\)\. Best, Philippe --- ## Post #29 by @thomas.beale that's quite simple - it's about making health data computable, which enables it to be used for decision support, risk analysis, personal medicine. Essentially it's the key to keeping medicine affordable for the future. Archetypes = library of re-usable domain data definitions (e.g. vital signs data elements etc) Templates = case-specific data-sets (e.g. diabetic 12 month checkup; specialist referral; etc) (+ Terminology to provide meaning and some inferencing) (+ Ontologies to provide strong inferencing) Without this (and of course other things like guidelines sitting on top), you can't compute with the unavoidable complexity of health information. - thomas --- **Canonical:** https://discourse.openehr.org/t/greetings-and-2-questions/13474 **Original content:** https://discourse.openehr.org/t/greetings-and-2-questions/13474