I have not seen much evidence of widespread uptake of HL7v3, indeed Stan and others have said in various places that it has been significantly lower than expected. CDA is the one thing that is getting use. The few large implementations have spent a MOUNTAIN of money to do what they did, and I know for a fact that the outcomes are not seen as good value.
See here for what appears to be a reasonable outline of the status quo - http://www.hl7standards.com/blog/2007/10/10/preparing-for-hl7-v3/ I don't believe changing the RIM, 21090 and other models (apart from CDA) would have that much negative impact on the industry as a whole, but if the changes were radical enough, they could help a lot. I currently don't have time to submit endless feedback to HL7 processes, especially when I know they will not be listened to. I can't imagine that HL7 is going fix its basic modelling methodology, which is what it needs to do. I have actually provided very detailed critiques in the past, and nothing has happened (other than blocking). Today I just have to be concerned with things that are going to be economically implementable by normal programmers, correct and safe. I realise that openEHR still has to solve some things to make that true (mainly to do with better and more openly available Operational Template downstream generators), but at least we don't (for the most part) have models that just cannot achieve interoperability. In openEHR, every single installation of any major version of openEHR, anywhere in the world, is 100% safe for data creation, readability, and interoperability. It is the same schema forever, for all clinical and demographic data, within any given major release. I believe that the openEHR methodology provides a pretty good framework for a) safe data, b) interoperable data, c) data reuse, d) implementable software, and e) being domain driven (via archetypes). I just can't use any HL7 models to do anything useful in the EHR space. - thomas p.s. if v3 was so good and easy, I am pretty sure Stan would have introduced it at IHC. On 25/11/2010 17:31, William E Hammond wrote: > HL7 is following basic modeling procedures in the minds of a lot of people. > HL7 and CDISC, for example, have worked together to produce BRIDG. A large > number of international technologists have and are contributing to HL7. I > agree that RIM has problems. RIM evolved early on from data models. > Decisions were made by a number of people who at that time believed that > was the approach. Stan Huff is leading a TF to look at some of these > issues (Graham is part of that TF). WHat changes will be made? I don't > know. The problem is further complicated in that the current model has > been used in a lot of applications. Thise applications work, even though > many of us believe there is a better way. Those changes have to be made > against an implemented set. I do urge you to submit your criticisms to the > HL7 Technical Steering Committee and to John Quinn, the HL7 CTO. Or to > Stan HUff. * * -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openehr.org/mailman/private/openehr-technical_lists.openehr.org/attachments/20101125/bb084551/attachment.html>
