We’ve recently started an informal and practically oriented regular contact with the Norwegian SNOMED CT NRC. One of the things they were interested in discussing was how to use postcoordinated SNOMED CT (expression constraint language) expressions with openEHR, which I know nothing about. Does anyone have any knowledge about or experience with this?
Kind regards, Silje Ljosland Bakke
Information Architect, RN
Coordinator, National Editorial Board for Archetypes
Nasjonal IKT HF, Norway
The UK has been trying to do this for over 20 years without success. It is a terminologists dream but implementers nightmare.
Make a start with high-value use cases e.g Allergy agent “Allergic to + causative agent” - so that you do not have to generate a new Snomed code for every potential allergen.
Perhaps consider laterality. Beyond that, you risk delaying SNOMED CT implementation, as has happened in the UK.
Post-coordination is like nuclear fusion - a damned good idea but tricky to do without blowing everything up.
I mostly agree with Ian, but with the small caveat that for very specific and well-known cases such as body laterality, you just might consider post-coordination on body site e.g.
Yes agree laterality is another potentially high value case.
They should probably talk to the CIMI folks. Stan Huff has a good an idea as anyone of both the potential and pitfalls of complex terminology handling. I suspect that really held CIMI back at a critical moment but at least they have a good understanding of te challenges.
In spite of decades of attempts, I don’t know of any national program that is using post-coordination to any extent.
Hi, just for clarification, you have mixed are two different things:
SNOMED CT postcoordinated expressions are structured combinations of one or more concepts to express a clinical idea. You use them to create new concepts not available in the SNOMED release. They are built using the SNOMED CT Compositional Grammar. In the archetypes they could be used for the semantic binding of the structure (atNNNN codes), or for data values in coded data types
Now, if your question is about how to combine the modeling of archetypes and the modeling of SNOMED CT expressions, this paper could give you some hints. It’s old, but still relevant. There are clear areas for modeling information as archetypes or with terminologies, but there is also a grey area where both solutions are applicable.
“Representing clinical information using SNOMED Clinical Terms with different structural information models”, David Markwell, Laura Sato, Edward Cheetham http://ceur-ws.org/Vol-410/Paper13.pdf
I’m with Ian on this. The only use of post coordination is hard and requires complex tooling support that is way beyond any value you get. I would say even laterality & other qualifiers should go in the information model and not in the terminology if you have the choice.
If you are instead talking about ECLs expression constraint language, then you might want to look at http://ontoserver.csiro.au/shrimp/ecl which provides a UI for constructing ECL expressions and evaluates them as well. It also has links to summary documentation and sample expressions.