I then imported the ‘openEHR-EHR-OBSERVATION.fundoscopic_examination.v0’ archetype from CKM as this has boolean default values. Once imported, the defaults were also missing.
Is this a defect in the tool, or am I missing something?
However, more importantly this is an ‘assumed value’ not a ‘default value’. In fact in ADL1.4 there is no such thing as a default value in archetypes, thoughthey can be set in templates.
The idea of an ‘assumed value’ is different - it is is not intended to appear in data, just gives guidance on what might be ‘assumed’ if a datapoint is empty altogether.
e.g in the Blood pressure archetype, if the body position datapoint is not provided, it might be assumed that it is normally ‘sitting’. So this is design-time’ human’ assumption not something that can be computed on.
It is a really tricky concept TBH and I’m am pretty sure it is not used and should be deprecated.
Any decision re the AD tool is for others to decide ( @matijap@borut.fabjan ). I suspect it either needs to be fixed or removed - it does appear to be defective at the moment.
Thankfully those were both in the draft state, but I’ve fixed them both. If you or anyone else come across more examples please add change requests to them
This is certainly an old modelling approach and we, well at least I, have not used this for a long time. I see my name against a not so ancient revision of the fundoscopic exam but clearly missed the hard to find and uncommon assumed value.
Would there be consensus to have ‘assumed value’ removed from the specs, or a least clearly deprecated? Removing would be breaking change in archetypes but not in CDR data.
This also opens up a question about ‘default values’ in archetypes that ADL2 will allow- needs some consideration by clinical community.
