Silje,
very nice analysis. Points 3 and 4 are a 'not applicable' idea, which
needs to be catered for as such I think.
- thomas
On 08/06/2016 14:50, Bakke, Silje Ljosland wrote:
Hi Gerard,
I don’t have time to look through the use cases right now, but I
thought I should give a general outline to how we handle negation in
openEHR archetypes.
Basically, there isn’t a single way that works for all use cases, but
we’ve worked out a few patterns that seem to work:
·Some concepts are very safety critical, where you should never be
able to confuse positive presence with negation. Examples of these are
medications, adverse reactions, problems/diagnoses, family history,
procedures, in some cases symptoms, and possibly implants. Negation of
these need to be handled by separate “Exclusion of X” archetypes,
which makes it impossible to query for something and inadvertently mix
up presence and absence.
oSymptoms are a special case, where we have both the ability to use an
explicit negation using the “Exclusion of a symptom” archetype, or the
softer “no more than usual” “Nil significant” boolean element.
·Some concepts aren’t as safety critical, or are more of a specific
status that should be updated in one single place of the health
record. An example of this is smoking status, where both “Former
smoker” and “Never smoked” could be seen as negations of “Current
smoker”. This is handled as a value set in the main data element of
the smoking summary archetype.
·Specific physical examinations sometimes need to be excluded, to be
able to say “I didn’t perform the examination of the left eye because
the patient has an artificial left eye”. This is handled using a
separate cluster that’s reused within each examination cluster.
·Specific observations or scorings sometimes need to be excluded, to
be able to say “Children’s Global Assessment Scale wasn’t performed
for this child, because they’re younger than 4 years old”. This is an
emerging pattern, but for now it looks like it’ll be similar to the
examination exclusion, using a separate cluster that’s reused between
the observations. We still don’t know for sure what to call this
cluster, as “Examination of observation” can easily be mixed up with
the archetype class OBSERVATION.
·There’s also the issue of how to specify the presence or non-presence
of something in the context of an examination of the same thing. A
good example of this is body fluids, which archetype is out for review
right now. If anyone would like to have their say on this and isn’t
invited to this review, please adopt this archetype to be invited:
http://openehr.org/ckm/#showArchetype_1013.1.2255
I might have forgotten something, but I think these are the basic
patterns of negation/exclusion we’re using/exploring as of now.
_______________________________________________
openEHR-clinical mailing list
openEHR-clinical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-clinical_lists.openehr.org
