Author vs. Informant
MSpielman
We're working through some ambiguities in the definition of author
vs. informant and its use throughout our CCD imports. Would appreciate
your thoughts on the best approach.
The best example of this is with medications; the information pieces
we are looking to populate is:
Entering Person (person who entered order into the sending system)
Entering Time (when the order was entered into the sending system)
Entering Place (where the user was located)
In the past, we have looked to <author> for the information about a
person and <informant> for information about an entity. So in the
examples above, we would look to:
Entering Person - substanceAdministration/author
Entering Time - substanceAdministration/author/time
Entering Place - substanceAdministration/informant/
representedOrganization
However, this appears counter-intuitive and duplicative... It seems
like we should really be pulling all three from either an <author> or
an <informant> as both structures are capable of representing all
three data points. If so, the question is, which one? Both can be
present for any clinical statement within a CCD according to the CDAR2
and PCC specs although author has a 1 or more cardinality vs informant
which is 0 or more -- which is moot because we can inherit from a
parent as well.
So the real questions are:
1) In the scenario above, or in similar scenarios in other content
modules which approach is correct to **accurately** represent all 3
data points? author, informant, or author+informant?
2) Is there anything resembling a formal explanation of the difference
in meaning between an <author> and an <informant> structure? (I can't
find one)