Open Sigfried opened 1 month ago
There is a distinction between id
which is an internal system identity column (e.g. an auto-incrementing id, a primary key) that might change if the system changes, and identifier
which is a globally unique id that will be consistent across systems (e.g. SSN). The variables are used in the way they are defined in the corresponding FHIR properties of id
and identifier
. As such the OMOP condition_occurence_id
is an example of an id
, not an identifier
.
relationship_to_participant
is underspecified currently, but should take a range of an enum that has values like Self
, Mother
, etc. This is to allow for the capture of both family history of a condition (where the relationship_to_participant
might be Mother
), and the individual's one history of a condition (where the relationship_to_participant
would be Self
)
yes, these should be consistently modeled and using slots for all occurrences makes most sense.
Referring to the yaml here.
identity
slot is a primary key for a given Condition record. Equivalent to OMOP'scondition_occurrence_id
.relationship_to_participant
attribute says it indicates "relationship between the Participant to which the Condition is attributed and the individual who had the reported Condition," but I don't see any attribute referring to the individual who had reported the condition. I don't think this attribute makes sense without that.associated_participant
is listed as an attribute. It is also the name of a slot and it appears as a slot in the Demography and Visit entities. Should it also be a slot in Condition?