Stream: implementers
Topic: Observation.$stats
Grahame Grieve (Nov 26 2016 at 17:46):
There's an operation on Observation called $stats. I've started working on implementing it, which has lead to some updates to it's definition. See http://build.fhir.org/observation-operations.html
Grahame Grieve (Nov 26 2016 at 17:47):
note that my interest is driven by ongoing discussions with openMHealth about what a collaboration might look like.
Grahame Grieve (Nov 26 2016 at 17:50):
As it stands now, the client provides 4 pieces of information that drive the stats;
- the subject id
- a code that identifies the kind of observations to include
- a duration / period for the stats to cover
- a list of stats to generate. The list of stats comes from openMHealth with one addition (total count)
In addition, the client can ask for the observations covered by the stats to be returned (or a limited subset) with the intent that this allows for a representative plot
Grahame Grieve (Nov 26 2016 at 17:50):
comments welcome
Brian Postlethwaite (Nov 27 2016 at 08:51):
The return value comments that it can be the observation directly, and then the example isn't. Should the comment be changed (as it is not a 0..1 cardinality?)
Grahame Grieve (Nov 27 2016 at 09:30):
thought I fixed all those. I'll update the comment
Last updated: Apr 12 2022 at 19:14 UTC
