Stream: implementers
Topic: Defining derived Observations
Rob Dingwell (Oct 12 2020 at 13:01):
Does anybody know of any work done in the area of defining derived Observations?
For example, take the concept of BMI. BMI is a derived value based off of height and weight and this value could be calculated using something along the lines of CQL. Is there a way to define not the specific instance of a BMI Observation, but more along the lines of a definition of the Observation along with the CQL that would be used to calculate the value for a specific instance?
I see ObservationDefintion in the spec and would imagine that applying an extension or 2 could enable this type of use. Is there any reason not to use ObservationDefinition for this? Has anyone done anything similar to this using ObservationDefintion or another means?
Lin Zhang (Oct 12 2020 at 13:51):
What do you want to define? The code/concept of BMI, or its calculation formula?
Related Links:
https://cql.hl7.org/examples.html
https://cds.ahrq.gov/cdsconnect/authoring
https://github.com/ahrq-cds/ahrq-cds-connect-authoring-tool
Rob Dingwell (Oct 12 2020 at 14:03):
I understand how to define CQL so that is not the issue. What I'm looking for is a way to define a template for an Observation along with the CQL the would be used to populate the value for the Observation. ObservationDefinition seems like this would be a logical fit as I could define the code/concept of an Observation and add an extension to it to define the CQL that would be used to derive the value.
Rob Dingwell (Oct 16 2020 at 16:31):
Does anybody see a reason why ObservationDefinition shouldn't be used like this?
Mark Kramer (Oct 16 2020 at 21:32):
In terms of representing the answer, unfortunately Observation.derivedFrom is limited to Reference(DocumentReference | ImagingStudy | Media | QuestionnaireResponse | Observation | MolecularSequence). That would work for your BMI example, but would not work for many other derived Observations.
Last updated: Apr 12 2022 at 19:14 UTC
