FHIR Chat · SEARCH PARAMETERS · CCDA / FHIR mapping stream

Stream: CCDA / FHIR mapping stream

Topic: SEARCH PARAMETERS


view this post on Zulip Roney A (Aug 05 2019 at 16:47):

Hi Lloyd, few questions around search parameters:

Q1. How do we resolve the search parameters type discrepancies?

Few search parameters have the input type defined as token, however the data stored in the DB is in the form of String or Boolean. 
For example: In the patient resource, "active" is a boolean field, and the input for the parameter is token type.
Similarly for parameter gender input type is token and the data is stored in the form of string. 

Search Parameter Link: (https://www.hl7.org/fhir/patient.html#search)
Token Format Link: (https://www.hl7.org/fhir/search.html#token)

Q2. How do we compare date and quantity type objects? It is time consuming to compare in the nested JSON fields?

Q3. How we address, for some of the composite functions which require search in two fields along with comparison of quantities?

For example: In MedicationRequest Resource DosageInstruction is a list that has event timings for every object. 

Composite type Link: (https://www.hl7.org/fhir/search.html#composite)
Quantity type Link: (https://www.hl7.org/fhir/search.html#quantity)

view this post on Zulip Grahame Grieve (Aug 05 2019 at 20:12):

you should ask a non-CCDA question like this on the implementers channel.

view this post on Zulip Grahame Grieve (Aug 05 2019 at 20:13):

A1: these are not discrepencies
A2: you don't need to compare date and quantity. (b): yes, it can be time-consuming. All servers prepare indexes in advance to help with this one way or another
A3: prepare indexes

view this post on Zulip Lloyd McKenzie (Aug 05 2019 at 20:54):

To expand on A1 - the type is 'token' because you're looking for an exact match (as opposed to 'string' search parameters where you're looking for a case-insensitive/accent-insensitive starts-with match). Think of the search types as "indexing behaviors", not data types.

view this post on Zulip Roney A (Aug 05 2019 at 23:31):

To expand on A1 - the type is 'token' because you're looking for an exact match (as opposed to 'string' search parameters where you're looking for a case-insensitive/accent-insensitive starts-with match). Think of the search types as "indexing behaviors", not data types.

Thank you Lloyd, and Grahame.


Last updated: Apr 12 2022 at 19:14 UTC