feature: Group by and having statement in FHIR Path

[mertgencturk]

Feature Request

Describe the Feature Request
I would like to have a list of patients having more than two chronic conditions. I am able to retrieve conditions from repository through FHIR query. What I need more is to group conditions by their subject and keep the ones having at least two condition resources.

Is your feature request related to a problem? Please describe
No.

[tnamli]

We have implemented a FHIR path function with the following signature;

groupBy(groupingExpr: expression, aggregationExpr: expression)

The 'groupingExpr' is an expression that returns a key value for each item in the collection. Then items in the input collection are grouped based on the equality of these keys.

The 'aggregationExpr' is an expression that returns an aggregation value from the grouped items. The result is a JSON array that contains a JSON object for each group with the following elements;

• "bucket": Where the value is the key value extracted by groupingExpr for corresponding group
• "agg": Where the value is the aggregated value calculated by the aggregationExpr for the corresponding group

We have also implemented the following aggregation functions for ease of use;

sum(valueExtractorExpr: expression)
Extract the numeric value (Fhir Path Number) by appliying the valueExtractorExpr on each item in the input collection and sum the results.
e.g. Return the sum of observation values within a bundle for each patient
Bundle.entry.resource.filter($this is Observation).groupBy($this.subject.reference, sum(valueQuantity.value))

avg(valueExtractorExpr: expression)
Extract the numeric value by appliying the valueExtractorExpr on each item in the input collection and take the average of results.

min(valueExtractorExpr: expression)
Extract the numeric value by appliying the valueExtractorExpr on each item in the input collection and take the minimum.

max(valueExtractorExpr: expression)
Extract the numeric value by appliying the valueExtractorExpr on each item in the input collection and take the maximum.