FHIR Chat · OO Block Vote for Thur Sept 13th · Orders and Observation WG

Stream: Orders and Observation WG

Topic: OO Block Vote for Thur Sept 13th

---

Eric Haas (Sep 04 2018 at 03:06):

OO FHIR Block Vote for Thursday Sept 13th OO Conference Call. Let me know if you would like to remove any item from the block for discussion.

Comment Submitters

• Julie Evans
• Lloyd McKenzie
• None
• Richard Townley-O'Neill
• Rick Geimer

Line Items

• GF#17571 The profiles of VitalSigns do not have their own names or titles (Richard Townley-O'Neill)
• GF#17654 Task.instantiatesReference should be instantiatesCanonical (Lloyd McKenzie)
• GF#17385 Extend Observation by adding transaction type (Julie Evans) Not Persuasive
• GF#17244 DiagnosticReport%3A change codedDiagnosis to diagnosis%5Bx%5D (Rick Geimer) Not Persuasive with Mod
• GF#17468 SupplyRequest.identifier -%3E 0..%2A and improve definition (Lloyd McKenzie) Persuasive
• GF#17677 SupplyRequest should have deliverTo search in Patient compartment (Lloyd McKenzie) Persuasive
• GF#17717 vital-signs profile rule vs-1 too strict and technical error (None) Persuasive
• GF#17516 Extend Observation by adding test type (Julie Evans) Persuasive with Mod
• GF#17636 In BodyHeight profile include guidance about body length (Richard Townley-O'Neill) Persuasive with Mod

Christine D (Sep 06 2018 at 13:25):

@Eric Haas and @Lloyd McKenzie I am not sure I understand the proposed solution for 17516. I would like @Angie Romano to look at this. I believe she has created an extension that meets our needs. What would be the process for reviewing Angie's extension and adopting that for this tracker?

Angie Romano (Sep 06 2018 at 13:41):

I have created a test extension that can be used. I intended to use it for Connectathon demonstration purposes only. I completely understand @Lloyd McKenzie comments that this is very specific to Clinical Trials and while we could get an extension created, adoption of said extension might be a challenge. Would it make more since for the extension to be a higher level resource like Encounter, or Diagnostic Report? All the results associated with a panel should fall into the same test type category. There might be a way for us to derive it too if we know what tests are planned as part of the study and at what visits.

Christine D (Sep 06 2018 at 13:50):

@Angie Romano wanted you to be aware of this block vote. I think we had discussed the need to know whether the data being sent was an update. There is some information in the tracker RE: looking at the section on FHIR Rest. Are we ok if we don't have this?

https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=17385

Angie Romano (Sep 06 2018 at 13:52):

@Christine D yes we will have to derive it. There is no way for the FHIR bundle to know. Some of the records could be new to us and some could be updates. That is something we will have to track on our side.

Lloyd McKenzie (Sep 06 2018 at 14:41):

@Angie Romano Keep in mind that context doesn't conduct across resources. An element declared on an Encounter or DiagnosticReport only applies to that resource - not to any Observations related to them. If you want something to be true for an Observation, it must be declared explicitly on the Observation. For example, if you report a panel of Observations, you declare the patient, lab, order, etc. on each one - you can't inherit from the panel Observation. And in this case, it would have to be on the Observation because one Observation might be tied to multiple studies with different values for each.

Eric Haas (Sep 06 2018 at 15:28):

@Christine D Can you suggest an edit to the tracker disposition? If not I will pull from the block ( scheduled for next Thursday)

Christine D (Sep 06 2018 at 16:12):

@Eric Haas is the disposition question referring to 17516 or 17385?

Eric Haas (Sep 06 2018 at 16:15):

GF#17516 you had not commented on GF#17385 disposition.

Christine D (Sep 06 2018 at 23:14):

@Eric Haas Does the extension work similar to this? (this is the one we had created) If so, then I am good with the modifications on 17516.
{
"resourceType": "StructureDefinition",
"id": "StudyTestType",
"meta": {
"versionId": "2",
"lastUpdated": "2018-09-05T16:46:24.978+00:00"
},
"extension": [
{
"url": "http://clinfhir.com/fhir/StructureDefinition/cfAuthor",
"valueBoolean": true
},
{
"url": "http://clinfhir.com/StructureDefinition/userEmail",
"valueString": "romano_angela_d@lilly.com"
}
],
"url": "https://fhirtest.uhn.ca/baseDstu3/StructureDefinition/StudyTestType",
"name": "StudyTestType",
"status": "draft",
"kind": "complex-type",
"abstract": false,
"contextType": "resource",
"context": [
"*"
],
"type": "Extension",
"baseDefinition": "http://hl7.org/fhir/StructureDefinition/Extension",
"derivation": "constraint",
"snapshot": {
"element": [
{
"id": "Extension0",
"path": "Extension",
"label": "CodeableConcept0",
"definition": "No Description",
"min": 0,
"max": "1",
"base": {
"path": "Extension",
"min": 0,
"max": "1"
}
},
{
"id": "Extension.url0",
"path": "Extension.url",
"representation": [
"xmlAttr"
],
"label": "CodeableConcept0",
"definition": "No Description",
"min": 1,
"max": "1",
"base": {
"path": "Extension.url",
"min": 1,
"max": "1"
},
"type": [
{
"code": "uri"
}
],
"fixedUri": "CodeableConcept0"
},
{
"id": "Extension.value[x]0",
"path": "Extension.valueCodeableConcept",
"label": "CodeableConcept0",
"definition": "No Description",
"min": 0,
"max": "1",
"base": {
"path": "Extension.valueCodeableConcept",
"min": 0,
"max": "1"
},
"type": [
{
"code": "CodeableConcept"
}
],
"binding": {
"strength": "required",
"description": "No Description",
"valueSetUri": "https://api-v5-r4.hspconsortium.org/R4ClinicalResearch/open/ValueSet/Study-TestTypeVS"
}
}
]
}
}

Eric Haas (Sep 06 2018 at 23:56):

@Christine D it does more, if the code is S the it can indicate via a reference indicates the studys it applies to.

Christine D (Sep 07 2018 at 00:02):

@Eric Haas I like that. Thanks!

Angie Romano (Sep 07 2018 at 13:00):

@Eric Haas and @Christine D this is great, but I do have a small question. We have an extension, http://hl7.org/fhir/StructureDefinition/event-researchStudy, that allows us to add a reference to ResearchStudy. Are you saying that the test type extension would also include the ability to add a reference to ResearchStudy? Is it OK to have more than one place to add the reference?

Eric Haas (Sep 07 2018 at 17:55):

That is a good point. I don't we should duplicate it. @Lloyd?

Lloyd McKenzie (Sep 07 2018 at 18:01):

They're saying two different things. One is saying "this Observation is relevant to study x". The other is saying "this Observation was planned as part of study X (or not planned as part of study X)". If you feel that you're always going to have the "was this planned as part of the study?" for all types of event resources (procedures, conditions, encounters, allergies, etc.) we could drop that extension and generify the other. Alternatively, perhaps some resources should have one and some have the other. But my guess is there will often be a need to declare "relevance" without commenting on whether it was planned or not. And having a consistent way to capture information across the resources is nice, even if sometimes that results in redundancy.

Eric Haas (Sep 07 2018 at 22:02):

If:

• extension for type : code = 'due to study' + extension event-researchStudy

means "this Observation is part of studyX and was planned (ie, due to) as part of studyx"

• extension for type : code = ' NOT due to study' + extension event-researchStudy

means "this Observation is part of studyX and was NOT planned (ie, not due to) as part of studyx"

• Is there a need for "this Observation is part of studyZ and was NOT planned as part of studyX but is still relevant to studyZ" ??

Lloyd McKenzie (Sep 07 2018 at 23:16):

I guess we could put an extension on event-researchStudy that says "due to study" / "not due to study", but it would have to be inside the event-researchStudy extension because a single event could be relevent to multiple studies

Angie Romano (Sep 12 2018 at 12:50):

The third scenario you're describing @Eric Haas would be rare, but I'm sure it could happen.

Then that puts us in the extension within an extension category as Lloyd suggested.

@Christine D

Angie Romano (Sep 12 2018 at 18:01):

@Lloyd McKenzie @Eric Haas how long would it take to get the extension created within the http://hl7.org/fhir/StructureDefinition/event-researchStudy extension? Is that quick or should we try to mock the extension up for the Connectathon?
@Christine D

Lloyd McKenzie (Sep 12 2018 at 18:48):

The extension is already defined in R4. The URL doesn't need to resolve for it to be a legal extension. When R4 becomes an official release, the URL will resolve and will continue to resolve from that point forward unless we remove it in some future release. Even if we did remove it from a future release and the URL stopped resolving, it would remain a valid extension for the release(s) it was valid for.

Angie Romano (Sep 12 2018 at 19:20):

@Lloyd McKenzie I see the event-researchStudy extension in the R4 Ballot #2 list of extensions, but not with the embedded additional extension for type:code='due to study' or 'NOT due to study'.

Lloyd McKenzie (Sep 12 2018 at 21:33):

Right. That hasn't been defined yet. And it will not appear on the website with that extension, it'll appear as a separately defined extension - though in instance, it'll appear one nested within the other.

Christine D (Sep 13 2018 at 15:11):

@Lloyd McKenzie Do we have an estimate on when the new piece will be created? Wondering whether we would need to mock up an extension for the Connectathon. @Angie Romano

Lloyd McKenzie (Sep 13 2018 at 15:32):

We can do it this week as a pre-applied change if we think we're getting consensus on this as being the correct solution.

Eric Haas (Sep 13 2018 at 18:33):

From OO perspective I would like to know where we have landed on GF#17513. My understanding is that we going add to the existing http://hl7.org/fhir/StructureDefinition/event-researchStudy extension and thus is the new proposed extension necessary?

Angie Romano (Sep 13 2018 at 18:48):

I may be getting lost in the string of comments. GF#17513 is for Planned Collection time, correct? We are planning to use existing extension http://hl7.org/fhir/StructureDefinition/cqf-relativeDateTime. Do you mean GF#17516 for test type? @Christine D please weigh in here.

Eric Haas (Sep 13 2018 at 19:11):

yes test type GF#17516 typo on my part.

Christine D (Sep 13 2018 at 19:47):

@Angie Romano

Christine D (Sep 13 2018 at 19:49):

Yes for GF#17516 I believe we were going to add an extension within the event-researchStudy extension to handle the test type. @Eric Haas @Lloyd McKenzie @Angie Romano

Eric Haas (Sep 13 2018 at 19:50):

OK to withdraw this tracker?

Christine D (Sep 13 2018 at 19:51):

Do you need the tracker to document the reason for the extension?

Eric Haas (Sep 13 2018 at 19:54):

Great question! @Lloyd McKenzie whose extension is this .... FHIR-I or should OO keep it and make the changes to the extension.?

Lloyd McKenzie (Sep 13 2018 at 20:08):

Owner for it would most ideally be BR&R. (Though it needs to be called something other than "test type")

Christine D (Sep 13 2018 at 20:13):

Hi @Hugh Glover , we had a mapping item (test type) that we identified as a gap between LAB and FHIR. The proposal is to add an extension to the event-researchStudy extention (new extension defined to allow us to tie an observation to a study to show "this observation was part of this study". I think we have two questions: ownership of the extension for test type and a new name for it (test type isn't very descriptive).
@Angie Romano

Eric Haas (Sep 18 2018 at 16:24):

@Lloyd this is an event extension so I can do it ( If I dare ever try GIt+build again) unless is more appropriate to punt

Lloyd McKenzie (Sep 18 2018 at 17:06):

This is an extension on an event extension - and it's only relevant to BR&R. Plus it's good for them to have practice :)

Eric Haas (Sep 18 2018 at 17:17):

punted....

Christine D (Sep 25 2018 at 14:27):

@Eric Haas and @Lloyd McKenzie I don't see event-researchStudy on the list of extensions and the URL above is giving me a 404 error. Was that extension pulled or am I just not seeing it?

Lloyd McKenzie (Sep 25 2018 at 14:54):

Look here: http://build.fhir.org/extensibility-registry.html

Lloyd McKenzie (Sep 25 2018 at 14:54):

The URL for new extensions won't resolve until R4 is published

Christine D (Sep 25 2018 at 16:21):

Thanks @Lloyd McKenzie ! @Hugh Glover , we have the list of extensions: http://build.fhir.org/extensibility-registry.html

Hugh Glover (Sep 25 2018 at 19:17):

Got it!

---

Last updated: Apr 12 2022 at 19:14 UTC