2015-10-15 Patient Care FHIR Call
|Patient Care FHIR Resources Conference Call
Location: Conference Call
Time: 5-6:30pm ET
|Facilitator||Michelle M Miller||Note taker(s)||Michelle M Miller|
|X||Michelle M Miller||Cerner|
|X||Elaine Ayres||NIH/Department of Clinical Research Informatics|
|X||Stephen Chu||Queensland Health|
|X||Rob Hausam||Hausam Consulting LLC|
|.||Laura Heermann-Langford||Intermountain Healthcare|
|X||Craig Parker||Intermountain Healthcare|
|X||Russell McDonell||Telstra Health|
|X||Viet Nguyen||Systems Made Simple|
|X||Lisa Nelson||Life Over Time Solutions|
|X||Lloyd McKenzie||Gevity (HL7 Canada)|
|X||Iona Thraen||Dept of Veterans Affairs|
|X||Eric Haas||Haas Consulting|
|Quorum Requirements Met: yes|
- Agenda review
- Approve previous meeting minutes Patient Care FHIR Minutes 2015-9-10
- Motion: Elaine/Emma Abstain - 3, Negative - 0, Approve - 7
- Review DSTU 2.1 timeline and frozen resources
- Recap Clinicians on FHIR
- Adverse Events - new resource?
- Protocol - new resource?
- gForge change request
DSTU 2.1 Timeline
FHIR Resources owned by Patient Care
- AllergyIntolerance (1) -- frozen
- Condition (1)
- Procedure (1) -- frozen
- CarePlan (1)
- Goal (1)
- Referral Request (1)
- Questionnaire (1)
- Questionnaire Answer (1)
- FamilyMemberHistory (2) genomics / Jonathan Holt – Vanderbilt - interested in contributing to this resource.
- Procedure Request (2)
- Flag (2)
- ClinicalImpression (3)
- Communication (3)
- CommunicationRequest (3)
Note: Contraindication and Risk Assessment are owned by CDS, not Patient Care
Note: Substantive changes to frozen resources in DSTU2.1 are prohibited unless FMG explicitly approves the substantive change
Review DSTU 2.1 timeline and frozen resources
AllergyIntolerance and Procedure are frozen
Condition is not frozen (need to resolve what was formally known as negation)
Anything that was not frozen (e.g. condition, clinical assessment), then it is a good use of time to polish.
From a prioritization perspective:
- Normative candidates
- For resources that are frozen, doing non-substantive changes to get them ready for quality criteria and moving up maturity levels.
- For draft or new resources, then be informed by what implementers are saying they need/want. Ping implementer community for feedback to help us prioritize.
Recap Clinicians on FHIR
Viet/Stephen's table was focused on family medical history. Good level of consistency. Created patient in advance. Didn't test extension for condition or parent/child. Stephen said that the negative family history wasn't well tested.
Russ's table tested MedicationOrder resource, which can contain a Medication resource, but didn't get to AllergyIntolerance. Found and reported glitches in the tool around timing. Went well. The Medication value set had substances, but not products. If a medication was ordered for a problem on the problem list (e.g. cough), but then the condition changes to something else (e.g. anxiety). Need to reference a "static" version of the condition that was relevant at the time of order.
Emma's table tested Procedure and CarePlan. A few tooling issues (saving to local cache instead of a server). Discussion about how FHIR handles protocols or standing orders.
- Participants from multiple working groups (genomics, pharmacy, patient care) helped go deeper and answer questions. Having other working group participants was worthwhile.
- Prep to build underlying resources (e.g. Patient)
- Helpful to have a technical person at the table to navigate resources.
Negation discussion on Friday. Ideas documented October_2015_WGM_Atlanta:_Oct_4_to_Oct_9#Patient_Care_Friday_Q3. If trying to make a statement "I didn't observe condition" -- discussion about whether to use lists, status, or value on Condition to convey this. Subtle differences between 'never existed' versus 'doesn't exist at this point in time' -- could be represented as an observation since it is a point in time assertion and could change in time.
Condition can be used for an admitting diagnosis (because it is never expected to change) and conditions can also be on-going. Use condition for signs, symptoms, and diagnosis. Condition boundaries need clarification. Lloyd will log a change request.
"Have you ever had high blood pressure?" is not a Condition.
Look for existing SNOMED codes (e.g. No Known Allergies, No Known Problems) to prevent double negatives. Ongoing discussions about how much should be handled via terminology.
Stephen is working on a negation wiki, which could get renamed: Representing_Negation. Structured Doc, OO, Pharmacy, Genomics, CDS are all interested in participating. Depending on how much FHIR DSTU 2.1 volume there is, then this topic will be discussed next Thurs, Oct 22 (and Tues, Oct 27 if needed).
Negation lessons from CDA:
- Given that the structure of observation has both a code and value, negation needs to convey whether negating code or value?
- Need to negate all act classes with same approach (it was a mistake not to think through supply or encounter)
Adverse Events - new resource?
Next step is Elaine looking at the equivalent ISO 27953 standard
Protocol - new resource?
Concepts of diagnostic protocol or evaluation protocol might be overlapping with clinical assessment.
Treatment protocols that go beyond order sets.
Russ believes there is merit in discussing protocols. Lloyd also believes protocol will come up as workflow as discussed further. More discussion about whether one single protocol resource or multiple (specialized) protocol resources are needed as well as how to link orders to protocols. Next step is to gather use cases in preparation for the upcoming workflow conference calls, including oncology.
FHIR notion of order is any kind of a request for action to occur. It is not limited to clinicians only. Trying to move away from "order" naming to "request" naming.
gForge Change Requests
46 change requests, 15 tied to clinical impression, the rest are new (some will be added to a block vote)
http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=8792 -- feedback for MnM is that 'note' has another meaning in health care and risks confusion; Discuss this as part of the general ClinicalImpression review. Next step: stakeholders can collaboratively edit this document to define note, comment, and annotation. Then, we can assess whether that semantic difference manifests in systems or not.
- Notes are....
- also known as clinical note or document; could be unstructured or structured; could require review and authentication; likely has a subject or type of note assigned to enable retrieval of note; could follow the SOAP format
- Comments are....
- unstructured, free form text that applies to the entire entity (e.g. condition, allergy, medication order, etc.). A comment is typically comprised of an author, date/time, and text only. Typically, comments don't have a status or title. Comments don't standalone, so there is typically no title of the comment or need for additional workflow to review or authenticate the comment independent from the overall entity's status.
- Annotations are....
- unstructured, free form text that is context specific with a given entity (e.g. annotated condition code display or annotation within a flowsheet or wave form for a specific point in time or time frame)
Adjourned at 7:09pm ET
|Next Meeting/Preliminary Agenda Items|
© 2012 Health Level Seven® International. All rights reserved.