Difference between revisions of "2017-01-27PC CIMI POC Call Minutes"

Latest revision as of 16:37, 27 January 2017

HL7 PC-CIMI-POC Meeting Minutes Location: Phone		Date: 2017-01-27 Time: 10:00-11:00 CT
Facilitator	Jay Lyle	Note taker(s)	Jay Lyle

Attendee	Name	Affiliation
y	Richard Esmond	PenRad
y	Galen Mulrooney	JP Systems
y	Jay Lyle	JP Systems / VA
y	Larry McKnight	Cerner
y	Susan Matney	Intermountain
y	Joey Coyle	Intermountain
	Gay Dolin
	Ash Davison	Intermountain
	Laura Heerman Langford	Intermountain
	Claude Nanjo
	Rob McClure

Agenda Topics

Minutes/Conclusions Reached:

Preliminary discussion: how to make CIMI artifacts simple enough for users
Question of model bindings: where to they go
- Evaluations have question codes
- Assertions have values
- Assertion properties: semantic binding is metadata. Can/Should it be in instances?
  - Body site: FHIR property is semantic equivalent of LOINC body site, and of SCT Finding site in finding context
    - So we can leave the LOINC code out? Is there any recordkeeping or administrative reason to have it?
    - Yes, for ordered panels, not so much for sub-panels?
Question of structure
- Site has a code or an associated structure
- How much do we need to worry about it here in the requirements stage vs design?
Cardinality
- Not just Tunneling but other properties may need to be 0:N, e.g., wound edge appearance, tunneling
- If so, do we need to fragment body site for respective bits? Is this too much detail?
- Tunneling has clock position: do we need some other way to designate sub-body-site? To assign an identifier?
Clinicians want these properties to be their own Evaluations
- maybe not Evaluation class, but to carry both question and answer.
Model details
- Remove Stage from pressure ulcer
- Rename to Pressure Injury?

Actions

Next Meeting/Preliminary Agenda Items

@@ Line 112: / Line 112: @@
 '''Minutes/Conclusions Reached:'''<br/>
-* See Patient Care
+* Preliminary discussion: how to make CIMI artifacts simple enough for users
+* Question of model bindings: where to they go
+** Evaluations have question codes
+** Assertions have values
+** Assertion properties: semantic binding is metadata. Can/Should it be in instances?
+*** Body site: FHIR property is <b>semantic equivalent</b> of LOINC body site, and of SCT Finding site in finding context
+**** So we can leave the LOINC code out? Is there any recordkeeping or administrative reason to have it?
+**** Yes, for ordered panels, not so much for sub-panels?
+* Question of structure
+** Site has a code or an associated structure
+** How much do we need to worry about it here in the requirements stage vs design?
+* Cardinality
+** Not just Tunneling but other properties may need to be 0:N, e.g., wound edge appearance, tunneling
+** If so, do we need to fragment body site for respective bits? Is this too much detail?
+** Tunneling has clock position: do we need some other way to designate sub-body-site? To assign an identifier?
+* Clinicians want these properties to be their own Evaluations
+** maybe not Evaluation class, but to carry both question and answer.
+* Model details
+** Remove Stage from pressure ulcer
+** Rename to Pressure Injury?
 ===Meeting Outcomes===