2019-07-09 Weekly Call Meeting Agenda

Time

Agenda item

Discussion Points

Who

Outcome/Follow-up

3:00 –3:15

Richard Esmond Check In

·       CIMI Feedback?

·       Submit to HL7: Process and Next Steps

·       Prep for CDS WG Meeting 7/10/19

Raja, Richard, All

Raja, Ioana

Raja, All

HL7 Ballot Process Details

3:15 –3:40

Decisions that we need to make by 7/15/19

·       Measure Semantic

Raja, All

AI:

Definition of measure semantic: (derived from Glip)

The measure semantic is a concept that represents the [intrepretation] of the numeric result. It is a codeable concept.

Q:

What value(s) goes into each Measure Semantic?

Suggestion:

For the purposes of the informative ballot, be generic? – What generally goes in this attribute?

AI: Text about cognitive dissonance (non-overlapping magistrate):  Cognitive dissonance is the mental discomfort experienced by a person who holds two or more contradictory beliefs, or values. This discomfort is triggered by a situation in which a person’s belief clashes with new evidence perceived by the person.(e.g.,  using a count type units of measure to describe diabetes present...)

Naming Measure Semantics:

·        Measure.semantic =

·        Countable quantity

·        Countable

·        Count

·        Occurrences

I'd like to propose we identify a few narrative use cases to tackle this question of what to name the semantic, and come up with some criterion about how we evaluate the suitablity of a choice. Then we put this in the ANF document, so show the reasoning, and then if they have others, they can try and apply them to the same criterion, and see if they come up with the same result.

Using Measure to specify that a text results exists?

-        No, TW/RC think that measure is numeric whereas “text” is in the Topic not measure. Unstructured circumstance

Intervals:  interval.lowerBound, interval.lowerBoundPresent, interval.higherUpperBound, and intervalUpperBoundPresent

Infinity rather than null…

We prefer x, we understand FHIR doesn’t support x, and present strengths and weakness of each. These are somethings that would need to change for ANF to do it the way we want.

FHIR not supporting IEEE754: This is a deficiency of FHIR we need to address seperately. Their decimal representation should really be IEEE 754. We can call that out in our informational ballot.  It is ok at this point, as we are informational. And we may have to give specific implementation advice to bridge the gap unti we can get the datatype added to FHIR R5...

Ioana: Infinity is different but I'm struggling with its representation in a Interval structure: I need to give both upper and lower bound a value and I don't want to use INF or -INF since they will not be available to implementers. Typically, if a upper limit is unknown, the value is left blank. "unbound"

Ordinal measures: Represented in Topic

3:40 –3:50

Reference Model

·       Review latest changes

Ioana, All

3:50 –3:55

Discussion

·       Narratives for Appendix

·       Textual Examples:

o   Formatting to help highlight what we are exemplifying

Tim

Raja

Continued Discussion

3:55 –2:00

Recap

·       Major Decisions & Action Items

Raja