CGIT concall 20160523

Return to Conformance

Meeting Information

• Date: Monday, May 23, 2016
  • Time: 4:00 pm, North America Eastern Time (New York, GMT-04:00)
  • Phone Number: +1 770-657-9270
  • Pass Code: 644843
  • Web Meeting: https://global.gotomeeting.com/join/647877325

Proposed Agenda Topics

• WGM for Baltimore
• Profiling Multiple Occurrences - question from Craig Newman
• Data Type Flavors
• Todo items from last WGM

Question from Craig Newman

I have a few questions about section 2.B.9.7 (Profiling Multiple Occurrences) of the version 2.8.2 base standard.

The existing immunization HL7 v2.5.1 implementation guide (not currently an HL7 balloted document) assigns meaning to a particular occurrence of a repeating field in several places. For example, the first occurrence of PID-5 shall contain the patient’s legal name and the first occurrence of RXA-9 (Administration Notes) shall contain the information source (using the NIP001 value set). I was under the impression that meaning should not be assigned to an occurrence of a repeating field, so I was aiming to eliminate that language in the next release of the IG, relying on name type code, equipment type code, coding system, etc to indicate the type of data being sent in a given occurrence. But now that I’ve read section 2.B.9.7 I’m not quite sure if this sort of thing is back in fashion.

On the surface, it doesn’t seem like 2.B.9.7 is encouraging assignment of meaning to a particular repetition, but it could be used for that purpose. For example, it seems like you could use the “number” attribute along with a very restrictive value set binding (or conformance statement) to limit the name type code in the first occurrence of PID-5 to L for legal. I’m assuming that while the examples in the base standard really just include Usage, that other profiling attributes such as value sets could also be differentially required. Is that within the spirit of what 2.B.9.7 is trying to accomplish?

Also, is there a mechanism for requiring a particular flavor of occurrence in a field. For example, if you wanted to define different patterns for XTN occurrences for phone number and email address using the “value” attribute, could you also require that every message have a phone number (that occurrence pattern itself has a usage of R) but not require an email address (that occurrence pattern has a usage of RE)?

Running List of topics

• Next WGM
• Data Type flavors
• Delete indicator

Participants

• Craig Newman
• Nathan Bunker
• Rob Snelick

Meeting Minutes

• Discussed WGM agenda for September Baltimore
• Profiling Multiple Occurrences - question from Craig Newman
  • Originally created to allow IG authors to be able to document requirements, but there was no discussion about when it was best to use this ability.
  • Craig is going to continue moving forward with reversing out the special meaning for the first repetition of certain fields.
• Data Type Flavors
  • Like to put stake in the ground and use the naming scheme CE_01, CE_02, etc. But then we reference those in a condition predicate, CE.1 etc. Always keep the constraint references as they were before, and only use the identifiers when the context is needed. Otherwise we just use the base reference.
• Past WGM
  • Delete Indicator needs to be still discussed and written up