PCD TC Message Profile Registration

From IHE Wiki
Jump to navigation Jump to search


Message Profile Registration Task Group

Each of the PCD integration profiles includes one or more message exchanges, for example, PCD-01, PCD-02, etc. Once these messages have been published for Trial Implementation, then they may be registered (e.g., with HL7 or some other public registry) and the resulting identifier used to perform conformance testing (static & dynamic), application configuration, etc.


Task Group Participants

Organization Participant
Iatric Systems Jeff McGeath (jeff.mcgeath@iatric.com) - Co-Chair, IHE PCD Technical Committee
GE Healthcare Robert Flanders (Robert.Flanders@med.ge.com)
Sherman Engineering, LLC Paul Sherman (paulrshermancce@gmail.com) - IHE PCD Technical Project Manager
NIST John Garguilo (john.garguilo@nist.gov)
Philips John Rhoads (john.rhoads@philips.com)
ProtoLink Al Engelbert (AlE@protolink.com)


List server: To send a message to the task group e-mail ihe-pcd-msg-reg@googlegroups.com

Task Group Meetings

<add page links to meeting agenda/minutes>


HL7 v2.x Message Registration Process

For HL7 v2.x messages, a registration web site is provided to formally submit message profiles and to obtain an ISO Object Identifier (OID) for that message. From the HL7 home pageselect Submit 2.x Global Message Profile (note: this may only be restricted to HL7 members).


Once an OID is obtained, it may be used in the MSH-21 field of the associated message, thus enabling formal dynamic conformance testing.

Message Profile Registration Activities

There are a number of message transactions supported within PCD profiles, all in various states of maturity. Once they have been finalized for Trial Implementation, they may be officially registered. The transactions below indicate those message profiles that are or have been registered.


PCD-01: Communicate Device Data

Registration OID: <TBD>

Registration Status: <tbd>

<link to message profiles on FTP site>


Baseline MWB Profile

Steps to Create Baseline Profile

The following steps were taken to establish a baseline MWB profile that conforms to the Technical Framework for Trial Implementation (Rev 1.1, Published 2006-08-15):

  • copied/renamed files as:
    • 20080815-PCD-01-V1.0.mwb
    • 20080815-HL7TableFile25-PCD-01-V1.0.mwt
  • Launched MWB
  • Loaded 20080815-HL7TableFile25-PCD-01-V1.0.mwt table file through Maint | Options, General tab, HL7 Table file (at bottom)
  • Loaded 20080815-PCD-01-V1.0.mwb through File | Open
  • Modified the profile as follows:
    • MSH.4.1 (Sending Facility | Namespace ID) -- changed "Table" to 0362
    • MSH.5.1 (Receiving Application | Namespace ID) -- changed "Table" to 0361
    • MSH.6.1 (Receiving Facility | Namespace ID) -- changed "Table" to 0362
    • PID.3.4.1 (Assigning Authority | Namespace ID) -- changed "Table" to 0363
  • Modified HL7 tables, as follows:
    • Table 0301 -- Added BBRAUN (for OBX.18.4)
    • Table 0361 -- Added PAT_DEVICE_BBRAUN (for MSH.3.1) and MWB (for MSH.5.1)
    • Table 0362 -- Added BBRAUN (for MSH.4.1) and MWB (for MSH.6.1)
    • Table 0363 -- Added HL7 (for MSH.21.2) and AA1 (for PID.3.4.1)
    • Table 0396 -- Added UCUM (for OBX.6.3) and ISO659 (for MSH.19.3)


Questions

  • Appendix B.1, Table 9
    • MSH-3 (Sending Application) says Table #0361, but in HD data type definition, Namespace ID is Table #0300
    • similarly for MSH-4, MSH-5, and MSH-6


Open Issues

  • PID-3.4.1 (Assigning Authority, Namespace ID) is marked as required in the profile, but was empty in some messages received at Connectathon and Showcase
  • PID-5.7 (Patient Name, Name Type Code) is marked as required, but was empty in all messages received at Connectathon and Showcase
  • From Peter Rontey: A couple of issues important to resolve and I think we need to bring Rob into the consideration:
    • OBX.20 – field doesn’t exist in 2.5, is only a place holder in 2.5.1. This can be supported in Message Work Bench, but what about NIST's Message Maker?
    • OBX.5 – Indeterminant data type (i.e., multiple possible OBX value data types), needs to be resolved for truly implementable profile. I can think of a way to model this in MWB with optional seg groups representing different combinations of OBR and OBX. Will that work for Message Maker?
    • Those are the big things; couple of small ones like erroneous CNS data type, Control char in name for LOINC table value, nonexistent table values etc. I’ll just correct.