HL7 Version 3.0 Task Force (V3TF) Meeting Minutes January 28, 1994 Doubletree Inn New Orleans, LA Agenda I. Appointment of Acting Secretary II. Statement of Principles/Introduction III. Health Data Sciences Letter IV. Object-Oriented Messaging Paradigm Attendees Dave Carlson Gary Dickinson John May John Quinn Wes Rishel Joe Schlesinger Mark Shafarman Abdul-Malik Shakir Wayne Tracy Mead Walker Absentees Dean Bidgood Shel Bridgman Pat Cahill Jim Demetriades Ed Hammond John Hoben Clem McDonald Mark Tucker Other Visitors Greg Forbes Brad Kowalski Hans Buitendijk Jonathan Tien Robert Burkhead Scott McPherson Wayne Moyers Shivajce Sarndarrhi James Chow Frank Wilcox Jesse Barber III Minutes Meeting was called to order. I. Appointment of Acting Secretary Dave Carlson was appointed acting secretary in John Hoben's absence. II. Statement of Principles/Introduction Wes handed out 3 documents: 1.HL7 Version 3 Statement of Principles (which will be balloted), 2.Introduction to HL7 Version 3, 3.Copy of the letter from Gary Dickinson of Health Data Sciences. There was a discussion on Conformance Claims. Wayne: Should it be in the Standard or in the Implementation Guide? John: Needs strong guidelines if we include it. Dave: It is important, but beyond the scope of this Task Group. A separate Task Group should be formed. Abdul-Malik Shakir: It would be enough to establish Claim by defining which Trigger Events are supported at each Conformance Level. Dave: Start a TG as a subgroup of the Technical Committee. Motion passed (1 No-Wayne Tracy, 7 Yes): Change wording to: "Recognize the importance of having a method for vendor implementations to claim level of conformance to the HL7 Standard and that a subcommittee of the Technical Committee be formed to address this issue." Discussion of Scenarios and Transactions. Mead: Notion of scenarios is critical, but the definition needs to be developed further. However, we should move forward as worded. John: Clarify intention so do not implicitly get into specific functionality and system architecture. Wayne: Scenarios should be illustrative, but not comprehensive. Wes: Remove "Initially" from the HL7 Intro, page 11, line 4. III. Health Data Sciences Letter Gary: Outlined content of letter. Two issues remain: Interoperability and the Computer Based Patient Record agenda. All the others Gary feels have been addressed. Interoperability. List issues. Mead: Does the SIG established out of the Microsoft presentation during the Control Group meeting address this. Wes: Not necessarily Wayne: Agree as a statement of overall goals of HL7, not necessarily for 3.0. Establish a working list of interoperability issues. Mark: It is an implicit goal we are working toward anyway. Version 3.0 is a movement toward this. Dave: Could this be relegated to the TG recommended in question I above. Motion passed (unanimously): Recommend the creation of an Interoperability Task Group of the Technical Committee to enumerate a list of interoperability issues one aspect of this is the development of HL7 conformance procedures. This supersedes the previous motion. Wes asked Gary to draft a subset of his letter that references interoperability and submit it as an initial guide to the Task Group identified in the immediate motion. Computer Based Patient Record agenda. List the objectives. Wayne: We are a data interchange standard, we are not a CPR organization. It is outside the scope of HL7. Mead: This is more appropriate for the CPRI of which HL7 is a member. John: Does not see this as something the CPRI is handling, but we don't need a specific group/action to address this. HL7 would develop appropriate messages as the business needs develop. Wes: There appears to be no group handling this. Wes: When a CPR is finally developed, we would review its impact on HL7 at that time. It is the consensus of the group we should not tackle this issue. It was the consensus that, subject to the edits implied by the discussion in this meeting, the work of the task force is ready to go forward. Motion passed (unanimously): Wes be empowered to edit the document for final submission to the Executive Committee with recommendation that it be balloted by the full membership. Meeting Adjourned at 11:30 am. IV. Object-Oriented Messaging Paradigm (This was an unofficial discussion.) Wes: An easy way to recast HL7 into an object-oriented paradigm would be to define objects which correspond to the system Roles, as described in the current V3TF Statement of Principles. So an object might be an ADT.Recording.System. These objects would contain data (subobjects) that corresponds to the patient, visit, clinical observation, etc. The exact representation of this data is hidden from the outside rule, according to O-O orthodoxy. The role objects would publish methods that correspond to HL7 trigger events. In particular the methods would correspond to the accepting object's view of the trigger event. So an ADT.Recording.System object would correspond have an Update.Admission.Data method only if it **accepts** such a message (in HL7 terms.) The parameters associated with these methods would correspond to the subobjects defined above (patient, visit, etc.) Mead: The important problem is the focus on the common view of the sub-objects. There is conceptual support for what Wes outlined. Mark: This is a useful construct. This approach is not included in the HL7 Version 3 Statement of Principles.