Meeting Minutes 29 March 2011
Notes from RPS WG Teleconference 29 March 2011
1. NOTES ON PREVIOUS MEETING
The notes from the meetings of 8, 15 and 22 March have been posted on the HL7 RPS wiki. There were no questions or comments on the notes from the last meeting.
2. REQUEST FOR A MORE INTEGRATED AND MORE TUTORIAL APPROACH TO DOCUMENTATION OF RPS
Keith reported that a member had expressed concerns about the difficulties encountered by members who were trying to understand the model in terms of the business processes they understood, and asked that the information essential to understanding model, including information about HL7 modeling rules and stakeholder requirements, be presented in one place in a more tutorial fashion.
There was general agreement that such an approach was needed, and considerable discussion followed, but no consensus was reached as to the form and content to be used. The main suggestions were:
- People want to see how the model plays out in the real world: build an explanation around a set of scenarios or build an explanation around a set of test case scenarios;
- People want a more basic walk-through of the structure and function of the model: use a narrative document and/or power point plus the glossary to create an “RPS For Dummies” or build a narrative against the model as recently re-organized (power point is not preferred for printing).
- People want to understand the model in terms of the eCTD that they already know: build a cross reference from eCTD concepts and actions to RPS concepts and actions (ICH is working on something along these lines)
- Vendors and IT people want to understand the XML, though users are not particularly interested in that level of detail: provide XML examples of each model construct (the Implementation Guide will have such examples) and/or a reference XML instances.
- People want to learn by doing (those who had worked on the R2 DSTU testing most strongly supported this point): build a tool or tool simulator to allow users to create test cases and see the effect as it would be seen, either from a rendered display of the message produced or of the effect of the message as seen by the receiver.
3. REVIEW OF COMMENTS AND QUESTIONS RECEIVED TO DATE ON THE MODEL WALK-THROUGH
Marc pointed out that we should not speak of things new to RPS R3, because HL7 does not recognize the existence of RPS R3. Since release numbers only increment from normative standards, and since R2 is still a DSTU, another iteration will still be R2. That leaves open the question of what to call this iteration of work on the standard.
Jason, Joe and Joel will note proposed changes to the walk-through and compare notes later.
The need for the reference to Medical Records was questioned; Jason reported that RPS1 had taken its life cycle model from that domain model and the Clinical Document Architecture.
There was some discussion about what to call the party responsible for a submission unit sent to a regulator; ‘applicant’ and ‘submitter’ were proposed but the point was left open for later decision.
There was a discussion of whether all key terms should be introduced up front, or explained as needed, or whether the reader should be directed to the glossary. No decision was reached.
It was suggested that the CoU and document life cycle sections should be rewritten; Keith said that this was one of the items listed in agenda item 5 that needed a detailed re-review.
The object states and life cycle of a submission were discussed. Jason asserted that the regulatory states could be mapped to the states defined in the HL7 RIM; Marc declared that the FDA wanted to communicate regulatory states using regulatory vocabulary; Keith supported Marc by pointing out that the state machine for an HL7 act has not only a different vocabulary but also a different configuration than one for regulatory states, and suggested the controlled vocabulary could created easily but that a new class might be needed in the model to communicate this information.
The call ran out of time at this point.
4. GLOSSARY STATUS
Keith reported that the glossary is now posted on the RPS wiki, with a link from the main page, and invited members to contribute missing definitions and make changes to existing definitions.
5. OUTSTANDING REVIEW ITEMS
Items to be reviewed (or re-reviewed) in the next few weeks:
- life cycle for submission unit, submission and application - life cycle for document, CoU, keyword use & definition, related app., product information - inheritance of keywords and titles through documents and CoU - applicability of keywords to all applications pertinent to a CoU - error responses from regulator
THIS ITEM WAS DEFERRED UNTIL NEXT WEEK
6. NEXT WEEK
Continue with discussion of Model walk-through and the outstanding review items.
7. FOLLOW UP ITEM: FIRST ENTRY IN THE JOE CIPOLLINA CHALLENGE
A regulator and a data modeler walk into a bar and take their usual table. Since there seems to be no wait staff on duty (it is the middle of the afternoon) the data modeler gets up and says, “I’ll go to the bar and get the first round. I’m having a Bud. What’ll you have?”
The regulator says, “I’ll have a Heineken, and if the bowl of big pretzels is on the bar, bring half a dozen.”
A few minutes later the data modeler returns and sets down a tray with one Bud and six Heineken.
The regulator says “Why did you bring me six beers?”
And the data modeler, looking somewhat surprised, replies “The bowl of big pretzels was on the bar”.