Canadian FHIR Baseline Profiles - Due Diligence Review - IPS – May 21st, 2-3pm EST

Attendees

Michael Savage
Sheridan Cook
Elliot Silver
Ken Sinn
Randy Nonay
Ryan Doherty
Shamil Nizamov
Rita Pyle

Review of Slicing for Medication.code

• Providing multiple slices / options for codesets (e.g CCDD, ATC, etc.) is good, but derived profiles may have issues with keeping all of these options defined as optional; don’t want to imply that these options are viable choices if the implementers actually don’t support those options for code sets
• One option is to actually strike those options out in the derived profile / IG
• Other option is to leave them defined as optional and tell the deriving IG guidance to say to NOT use those options in the implementation

Observation

• .reference
o CAB references only Patient, IPS references more (Patient, Group, Device, Location)
o Suggestion to leave CAB reference as-is and revisit if future reviews show that we should be providing option to reference more than just Patient; can note we’re finding the reference to Patient to be the more common need, but are open to opening it up as future reviews happen
o Decision going forward is to relax back to referencing Group | Device | Location as well, will elicit feedback

• .dataAbsentReason
o CAB flags as MS, while IPS does not
o Decision to keep the CAB’s current constraints, and flag our constraints to IPS team to see if they see value in inheriting those constraints (invariants and MS flags as well)

• .component
o Same situation re: value and dataAbsentReason

• .hasMember
o Feedback is that it would be expected to have this element as MS in the CAB
o Suggestion to expand CAB invariants to include hasMember
o Plan is to leave as-is and revisit during next meeting, hopefully having looked at a few other iGuides

To look at next time:

• Revisit use of invariant + MS flags for Observation value/DAR/component/hasMember
• Observation (Smoking Status)

Hi all!

We will be continuing with our Due Diligence Review of IPS for our Friday May 21st session. The planned agenda for profile comparison is as follows:

1. Review of slicing approach for Medication.code (specifically rules of derived profiles needing to inherit open slices)
2. Observation
3. Observation (Smoking Status)

(We're planning to save DiagnosticReport for a future session, where we can protect the entire agenda time, as there are a lot of profiling decisions to work through)

Thank you!