This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

20170713 OO FHIR conCall

From HL7Wiki
Revision as of 21:10, 13 July 2017 by Ehaas (talk | contribs) ()
Jump to navigation Jump to search
HL7 OO on FHIR (for Orders and Observations)

Call in details:
Phone: +1 770-657-9270, Passcode: 398652

Join the meeting at:

Date: 2017/07/13
2015 - 02:00 PM (Eastern Time, GMT -04 DST)
Quorum = chair + 4 no

scope="col" Co chairs Chair Notetaker
Riki Merrick X
Rob Hausam X
Lorraine Constable
Patrick Lloyd
KD Nolan
Hans Buitendijk

X Eric Haas
X Riki Merrick
Hans Buitendijk
Jose Costa-Teixeira
Dan Rutz
Jonathan Harber
David Burgess
Lloyd McKenzie
X Rob Hausam
X Andrea Pitkus
Francois Marcary
Robert Dieterle
Brian Reinhold
John Rhoads
KD Nolan
Ben Atkinson
Elliot Silver
Serafina Versaggi

  • Roll Call
  • Agenda

Progress Report

Discussed the Boundaries between Media and Observation + valueAttachment and the attachment datatype.

  • Next Steps:
    • Schedule time with FMG to discuss
      • Reminder Media is still owned by FHIR-I and decision whether to go to OO or Imaging is still pending the outcome of this discussion

Transfusion of Blood Product using FHIR

This is the normal PC on FHIR call time which is being scheduled to host a joint call between OO, PC and Pharm to discuss these trackers prior to the Sept WGM Wed Q3:

- GF#12966 and - GF#12993

The Goal of the call is to lay out a roadmap for how to represent transfusion of blood products in FHIR. Then we can follow up at the San Diego WGM.

HL7 audio will be used: +1 770-657-9270, Participant Code: 943377

Join WebEx meeting | 596 965 698

Join from a video conferencing system or application


#13318 Observation example for findings/signs/symptoms

also see the related Zulip chat

Discussed no vote - ready for vote will bring up on general OO call for discussion and vote

  • Rob and I wrote up some guidance text surrounding the Question-Answer pair pattern and the variations on that theme.
  • Add examples
  • Add text to code and value comments to stating that need to look at both

#9980 Observation example for findings/signs/symptoms

Discussed no vote - ready for vote will bring up on general OO call for discussion and vote

  • add extension for event pattern `.eventHistory` which references Provenance (+ do this for all OO event resources Obs, DR, SupplyDelivery, Task. See how similar element implemented in DiagnosticRequest `.relevantHistory`

meeting minutes

Need to figure out if this is the tracker we have been working on with the modifier discussion
When code normally used in .value is used in .code must be careful to review the .value to know what it means  Eric and Rob H were exchanging a document  = may be on a different tracker?
Code and value MUST always consider together – meaning MUST always be derived from BOTH
In lab results for example we use qualifiers all the time for detected and not detected
Grahame likes to have the asserted codes in .code rather than have a meaningless code like assertion in .code and then report the finding in .value and a modifier
OBX-3 = question – which can be clear or not
OBX-5 = finding (typically we use organism as a shortcut to avoid having explosion of the finding terms, for example when detecting the DNA of an organism – which is really a substance) – or numeric values
OBX-8 = interpretation of the finding, especially when we have numeric results
May be need to profile observation for the different approaches
Use examples to see if you can properly support the business rules and regulations in FHIR resources
Creating specific examples for each of the scenarios we discussed
For clinicians on FHIR work on lab reports – not sure, if that will be on the docket - Andrea won’t be in San Diego
Labs are not using FHIR as far as we can see, though Cerner reported they are using FHIR for lab – think that may be for provider access to data from their EHR-s.
Reach out to Michelle or Jenny for more info – Eric to do
Cerner: Pathnet millennium, CoPath, millennium AP
EPIC: Beaker (more recent)
Meditech, McKEsson, Orchard, Sunquest are ONLY LIS vendors
Person or device to finalize the result – who did it = not clear as current, butt can use the provenance resource to find that out. Not many resources use this currently
Event pattern has relevantHistory attribute to point to the past – but how do you point to the current provenance, so need to work out how that works a little more
#12896: Timing can reflect current and past events – was approved, so could we use that? = there seems to be an extension that might overlap

Back to OO_on_FHIR