AP Domain Minutes 23Sept2014
|Raj Dash, MD, FCAP- Co Chair||Duke University/College of American Pathologists|
|Francois Macary||ASIP Sante’|
|Mary Kennedy||College of American Pathologists|
- A RECORDING OF THIS CALL IS AVAILABLE IF REQUESTED. PLEASE CONTACT SKREJCI@CAP.ORG
Discussion of ONC Structured Data Capture and its Relation to ONC (See Slides Media: 2014-08-27_ICEH-StructureDataCapture_lb_(2).pdf)
Highlights and Discussions include:
- Past initiatives issues have been the data elements, since everyone is designing their own forms. While its great that the framework is standard, the content needs to be standard as well;
- There needs to be a committed group who will curate the standard data elements. This would be a time consuming project and would require support. This activity would need to be sponsored from by government funding. The need for AP experts to contribute to the content.
ACTION: MK to find out (from Sandy Jones @ CDC) who is working with CDC to participate at the Connecathon of the initial SDC profile.
Action: Rich Moldwin of CAP to provide a presentation on the SDCC Initiative to the IHE AP soon. _____________________________________________________________________________________________
The proposal was to document every operation that happens to a specimen in the Procedure Steps Section of the APSR report. See slides for the evolution of the APSR Report Media: APSR_Evolution_.pdf
The group reviewed the questions that were presented by Gunter and the responses that were provided by François and Raj
Q1: Should one regard the technical procedure for special stain and IHC, too, in the construct for describing the production of a slide from a block? I tried this by using a qualifier for HE, IHC etc. in the description of making a slide. If this is formally not correct, we should look for another solution, but to have this information is very important.
Francois Response: Any operation performed on the specimen, considered as important for tracking in the report should be represented by a procedure step, and be structured as such in the Procedure Steps Section. In the case of staining, my understanding is that it operates on one slide, there is no new slide produced, therefore in this case, the procedure has no <participant @typeCode="PRD">. In all cases, the precise operation performed is represented in a procedure step by the procedure/code. We then need a vocabulary which includes surgery, biopsies, slicing, staining, ... We do not need any qualifier. Just the appropriate code taken from an appropriate vocabulary: SNOMED CT or PathLex.
Raj’s Response: An IHC slide is a new slide from an existing parent block. There may be unstained slides present that can be used, but these are produced off the parent block as well. They are often cut at the time of the original H&E so that the block does not need to be “faced off” again to generate the unstained slides on which to perform IHC (because tissue is lost each time).
Q2: Is it important to have the LOINC code you choose for the Procedure steps section? The alternative I would prefer (68992-7, Specimen-related information panel) seems to be better suited for the "new role" this section plays.
Francois Response: LOINC is not an appropriate vocabulary for procedure codes. Better using SNOMED CT or PathLex. For SNOMED CT the constraint is that you have to have a license for using it.
Raj’s Response: Agree with François. We should have the procedure encoded in SNOMED CT. Mary, do we have update from IHE HQ on the letter to the IHTSDO?
Q3: Is it important to have the LOINC code you choose for the Procedure steps section? The alternative I would prefer (68992-7, Specimen-related information panel) seems to be better suited for the "new role" this section plays.
Raj’s Response: As I recall the “procedure steps” is referring to the steps involved in processing the sample that eventually result in blocks, slides, etc. I do not think a terminology exists for describing this process.
Q4: Should one use the "methodCode" item in the observation for making clear at what level the observation was done, as it is possible in the genuine HL7 observation template?
Francois Response: methodCode can be used with observation as well as with procedure, to describe the method that was used. This requires a coded vocabulary to describe this method. I don't know if such vocabularies exist.
Raj’s Response: Need some examples of methodCode to better understand its applicability.
Q5: Should one orientate generally to PathLex codes and value sets, or should other terminologies not be excluded (SNOMED CT and LOINC if the appropriate codes are available)? Are there HL7 rules for postcoordination processes in SNOMED CT?
Francois Response: SNOMED CT is usable under a license, to code procedures, anatomic sites, observable entities, and lots of other things. PathLex is a temporary terminology, which does not need a license and can be used in countries not affiliated to IHTSDO (e.g.; Germany and France).
Raj’s Response: I do not know if there are rules per se, but guidance has been proposed by both HL7 and SNOMED CT. I do not know if there is formal documentation in this regard. We could reach out to the IHTSDO in this regard.
Additional Highlights and Discussions Points:
- Gunter has drafted an example based on examples that came from Raj;
- It was presented at a session and was satisfied with the result;
- The result is that everything that happens to the specimen will be documented as structured data so it can be processed by the receiving system;
- Other sections of the APSR report become simpler; instead of describing the specimen, they will just need to reference it;
- There have been vocabulary issues when it comes to staining or producing a new slide;
- Mary spoke with Chris Carr at the HL7- he had the draft MOU that this group sent to him. He was going to be meeting with the IHTSDO and was going to present the draft document to them. She hasn’t’ heard the outcome of the presentation and will follow up; and
- Hopeful that the IHTSDO will allow limited use of SNOMED CT during the development of this report. We don’t have the resources to maintain PathFlex, which is why SNOMED CT would be a better vocabulary to move forward with.
Action: MK to follow up with Chris Carr