2023-11-27 Vocabulary WG Meeting Notes

Owned by r.dash
Last updated: Nov 27, 2023
5 min read

Participants for today’s call:

 

Membership:

Name

Organization

Role

Name

Organization

Role

Scott Campbell

UNMC

Co-Chair

Steering Committee member

Raj Dash

College of American Pathologists (CAP)

Co-Chair

Steering Committee member

Dan Rutz

Epic

Steering Committee member

Muktha Natrajan

CDC

 

Sandy Jones (Secondary)

CDC Cancer Surveillance

 

Anne Peruski (Secondary)

CDC

 

Andrea Pitkus

University of Wisconsin-Madison

Steering Committee member

Xavier Gansel

bioMérieux

Steering Committee member

Stan Huff

Graphite Health

Steering Committee member

John Snyder

NLM

Steering Committee member

Rob Hausam

Hausam Consulting

 

Marjorie Rollins

Regenstrief

Steering Committee member

Amy McCormick (secondary)

Epic

 

Nanguneri Nirmala

Tufts Medical Center

Steering Committee member

Mehdi Nassiri

Indiana University/Indiana University Health/Association for Molecular Pathology

Steering Committee member

Eza Hafeza

Regenstrief

Steering Committee member

Jim Case

Snomed International

Steering Committee member

Mary Kennedy

CAP

 

++++

 

Discussion topics

Item

Notes

 

Item

Notes

 

Item

Notes

 

Item

Notes

 

Review: Identify tangible activities for this working group

Identify use cases / challenges in lab interoperability.

  1. D-dimer (DIMER - Overview: D-Dimer, Plasma (mayocliniclabs.com))

 

Use case review

  1. D-dimer reviewed by Dr. Dash and Dr. Campbell with actual examples of result reporting in patient charts (de-identified). See attached word documents.

  2. Duke does not use LOINC codes except as required regulatory reporting (e.g. MU stage 2, physician quality report, etc); and method-less whenever possible to preclude assignment of an inaccurate code (not required or coded for D-Dimer at Duke).

  3. UNMC does have a LOINC code assigned but DDU vs FEU unit of measure has the same LOINC code assigned.

  4. LOINC does not differentiate tests with different units of measure if methodology is the same.

  5. LOINC code look up for D-Dimer above

  6. Normalization of units of measure (UOM) is a necessary but insufficient condition to support equivalency of results

  7. What might be helpful is to establish a framework of guidelines that would support equivalency of results (namely same instrument, methodology, following manufacturer’s procedure for QC and calibration).

  8. Demonstrating equivalency for other situations (e.g. different instruments with different methodology) requires “harmonization” using a process that needs to be established (e.g. looking at patient result data cohorts across various demographics and disease conditions).

  9. As a starting point, LIVD codes could be mapped to a “harmonization data element” where equivalency has been established.

  10. Is there a use case for LOINC groupers? LOINC does not add groupers for less than fully defined concepts (where all 5 axes defined).

  11. It may be helpful to have some axes more generically defined for using LOINC for certain situations, such as when placing orders (where details for lab testing, such as method, might not be readily available).

  12. Once SCT incorporates LOINC in the LOINC extension - can group based on defining characteristic or based on independent selection. Create refset (subset) that describes the group (if cannot be automated). All concepts need to be as fully defined as possible.

  13. ECL = expression constraint language...a way of really digging into SNOMED...graphy

  14. Challenge in harmonizing ECL across organizations.

  15. Labs don’t have the resources to maintain ECLs in a federated manner. Some central resource would need to provide and maintain.

  1.  

Next steps

  1. AST (AST - Overview: Aspartate Aminotransferase (AST) (GOT), Serum (mayocliniclabs.com))

  2. Urine culture and susceptibility testing (DUHS Electronic Test Catalog (duke.edu))

 

Cloud recording:

Video Conferencing, Web Conferencing, Webinars, Screen Sharing

From Chat:

Andrea Pitkus to Everyone 1:31 PM
thanks Pam, for forwarding info
https://testcatalog.duke.edu/#/test-detail/LAB131/1230100036

You to Everyone 1:35 PM
https://www.beckmancoulter.com/wsrportal/techdocs?docname=/cis/A18460/%%/EN_AST.pdf

Andrea Pitkus to Everyone 1:41 PM
so d-dimer specimen tested is platelet poor plasma, right?
you should have documentation of instrument change if that occurred as well as communication to providers of the changes
PPP is used on most coag analyzers uncless point of care testing

Andrea Pitkus to Everyone 1:48 PM
Do you all have the IVD devices and model info for these assays?
Does your coag technical specialist/supervisor?
know? agree w Pam

Scott Campbell to Everyone 1:54 PM
Actually, Raj and I can actually work on vehicles

Andrea Pitkus to Everyone 1:57 PM
Have you done literature reviews between the different methods, different units here like reported by CAP in literature for d-dimer, etc.?
one way is to convert all values from different units to one set of units. (Not that I'm advocating for it)

You to Everyone 2:00 PM
^ that will not guarantee interoperability
but is necessary
necessary but not sufficient

Andrea Pitkus to Everyone 2:01 PM
right, because there may be other factors
Do you have the IVD info from the IVD vendor such as the package inserts for these per the original homework?

Pam Banning - 3M Health Info Systems to Everyone 2:02 PM
I didn't go that far Andrea in gathering from the reference labs.

Andrea Pitkus to Everyone 2:02 PM
The other need I hear is how do we indicate apples to apples and apples to organges

Scott Campbell to Everyone 2:02 PM
That's a good call, Andrea. We should get our package inserts for the tests in our homework

Andrea Pitkus to Everyone 2:03 PM
Let's do that, Scott. Pam, I started looking up the one's from Mayo as they overlap with the Roche LIVD maps you provided

Jim Case to Everyone 2:03 PM
I am not sure you can make a pure equivalence between test results from year to year. Things change all the time

Scott Campbell to Everyone 2:04 PM
@Jim Case True, but could/should there be something that indicates a change in process over time?

Amy McCormick to Everyone 2:05 PM
Unit conversions concern me. Especially when there are reference ranges in the component-related comments (because that won't be converted).

Andrea Pitkus to Everyone 2:06 PM
or a clinical decision support tool per the use case
challenge, tied, stimulant results versus not
timed

Scott Campbell to Everyone 2:09 PM
brb

Andrea Pitkus to Everyone 2:14 PM
the research community is trying to use generic methodless codes as a grouper as they think it provides a parent-child relationship like SCT.

You to Everyone 2:15 PM
Does anyone know if we can attach word documents onto the web site?

Andrea Pitkus to Everyone 2:15 PM
It's desired
faint

Scott Campbell to Everyone 2:16 PM
ECL = expression constraint language...a way of really digging into SNOMED...graphy

Andrea Pitkus to Everyone 2:19 PM
Thanks for sharing ECL Scott. Seems like similar issue with the lab test knowledge you mentioned earlier. If curated groups, it would help the end user.
One other aspect I mentioned this am, is how we name things (will be discussing in future). D-dimer by FEU vs D-dimer DDU for example so we know the distinctions that would be grouped/flagged, etc.
Jim, good point. a number of order LOINCs are methodless for that order user case
I'd recommend starting with similar tests (avoid reflex and complexities but work that way ) once we get the foundation/building blocks established
focus on units used in US
exclude intl used units
start with IFCC harmonized results/methods

Andrea Pitkus to Everyone 2:25 PM
I wouldn't expect labs to use ECL
many don't have resources or expertise
especially critical access hospitals

Jim Case to Everyone 2:30 PM
gotta run...

Andrea Pitkus to Everyone 2:31 PM
great discussion