2026-07-02 LabMCoP Meeting Notes
Date
Jul 2, 2026
Attendees
Present | Name | Organization |
|---|---|---|
- | Nancy Cornish | CDC |
Manjula Dharmawardhana | CDC | |
X | Riki Merrick | APHL |
X | Christina Gallegos | APHL |
- | Amy Liu | Inductive Health / APHL |
- | Raj Dash | Duke / CAP |
- | John Snyder | Pragmatic Terminologies, LLC |
X | Andrea Pitkus | UW |
X | Kathy Walsh | Labcorp |
Rob Hausam | Hausam Consulting | |
Pam Banning | 3M | |
Elissa Passiment | ||
Kristina Betz | CDC - NHSN | |
Jake Bunn | CDC | |
Marissa McMeen | CDC |
Upcoming OOO |
|
|
Reflex testing in FHIR follow up | Andrea | USCDI V6 has been listed in SVAP - need to keep an eye on how folks are implementing - look at USCore_ Andrea is working on a reflex workflow layout to check how the orders / grouping and specimen reject might work: reflex orders are not often included in current messages, but PH would really like those - how would that look in FHIR, could we fix it there? for blood culture is volume included as OBR (is it reflex in the LIS, or it may be always part of the panel, especially if this is always included, even if no growth on the cultures if growth (which may or maynot be reported out) reflex to gram stain - this should be grouped; reflex to organism ID for each culture grown - need to figure out what methods: MALD-TOF, PCR, others?; need to have linkages to the correct cultures / organisms for each organism would reflex to the apppropriate ABx susceptability panel reporting of susceptibilities has several choices:
Note there was a Biomerieux webinar on blood cultures earlier this week- will see, if recorded and then update the list of Abx to use We will probably discuss these on the Lab calls (UV reporting and generic Lab) - or when we pick up the grouping discussions again; might also look at parts of this on specimen call when we are looking at the reject reason scenarios Copied from email sent after call: Some background. It does not include everything (e.g. Specimen Resources/details), but focuses mainly on orders and results. These are the terms which would be mapped to LOINC and built in LIS dictionary builds and used in LIS ordering workflows from ordering/reflexing and reporting. When reports are released from the performing lab, they flow into the EHR either without interfaces via the internal database items or via interfaces to the EHR or PH. Either way, the receiving system also builds the order and result terms so there is a place to "file" (using Epic-speak) or be remapped to the EHR terms in the database (and then the HL7 message such as v2 is destroyed). The top level order is what is ordered by the clinician. Any reflex orders are not ordered by the clinician by added by the performing laboratory if reflex criteria are met. The test order terms are used. In Epic there is a flag that indicates these reflex orders are not physician orderables. Do folks know how these are designated by Oracle, Meditech and other EHR vendors? Ordering workflow is being discussed on HL7 COW calls from a global perspective (and any order, not just lab). There are no FHIR Lab Order Implementation guides under development yet either. However, ONC this week posted the following: USCDI v6 is now included in the SVAP and Health IT vendors can start providing functionality including FHIR in August as part of this process. As such, vendors will likely be "interpreting" how lab order (now included in USCDI v6) exchange functionality would occur. As seen with their previous FHIR specifications. Different vendors may choose different FHIR Resources (FHIR Service Request, FHIR Procedure, and FHIR Observation) to support for lab orders. This burdens all if all 3 need to be supported for queries, exchanges and receipt, especially with exchanges across different vendors or for those transforming V2 lab, pathology, genomics messages into FHIR. COW and OO calls discussing lab orders have not started working on complex examples like cultures, reflex orders, 1 click physician ordering as they are still working on simple order resultable examples like Potassium (K) or simple panels like CBC or BMP. There are some US Core examples that are pretty basic too. Thanks Riki for confirming US Core v 9 profiles are what represent USCDI v6. We discussed taking a look at those. In the Blood Culture simplied example, I have the top level order from the ordering clinician in the top left and once received by the laboratory, the first step is to ascertain if there is growth or not. The Blood culture bottles are incubated in a blood culture instrument. The attached example shows if there is no growth a result with text used by laboratories in preliminary reports such as No Growth at 24 hours (Please share your real world examples as we can update the content.). This would be an OBX in v2 messaging and FHIR Observation as shown. It is reported in FHIR in a Diagnostic Report Resource. I've started added LOINCs too to show what codes may be used order and result data dictionary maps exchanged internally, via V2 OBXes, and in FHIR Codeable concept. (The arrows represent laboratory workflow and when used in FHIR, many would be reversed.) I've included the Sub order (which may be spawned from the Blood Culture Order if expected for all blood cultures as shown here. We've discussed multiple places in the lab workflow so this could change. Let's say it's noted upon receipt of the specimen/blood culture bottles as discussed. A test order of Blood Culture Bottle Total Volume (or similar name) built in the LIS order dictionary would be mapped to a new LOINC for this order panel. Once decided, this could be requested from LOINC. The idea is th standardize the panel so each lab doesn't need to invent their own structure for it. With the O&O discussions, Hans recently presented and indicated in FHIR not only would the Order panel (shown in red) be needed but a FHIR Observation of the same order panel/thing is needed to serve as the grouper of the results in the the Diagnostic Report. Thus we see that in blue in the diagram. Each result for Bottle 1, 2, etc for as many collected and a total volume of all the bottles, which can be a calculated value like timed urines are reflected as their on OBXes or FHIR Observations below the order panel/grouper Observation. These would have their own result LOINCs too. There was discussion of using an exisitng LOINC or creating new separate LOINCs for each Bottle so kept distinct. Need to finish that discussion and update accordingly. Nancy indicated she was going to work on standardized terms for a number of these items and some related responses. In this example, no growth was seen so no additional relfex orders occurred and items shown are reported in the Diagnostic Report. With the new O&O decisions on "grouping items" in their respective panels t reflect lab test structure/modeling and thus mapping to codesystems, the option of using Observation as that grouper is reflected in Blue. Would a second Grouper be needed before the Growth items at the top to keep those items in their own group or is it sufficient to use the relationship with the order as shown and only have groupers in the sub orders? This may be discussed on future HL7 calls. We also know that some LIS vendors lack functional especially in micro and blood bank workflows to exchange reflex orders and likely sub orders outside of the LIS. These sub orders may be in V2 messages or not. If not, then the question becomes how/who will create the Bottle Total Volume order grouper with it's appropriate LOINC in codeable concept, performing lab's test name and associate it with the bottle volume results as shown? As folks mentioned, it's challenging to create something out of nothing. I mentioned SHIELD may want to make a recommendation that when LISs currently "suppress" the sub/reflex order name (in the darker red for the Total Volume of the Bottles), from being released from the LIS to the EHR, PH and other downstream consumers of reports, these should be available in the report to the EHR, PH and other Health IT like other sub orders (e.g. CBC with Manual Diff) as FHIR will need those for accurately grouping in FHIR Diagnostic Report. These complex needs and real world examples haven't yet had discussions for all the FHIR needs to represent them yet. There has been extensive disucssion on grouping in FHIR Diagnostic Report from a technical perspective.. Of the many options discussed, it was whittled down to two globally. There is not yet discussion of a US FHIR Lab IG which meets CLIA requirements for exchanges too. Until a US Lab FHIR IG exists, vendors may again "interpret" results reporting with LOINC in different ways as seen. There is the Blood Culture FHIR IG for Clinical Quality Measure Reporting so need to continue discussion on flow of data from laboratories to EHRs that would create the FHIR messages with the content needed for those. Blood Culture Bottle Total Volume is one of the content items. What other questions do folks have for future Lab Cop and/or HL7 discussions? Do want to ensure exchanges with the new paradigms are supporting laboratory (clinical and Public health) reporting needs. Appreciate real world examples. I should note, we know FHIR Specimen Resource discussions are needed too to work through the relationships of the parent specimen and derived specimens like isolates are needed in FHIR SPecimen Resource. It is also needed for examples like this with FHIR Observations and results made on the parent specimen as well as islolates (e.g. identification and susceptibilities) all reported together for each organism in FHIR Diagnostic Report. That is another layer of complex global lab requirements not yet worked out. Another aspect which recently arose with USCDI v6 including more lab data such as "Specimen Condition Acceptability" are workflow needs in FHIR to support this new item. Laboratories have been reporting for decades as part of CLIA requirements, but often free text comments or reasons for why an order is canceled (e.g. broken tube or bottle or slide) upon receipt in the laboratory. With USCDI you'll note it is a discrete item, and also has codesystem requirements fro SNOMED CT. CLIA also delineates what is considered an acceptable specimen or not and it is rejected leading to the order cancellation, as well as noting the Specimen Condition(s) (e.g. hemolysis, icterus, lipemia) that impact one or more results reported. This is a whole other discussion. |
Call adjourned | 12:00 PM EDT - BELOW NOT DISCUSSED | |
Working with DHQP for NHSN reporting | Nancy | Follow up list:
|
Question on Karius Testing | Riki/Christina | Next Steps - working through these:
|
Follow up from AndesVirus codes question | Riki - no update |
|
SNOMED International proposal for Serologic terms | for Jul 9, 2026 | SNOMED International is purposing to revise content with “Serology” or “Serologic” in the FSN and remove this wording from FSN across clinical findings, procedures, and observable entities. I would like LabMCop team to review and provide any feedback on the topic during the meeting. Prior discussions: Next Steps:
|
HPV screening/confirmatory testing workflow | John | Prepare for a call on Jul 2, 2026 or Jul 23, 2026 : We’d like to discuss with the group the current modeling and naming policy for procedures connected to detecting HPV and/or pre-cancerous/cancerous cells in the cervical region. From Andrea Pitkus' email: US protocols are usually to reflex to HPV testing with PAP Smears. Most use liquid based methods such as Thin Prep (brand). Technically, cytopathology is the term used for this area of laboratory testing. I'm sure Nancy will have more thoughts too. Robyn Temple-Smolkin (AMP) email: Our reference lab used to run HPV from ThinPrep or SurePath. Sample would go to the hospital cytopathology lab for preparation of the pap smear then the atypicals were reflexed as sendouts to us for high-risk HPV testing by PCR. The entire ThinPrep/SurePath vial was the specimen we received. We also did CT/NG off this sample when ordered. I believe this is still a very common workflow performed either within one institution or as a sendout, but defer to others' input on newer workflows as I have been away from day-to-day operations since joining AMP. Discussion today: = example of the collection kit the collection is the swab/brush/spatula of cells, this is really the collection container though Here is the labcorp example: which shows the different colelction methods depending on container used at the bottom Hologic ThinPrep package insert page: Try to see, if Jenna from Arup is available on either of those dates Note: Andrea and Riki are both out Jul 23, 2026, might hav eto find a later date, if July 2 is not working; John will call in only on Jul 2, 2026 |
Specimen CMT - Hosting Options | Riki | Any Updates on FU with: Clinical Architecture: They could create conceptMap based on the content model they built, need to follow up about FHIR server usage ONC: Pull up email from Carmela |
Specimen CMT terms | Christina | Review collection method concepts (initiated by USCDI valueset creation: ): retired concepts: 24139008 Endoscopy of urinary bladder (procedure) not in procedure hierachy: 258429002 Rectal scrape specimen Using preferred name, not FSN (these should not be an issue in the final version, since FSN or preferred name would be pulled based on the concept code, correct?): 122462000 Drainage procedure only NHSN terms left for questions: |
CMT Governance process | ||
CMT proposed maintenance process | Start writing this up on this confluence page: Questions from Zulip thread that should be covered by the process we come up with here:
User Guide starter | |
Follow up items |
| |
Lab tests as procedures or orderables | Do we still need this? | Recommendation by SNOMED is that the Observable entity hierarchy be used for both ordering and resulting. As for the technique hierarchy, that in no way is intended to be used for either ordering or resulting laboratory tests. Those concepts are used to model both observables and procedures. SNOMED LOINC extension: The problem is that major EHR-s vendors have set up CPOE using Procedures for orders to initiate a workflow (that creates the triggering event in the system) - that’s where the push-back comes from. There is a CPT to SNOMED CT mapping (as a paid mapping available from AMA), but no LOINC mapping. Can we reach out to CAP Informatics, ADLM Informatics and ASCLS Informatics to get their take on where Lab tests should live We had said we would work through the Colonoscopy example: Reference links: clinical guidelines: Official journal of the American College of Gastroenterology | ACG Quality Indicators: Official journal of the American College of Gastroenterology | ACG
There is this question in the LOINC Community: it is related to this US Core FHIR Change request: https://jira.hl7.org/browse/FHIR-54415 Answer to this one is that in USCDI Procedure (https://isp.healthit.gov/taxonomy/term/781/uscdi-v6) does not list LOINC as applicable Vocab standard in any verion, so remove it. Often folks think of lab tests as procedures, because they can be ordered in CPOE (and outsidde the US, in the UK for example that’s how folks have modeled those, which is why LOINC was adding more of the high level order codes in 2.81 release, so we still should tackle the LOINC community question. John was working on creatign an intesnional valueset definition based on class + type and maybe a few other attributes |
European Semantic work | Link to the German FHIR IG around suceptibility testing:
Semantic Example section: Confluence page: Ask if Rob can keep us updated | |
Specimen CMT pilot implementers |
| |
Specimen CMT - education |
| |
Specimen CMT - tracking implementation impact
|
| |
Future projects for this call after CMT |
|
Recording:
Please reach out to riki.merrick@aphl.org for access to the recording.
Chat:
Action items
Quick decisions not requiring context or tracking
For quick, smaller decisions that do not require extra context or formal tracking, use the “Add a decision…” function here.
- Add a decision…
Decisions requiring context or tracking
For decisions that require more context (e.g., documentation of discussion, options considered) and/or tracking, use the decision template to capture more information.