Stream: terminology
Topic: COVID-19
Grahame Grieve (Mar 11 2020 at 01:17):
here's a value set for all the COVID-19 related LOINC codes:
Grahame Grieve (Mar 11 2020 at 01:17):
Grahame Grieve (Mar 11 2020 at 01:17):
I think that's correct...
Grahame Grieve (Mar 11 2020 at 01:21):
And snomed codes for the Condition resource: ValueSet-covid-19-cond.json
Abbie Watson (Mar 11 2020 at 01:34):
(deleted)
Peter Jordan (Mar 11 2020 at 01:51):
The descriptions in that SCT ValueSet don't reflect the Technical Corrections that SNOMED International made to those concepts after the WHO officially named the condition COVID-19. See https://www.snomed.org/news-and-events/articles/changes-coronavirus-descriptions-(1)
Grahame Grieve (Mar 11 2020 at 01:54):
how do they not? that's the page I copied them from
Grahame Grieve (Mar 11 2020 at 01:54):
I don't see a snomed CT Concept for 'testing for Covid-19"- I would've thought that would be a relevant and important SCT concept? I suppose we just have to fall back to 121973000?
Peter Jordan (Mar 11 2020 at 02:03):
You may have copied the descriptions from the 'original description' column, rather than the 'new description' one?
Ironically, SI don't appear to have made that technical correction themselves in the 20200131 version of the International Edition that's available in their browser or SnowStorm Terminology Server. As a layman, I was surprised that only one of the new descriptions includes 'COVID-19'. However, there are more of such descriptions in the latest Australian Edition which can be seen using Michael's tool at http://ontoserver.csiro.au/vstool/.
Grahame Grieve (Mar 11 2020 at 02:18):
I guess you're actually saying that you want the preferred term? I guess I can do that
Grahame Grieve (Mar 11 2020 at 02:20):
Josh Mandel (Mar 11 2020 at 02:37):
"date" : "2020-30-11",
from the LOINC file looks like like a month/day swap.
Peter Jordan (Mar 11 2020 at 02:39):
...or accidental trailing 0 on the month element?
Josh Mandel (Mar 11 2020 at 02:41):
Yeah, that's better!
Josh Mandel (Mar 11 2020 at 02:41):
(Sorry, I thought this was a valueset from the future.)
Grahame Grieve (Mar 11 2020 at 02:43):
yes oops - typo
Grahame Grieve (Mar 11 2020 at 05:56):
why doesn't LOINC have a code for COVID-19 as observed by a CT of the lungs? That's currently how the diagnosis is being done in China right?
Grahame Grieve (Mar 11 2020 at 06:02):
nor do i have any idea how a normal person would differentiate between all these tests. I can't find a method-less test code, and I don't know how to differentiate the tests.. the method isn't exactly descriptive, for example
Can someone more up to date with this stuff give me any guidance here?
Grahame Grieve (Mar 11 2020 at 06:26):
@Swapna Abhyankar
Rob Hausam (Mar 11 2020 at 11:19):
I don't know what China is doing, but a chest CT would not be specific for COVID-19 - a positive finding could be COVID-19, or it could be influenza or any of a number of other viruses or it could possibly be an atypical or other bacterial infection. So the CT would only be useful as a screening test for COVID-19 in a very high prevalence situation - which is exactly where they have been lately (but maybe are starting to be not so much now). I think the appropriate LOINC code to use for that would be one of the already existing ones (I haven't looked for a particular one), not a COVID-19 specific code. But it will be good to know what Swapna's take is on it.
Grahame Grieve (Mar 11 2020 at 11:39):
no obvious one stands out for an observation
Swapna Abhyankar (Mar 11 2020 at 12:04):
@Grahame Grieve , I agree with what Rob said, a chest CT would not be specific for COVID-19. The LOINC term would be one of the existing chest CT terms, and if it was suspicious for COVID-19, that would be one of the findings.
There are no methodless LOINC codes because all of the tests are PCR-based (looking for the virus) or immunoassay (looking for the antibodies). We're making codes specific to the genes tested because that's important in the early stages of this in order to be able to evaluate the situation, and the labs doing the tests will know exactly which codes to pick.
Please let me know if you have any questions. I'm not sure which of the above comments apply to LOINC versus SNOMED CT versus general, so I may have missed something.
Robert McClure (Mar 11 2020 at 15:13):
@Grahame Grieve I've reached out to Suzy Roy from SNOMED but she's out through much of this month, so I'm looking elsewhere for official SNOMED input. That said, SNOMED has created a variety of concepts, beyond just conditions as listed here and I think if we have permission we should craft a general value set that contains all of them. I do see that the latest international version published 2020-03-09 browsable here has the updated preferred terms and your value set of conditions does not. I can help - let me know.
Grahame Grieve (Mar 11 2020 at 19:39):
hi @Swapna Abhyankar the problem is, not everyone who needs to record a positive test is a lab who knows how the positive test was done. there's the same need for a methodless test here as in any other case. I guess this means I can't use LOINC for a population based system that isn't getting a direct feed from a lab
Grahame Grieve (Mar 11 2020 at 19:40):
@Robert McClure I wasn't sure whether to create a general value set for all covid-19 related concepts from disparate locations in snomed or not. For me, it wouldn't be useful - I need to split out by condition | procedure | observation (loinc). But it could be useful if you wanted to preload terms I guess
Swapna Abhyankar (Mar 11 2020 at 20:19):
Hi @Grahame Grieve . If someone is self-reporting that they're positive for SARS-CoV-2, can you use the ICD-10 or SNOMED CT code for COVID-19? Also, even if there's no methodless code, if someone is tested for the virus in any respiratory specimen, you can use 94500-6 to record the result, or if the specimen is not known, 94309-2.
Grahame Grieve (Mar 11 2020 at 20:22):
If someone is self-reporting that they're positive for SARS-CoV-2, can you use the ICD-10 or SNOMED CT code for COVID-19?
yes, that would be appropriate for that fact
if someone is tested for the virus in any respiratory specimen, you can use 94500-6 to record the result, or if the specimen is not known, 94309-2.
hmm. those still are NAA/Probe - but I guess that's a reasonable assumption at this point
Grahame Grieve (Mar 11 2020 at 20:22):
thanks
Robert McClure (Mar 11 2020 at 23:29):
Swapna is essentially telling your that yes, all the tests are NAA/Probe because they are for now.
I've spoken to Jim at SNOMED and they are working on clarifying IP and how to push through new editions for those of us the expect to use them. In the USA we are _required_ to use the US edition. There is a good list for all the SNOMED SAR-COV-2 related codes here. The best spot to find them may change over time.
@Grahame Grieve It would be good if you updated to the current preferred displays in your value set. I'm working on determining the version identifier for this interim release.
Grahame Grieve (Mar 11 2020 at 23:32):
Grahame Grieve (Mar 11 2020 at 23:32):
I can't tell how these displays are not the current preferred displays
Robert McClure (Mar 11 2020 at 23:40):
I take that to mean you are not finding the preferred flag in the data?
Screen-Shot-2020-03-11-at-5.39.22-PM.png
Grahame Grieve (Mar 11 2020 at 23:42):
I was working from here: https://www.snomed.org/news-and-events/articles/changes-coronavirus-descriptions-(1) - is that wrong?
Rob Hausam (Mar 12 2020 at 00:09):
That should be a good source to work from.
Grahame Grieve (Mar 12 2020 at 00:13):
they appear to be different
Rob Hausam (Mar 12 2020 at 00:23):
This is the official announcement from today:
March 2020 SNOMED CT Interim International Edition release available
Dear all,The Production release of the March 2020 SNOMED CT® Interim International Edition is now available to everyone, from the MLDS distribution site.
This INTERIM International Edition Release has been created as part of the ongoing global effort to tackle the Coronavirus outbreak, and as such includes the updated Coronavirus descriptions for Members and implementers. This is a full International Edition Release package containing changes to the content detailed in the news item (https://www.snomed.org/news-and-events/articles/changes-coronavirus-descriptions) under the following conditions:
- following SNOMED International's critical incident policy, there will NOT be a recall of the January release, simply an update in response to international renaming, to ensure that SNOMED CT is current when assisting with the global Coronavirus effort
- implementers are encouraged to update to the new edition, however, they are NOT obliged to do so as the concept modelling and identifiers are not being modified
- the entirety of these changes will also be incorporated into the upcoming July 2020 International Edition Release
All files distributed for the March 2020 International Edition are available from the “March 2020 - v1.0 INTERIM RELEASE” version in the “SNOMED CT International Edition” project, which can be found on the “SNOMED Releases” page in MLDS, accessed here:
https://mlds.ihtsdotools.org/#/ihtsdoReleases
The SNOMED CT International Edition contains SNOMED CT files in Release Format 2 (RF2) only, distributed in a zipped file archive:SnomedCT_InternationalRF2_PRODUCTION_20200309T120000Z.zip
The Release Notes ("doc_SnomedCTReleaseNotes_Current-en-US_INT_20200309.pdf”) can be downloaded as distinct files from the same MLDS release version, or accessed online:https://confluence.ihtsdotools.org/display/RMT/SNOMED+CT+March+2020+International+Edition+-+SNOMED+International+Release+notes
Those Members using MLDS as their distribution site will find that their affiliates have access to the new release automatically.All vendors and affiliates should please contact their National Release Centre (NRC) for information on when the latest release package will be available for download from the NRC website. Contact details for all NRC's can be found here - http://www.snomed.org/members/.
Please refresh your memory of all changes to this release (and future releases), as published on the Early Visibility page here: https://confluence.ihtsdotools.org/display/RMT/July+2020+Early+Visibility+Release+Notice+-+Planned+changes+to+upcoming+SNOMED+International+Release+packages and in the Release Notes.
If you have any questions about the March 2020 International Edition or have difficulties downloading the files from the distribution site, please contact SNOMED International at support@snomed.org with “March 2020 International Edition release question” in the subject line.
Kind regards,
Andrew Atkinson | Release Manager
As it says, the SnomedCT_InternationalRF2_PRODUCTION_20200309T120000Z.zip file (from MLDS) is the official source.
Rob Hausam (Mar 12 2020 at 00:28):
Looks like it's available from the NLM now, so I can go ahead and update tx.fhir.org.
Grahame Grieve (Mar 12 2020 at 00:35):
yes please
Rob Hausam (Mar 12 2020 at 05:04):
done - SNOMED CT March 2020 Interim Release (International Edition) imported on tx.fhir.org
Grahame Grieve (Mar 12 2020 at 05:44):
btw, tx.fhir.org can host more than on version of the same edition as of this release
Rob Hausam (Mar 12 2020 at 05:46):
Good to know! I just made a comment about that in IG Creation here.
Rob Hausam (Mar 12 2020 at 05:47):
You'll have to fill me in on how that is set up - because I definitely want to do it.
Grahame Grieve (Mar 12 2020 at 05:54):
just list them in the ini file in order - first is default for the edition when referenced with out a version e.g.
nameX=type: snomed; source: [cache file 1]; default: true nameY=type: snomed; source: [cache file 2]
Grahame Grieve (Mar 12 2020 at 05:54):
only one of any of the snomeds can be overall default
Rob Hausam (Mar 12 2020 at 05:56):
great - I'll give it a try
Rob Hausam (Mar 12 2020 at 05:58):
is the order itself (not just the default) significant?
Grahame Grieve (Mar 12 2020 at 06:01):
yes - the first is the default version for that edition. after that order doesn't matter
Grahame Grieve (Mar 12 2020 at 06:01):
@Robert McClure Coming back to this, I have COVID-19 as the display for 840539006. I'm not at all sure why you think that's not right based on the current release:
Grahame Grieve (Mar 12 2020 at 06:02):
Grahame Grieve (Mar 12 2020 at 06:02):
it's the preferred synonym for USA?
Jim Steel (Mar 12 2020 at 06:03):
That's the correct one for SCT Intl 20200309. It was different ("Disease caused by 2019-nCoV") in 20200131 and any national extensions based on that.
Grahame Grieve (Mar 12 2020 at 06:18):
ok. summarised here: http://www.healthintersections.com.au/?p=2988
Patrick McLaughlin (NLM) (Mar 12 2020 at 09:00):
Patrick McLaughlin (NLM) (Mar 12 2020 at 09:00):
840544004 should be "Suspected COVID-19"
Grahame Grieve (Mar 12 2020 at 10:14):
hmm. the web page is wrong there (@Peter Williams )
Grahame Grieve (Mar 12 2020 at 10:15):
updated the value set and my blog
Peter Williams (Mar 12 2020 at 10:44):
Thanks @Grahame Grieve , on it.
Peter Williams (Mar 12 2020 at 11:56):
Thanks again for bringing that out of date information to our attention. The news page from Feb 13 now only contains a link to the updated information here: http://www.snomed.org/news-and-events/articles/march-2020-interim-snomedct-release-COVID-19
Rory Davidson (Mar 12 2020 at 12:36):
Just to add to this, we've put up a more user-friendly information page, http://snomed.org/covid-19 , about the content and map and also to note that these concepts will be made available in the next GPS release and in the meantime, can be used freely (under the Creative Commons Attribution 4.0 International License)
Rob Hausam (Mar 12 2020 at 13:15):
@Rory Davidson @Peter Williams So this is basically a "patch" addition to the current GPS? When we publish the full GPS value set (which I'm building right now) in the IPS IG in the next few days, should we go ahead and include this additional content (I'm thinking that sounds like a good idea)?
Rory Davidson (Mar 12 2020 at 13:18):
@Rob Hausam that's correct, although we are unlikely to produce an interim GPS release at this stage. So in the meantime, you can absolutely include this content wherever necessary. I can't comment about it being in the IPS though, which is only a subset of the GPS content. It will probably end up there, but that hasn't yet been discussed afaik.
Rob Hausam (Mar 12 2020 at 13:38):
Right, thanks. As a convenience (e.g. making it available as a FHIR value set instance), we are planning to publish the entire GPS as a value set itself (not directly used in any specific profiles). And we will want to include this new content in our "organisms-gps" value set that is part of the laboratory results observation profile.
Robert McClure (Mar 12 2020 at 14:13):
Many thanks to @Rory Davidson and Jim Case for the quick clarifications and making this easy for all to use. The SNOMED CT Coronavirus Content page is perfect. Everyone note that for this interim international release the version url is: http://snomed.info/sct/900000000000207008/version/20200309/
Grahame Grieve (Mar 12 2020 at 20:35):
I added this link to my blog post
Abbie Watson (Mar 12 2020 at 21:23):
Do we have procedure codes for coronavirus testing?
Grahame Grieve (Mar 12 2020 at 21:30):
I'm talking to both Snomed and Loinc about that now. Here's my summary of where I think we are:
- the LOINC codes can be used for test requests. The default LOINC code would be 94306-8 (SARS Coronavirus 2 RNA panel )
- but that's for the 'procedure' of the lab test
- neither snomed CT nor LOINC have a code for the patient test perspective - of getting a test done
@James T. Case @Suzy Roy @Peter Williams @Swapna Abhyankar move our email thread to here?
Swapna Abhyankar (Mar 12 2020 at 22:20):
@Grahame Grieve , sure we can move our email thread (not sure how to do that!).
About LOINC codes -
- We have tests for the SARS-CoV-2 virus and tests for antibodies to the virus (see https://loinc.org/prerelease).
- I would not say that 94306-8 is the default LOINC code because it depends on 1. whether it's a virus versus antibody test, 2. the specimen, and 3. whether a panel of tests is being done like the CDC assay, or if it's just one single test. 94306-8 represents the CDC and other panels in which multiple tests are done and the combination of results is used to produce a final result for whether or not the virus is present.
- The most generic LOINC for testing for the virus in an unspecified specimen is 94309-2. If a single test is being done, this code is used to order the test, and to report the result.
- Panel codes only serve as the order, and should not be used to report results. The individual codes in the panel are used to report each individual result.
Swapna Abhyankar (Mar 12 2020 at 22:53):
@Grahame Grieve ,
-
You can use 67803-7 History of Procedure to record what procedures were done, and we could create an LA (LOINC Answer) code for SARS-CoV-2 test.
We also have two terms to report the status of an order and an observation. These are not specific for SARS-CoV-2. -
92235-1 Lab order result status (https://loinc.org/92235-1/) --> answers include order received but specimen not yet received; No results available/procedure scheduled but not done; Preliminary; Order canceled etc.
- 92236-9 Lab observation result status (https://loinc.org/92236-9/) --> answers include Final; Specimen in lab;/results pending; prelminary; partial; etc.
Grahame Grieve (Mar 12 2020 at 23:00):
we could create an LA (LOINC Answer) code for SARS-CoV-2 test
well, you probably should do that. but it still feels to me not the code I'm looking for to use in Procedure.code.
Grahame Grieve (Mar 12 2020 at 23:02):
I'm not sure what the processes are right now - do clinicians ask for antibody vs virus test? or which specimen? Or is that deferred to the testing process?
Swapna Abhyankar (Mar 12 2020 at 23:10):
In the U.S. I believe that nearly all the testing is for the virus. And nearly all specimens being tested are respiratory. Virus testing tells you if the virus is present or not. Antibody testing gives you an indirect answer - that you were exposed to the virus and your body has mounted an immune response. IgM antibodies show that someone has recent/current infection, and IgG tests show past infection.
Grahame Grieve (Mar 12 2020 at 23:11):
right. but do they ask for virus testing directly? Really, though I'll have to ask here in Australia
Swapna Abhyankar (Mar 12 2020 at 23:13):
The specimen and which specific test is done depends on what is being recommended by the local jurisdiction/hospital/country.
Yes they ask for virus testing. Usually a "respiratory panel" to check for influenza and other respiratory viruses, plus a separate test for SARS-CoV-2...if the test is available.
Swapna Abhyankar (Mar 12 2020 at 23:25):
According to this PHLN is recommending respiratory testing for the virus. Antibody/serology testing is not available yet in Australia according to this document https://www.health.gov.au/resources/publications/phln-guidance-on-laboratory-testing-for-sars-cov-2-the-virus-that-causes-covid-19
Grahame Grieve (Mar 12 2020 at 23:27):
thanks
Matt Cordell (Mar 13 2020 at 04:25):
Just to add. The AU edition of SNOMED (February) also has a concept 1445431000168101|COVID-19 nucleic acid assay|.
The next release will have some specimen specific subtypes (sputum, nasopharyngeal swab etc)
We're also adding a "COVID-19 serology", but as far as we can see there's no such (blood) test available at this time.
Grahame Grieve (Mar 13 2020 at 05:35):
why add that to SNOMED CT if it's in LOINC?
Michael Lawley (Mar 13 2020 at 06:13):
Not all parts of the world use LOINC ;-)
Grahame Grieve (Mar 13 2020 at 06:41):
@Matt Cordell is that a procedure?
Jim Steel (Mar 13 2020 at 07:23):
Matt Cordell (Mar 13 2020 at 08:54):
Yes, a procedure. The PITUS requesting set use SCT procedures.
John Snyder (Mar 13 2020 at 13:16):
Matt Cordell and Grahame Grieve
Is 1445431000168101|COVID-19 nucleic acid assay in the international edition? The NameSpace identifier (1000168) would indicate that this may be in the Australian National Extension and I am not able to find it in either the main SNOMED CT browser or the Daily build browser under the international edition.
I appreciate the national extensions want to get out ahead of this, but shouldn't this be handled at the international level so that when we go to do data analytics we are looking for a single set of SNOMED concept IDs and not having to reconcile COVID-19 concept IDs from each national extension?
Thanks
John
Robert McClure (Mar 13 2020 at 14:46):
FYI, here is the US CDC page on CDC Tests for COVID-19.
@Swapna Abhyankar I see that the CDC names the test kit "Centers for Disease Control and Prevention (CDC) 2019-Novel Coronavirus (2019-nCoV) Real-Time Reverse Transcriptase (RT)-PCR Diagnostic Panel", can you confirm that is what you have in LOINC as SARS Coronavirus 2 RNA panel - Respiratory specimen by NAA with probe detection?
Based on that page, as of today, it does not seem that there are currently antibody serology tests available in the USA.
Swapna Abhyankar (Mar 13 2020 at 15:34):
@Robert McClure , the LOINC panel that corresponds to the "Centers for Disease Control and Prevention (CDC) 2019-Novel Coronavirus (2019-nCoV) Real-Time Reverse Transcriptase (RT)-PCR Diagnostic Panel" is 94306-8 SARS Coronavirus 2 RNA panel - Unspecified specimen by NAA with probe detection (https://loinc.org/94306-8/). That panel contains the specific N1 and N2 primer-probe codes that correspond to the CDC kit. The CDC did not want respiratory-specific codes.
The respiratory panel was created for IVD manufacturer-developed tests. We are asking the manufacturers if we can publish their names/test kit info and corresponding LOINCs on our website.
Grahame Grieve (Mar 13 2020 at 18:38):
ok I updated the page and will update further as @Swapna Abhyankar releases more information
Swapna Abhyankar (Mar 13 2020 at 20:47):
We just published an info page about LOINC/SARS-CoV-2 codes - https://loinc.org/sars-coronavirus-2/. Please check it out and let me know if there is anything important we missed. At the moment we have not gotten permission to list any manufacturers but will update if/when we get permission.
Grahame Grieve (Mar 13 2020 at 20:53):
I added a link to this page to my blog post
Swapna Abhyankar (Mar 13 2020 at 21:18):
Grahame Grieve said:
I added a link to this page to my blog post
Thank you!
Grahame Grieve (Mar 15 2020 at 19:55):
CPT have added a code:
87635 Infectious agent detection by nucleic acid (DNA or RNA); severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (Coronavirus disease [COVID-19]), amplified probe technique
The definition is very LOINC like - when you use one and not the other?
Robert McClure (Mar 16 2020 at 15:42):
@Grahame Grieve I agree it is kinda LOINCish, (@Swapna Abhyankar - thoughts?) but this seems to be a combination of a somewhat non-specific test AND the result being positive. So actually more of a finding, aka SNOMED. I do not think there is such a finding in SNOMED CT at the moment. @Suzy Roy - Welcome to zulip and thoughts about this?
Rob Hausam (Mar 16 2020 at 15:53):
It's a CPT code, so I'm sure that it's not intended to be a finding. When it says "Infectious agent detection by nucleic acid (DNA or RNA)" it's referring to the procedure (test) to detect the infectious agent. Even though the wording is somewhat obscure I don't think that it actually implies anything about whether the result of the test is positive or negative.
Robert McClure (Mar 16 2020 at 16:09):
@Rob Hausam Yes, you are probably right - a CPT code to get paid for doing the test is definitely the likely intent.
Suzy Roy (Mar 16 2020 at 16:16):
I'm going to agree with @Rob Hausam here and say that it sounds like it is referencing the test procedure rather than the finding of the infectious agent.
We do have 840539006 | Disease caused by severe acute respiratory syndrome coronavirus 2 (disorder) for the reporting of the actual result of the test.
And Hi @Robert McClure - thank you for the invite :)
James T. Case (Mar 16 2020 at 16:23):
There is a difference between the result of the test and the presence of disease (i.e. COVID-19). The use of 840539006 |Disease caused by severe acute respiratory syndrome coronavirus 2 (disorder)| should only be used in the presence of detection of the virus AND clinical signs. Without clinical signs there is no "disease" and many cases are asymptomatic. We will be releasing an "Asymptomatic infection with severe acute respiratory syndrome virus 2" concept in the July release.
James T. Case (Mar 16 2020 at 16:31):
@Grahame Grieve , what is meant by "...a code for the patient test perspective - of getting a test done"? I don' understand the context. And how would it be used?
Robert McClure (Mar 16 2020 at 18:19):
@Grahame Grieve - to be added to you blog: Just notified of a new (US-centric) HCPCS code:
From a PCORTF announcement: "CMS has developed a second Healthcare Common Procedure Coding System (HCPCS) code that can be used by laboratories to bill for certain COVID-19 diagnostic tests to help increase testing and track new cases. In addition, CMS released new fact sheets that explain Medicare, Medicaid, Children’s Health Insurance Program, and Individual and Small Group Market Private Insurance coverage for services to help patients prepare as well. HCPCS is a standardized coding system that Medicare and other health insurers use to submit claims for services provided to patients. Last month, CMS developed the first HCPCS code (U0001) to bill for tests and track new cases of the virus. This code is used specifically for CDC testing laboratories to test patients for SARS-CoV-2. The second HCPCS billing code (U0002) announced today allows laboratories to bill for non-CDC laboratory tests for SARS-CoV-2/2019-nCoV (COVID-19). On February 29, 2020, the Food and Drug Administration (FDA) issued a new, streamlined policy for certain laboratories to develop their own validated COVID-19 diagnostics. This second HCPCS code may be used for tests developed by these additional laboratories when submitting claims to Medicare or health insurers. CMS expects that having specific codes for these tests will encourage testing and improve tracking.
The agency also released three fact sheets that cover diagnostic laboratory tests, immunizations and vaccines, telemedicine, drugs, and cost-sharing policies. Medicare Fact Sheet Highlights, Medicaid and Children’s Health Insurance Program (CHIP) Fact Sheet Highlights, Individual and Small Group Market Insurance Coverage."
Grahame Grieve (Mar 16 2020 at 21:08):
@James T. Case I was in the hospital emergency dept last night - rapidly turning into a crazy (well, crazier) experience. So I overheard the triage nurse saying to a patient "OK, you meet the criteria (! - hardly anyone does these days), so we're going to get you tested. Go to [address] and get the test done right now"
Grahame Grieve (Mar 16 2020 at 21:09):
that procedure - that's what I want a code for. And I don't know what the code is. There's a LOINC code but that's actually for the lab procedure. The patient doesn't do the PCR themselves. And what they really do is get a swab done, but what's a code for 'swab for COVID-19' ?
Grahame Grieve (Mar 16 2020 at 21:16):
@Robert McClure updated
Grahame Grieve (Mar 16 2020 at 21:17):
I presume that the CPT code is just for billing and I wouldn't use it clinically
Robert McClure (Mar 16 2020 at 22:03):
@Grahame Grieve Actually I do think the CPT code is used clinically in the USA and in the outpatient world, which aligns with your example (and WTH where you doing in the hospital! Not a place to be...) That said, I'm not sure we normally capture the fact that a patient has gone to provide a specimen for a specific test. Perhaps what you are asking for is a finding that the specimen has been collected for use in determining SAR-CoV-2 infection status?
Grahame Grieve (Mar 16 2020 at 22:05):
I'm looking for a code for 'please do this' and 'this has been done' where 'this' is provide a specimen for a specific test. "I'm not sure we normally capture the fact that a patient has gone to provide a specimen for a specific test" - I'm really surprised since it matters in many clinical workflows, and is legally mandated in a few (e.g. drug tests)
Grahame Grieve (Mar 16 2020 at 22:06):
and I was in hospital with my daughter who had an acute asthma attack (looked liked covid-19 minus the fever)
Swapna Abhyankar (Mar 16 2020 at 23:19):
@Grahame Grieve , I hope your daughter is doing okay.
Agree that the CPT code is for billing. The CPT is just saying that a molecular test used to detect SARS-CoV-2 was done, without saying which test. It shouldn't be used in clinical care other than to record that a test was done. If it's done in the doctor's office (not happening now), the CPT would be submitted to insurance by the doctor's office, along with whatever other tests were done (Flu, RSV). If it's done at a lab, the lab would submit the CPT to the insurance company, but that CPT code wouldn't wind up in the patient's primary record.
'Please provide a specimen' - as far as I know there's no code for this. It might be documented in the clinical note ("I sent Mr. A to lab X to get tested for Z") , but it's not directly coded. However, it is captured indirectly in the form of an order either as -
- An electronic order sent to the performing facility.
- An order written on paper - "Please go to lab X and get a nasal swab for SARS-CoV-2" and documented in the clinical note.
'A specimen was provided' - again, you can get to this indirectly -
- If there's an electronic order, it can be tracked from order placement, specimen collection, prelim results, final results, etc. If the patient doesn't show up, you still have a record of no specimen being provided.
- If there's no electronic order, it can't be tracked until the patient actually goes to the performing facility. If they never show up, the performing facility will never know that there was an order or a specimen that needed to be collected.
Until recently, most patients weren't documenting medical information themselves, which may be a big part of why there aren't any codes from the patient perspective.
Grahame Grieve (Mar 16 2020 at 23:21):
right....
Grahame Grieve (Mar 16 2020 at 23:21):
(oh and Mel is fine now)
Robert McClure (Mar 17 2020 at 00:07):
Now there are SCT codes for specimen collection (a procedure) but none that I know of that are specific to a _reason_ for the collection. I suppose we could create an procedure request that had all that info in it, but I doubt we'd get a code with all that in it. Perhaps @James T. Case or @Suzy Roy could chime in. As for the 'this has been done' code - that is actually a finding.
Grahame Grieve (Mar 17 2020 at 03:04):
a procedure with a status of complete, or an observation with a finding? wouldn't that depend on your perspective ?
James T. Case (Mar 17 2020 at 14:17):
So, @Grahame Grieve , it seems like what you are looking for is a specimen collection procedure, with all the appropriate life cycle statuses that that entails. Is that correct?
James T. Case (Mar 17 2020 at 14:22):
@Robert McClure , we do not pre-coordinate a "reason" for a specimen collection procedure with the actual procedure itself. I would think that within the medical record there would be a chief complaint or reason for visit as well as a differential diagnosis that would indicate why a specimen was taken, no?
James T. Case (Mar 17 2020 at 14:24):
@Grahame Grieve re: "a procedure with a status of complete, or an observation with a finding? wouldn't that depend on your perspective ?"...My experience in lab is that a lab procedure was not "complete" until the finalized, reviewed and approved results were released.
Grahame Grieve (Mar 17 2020 at 18:44):
@James T. Case yes that's how a lab procedure works. But the test collection procedure is different. I don't think that it's reasonable to infer the purpose of specimen collection from the chief complaint
James T. Case (Mar 17 2020 at 18:55):
@Grahame Grieve , what would be a use case for requesting a specimen collection without also ordering the associated testing procedure?
Grahame Grieve (Mar 17 2020 at 19:29):
I don't think there's any use case for that. But I think there's a use case for recording the ordering of a specimen collection for a purpose without knowing what the specific laboratory test codes are
James T. Case (Mar 18 2020 at 00:35):
@Grahame Grieve , specific test codes, yes, generic testing as an order, no. That is why there are generic order codes that allow for discretion on the part of the lab as to what they fulfill.
Matt Cordell (Mar 18 2020 at 00:40):
@Grahame Grieve Are you asking for a code vague/generic "collecting something for SARS2"? I'd view this as the same as just saying "Test for SARS2". (kind of what Jim just said).
Also "collecting the specimen" is an implicit part of requesting the test - whether collected by Doctor or collection centre - the request is the same. (Though there's a cost code associated with specimen handling IIRC, at least in AU)
Matt Cordell (Mar 18 2020 at 00:47):
@James T. Case Will that "Asymptomatic infection with severe acute respiratory syndrome virus 2" be in the Daily in the next couple of weeks?
Also, we've generally been saying COVID-19 is the preferred term for most concepts (acknowledging it's not necessarily accurate, but reflecting popular usage). So we'd probably call that "Asymptomatic COVID-19 infection"
James T. Case (Mar 18 2020 at 01:03):
@Matt Cordell , I reviewed and approved the "Asymptomatic" concept today, so it should show up in the daily build in the next few days as it goes through the promotion process. Asymptomatic COVID-19 is included as a description on the concept.
Grahame Grieve (Mar 18 2020 at 05:14):
collecting the specimen" is an implicit part of requesting the test - whether collected by Doctor or collection centre - the request is the same
So my point is 2-fold: (a) it's something that is completed and done independently of the whether the lab work is done or not. and (b) you might know that you collecting a specimen for a purpose without knowing which specific lab test is being done (though presumably someone would know that)
James T. Case (Mar 18 2020 at 17:25):
In my experience (welcome to hear other points of view) it is the ordering of the test that determines the specimen type that is collected. It would be rare that anyone would say "take this specific specimen type for this testing purpose". It is the requested test that drives the collection.
Grahame Grieve (Mar 18 2020 at 19:30):
that's my experience too. And inside the system that's how it works. But that's not the perspective from the consumers point of view
Grahame Grieve (Mar 18 2020 at 19:30):
but what I'm hearing is that this pov doesn't matter for SNOMED
James T. Case (Mar 19 2020 at 23:32):
I suppose I just do not understand the use case
Grahame Grieve (Mar 28 2020 at 18:55):
More on Covid-19: http://www.healthintersections.com.au/?p=3017
@James T. Case @Suzy Roy this is still exploring the patient perspective; I don't know if SCT is interested in new codes for the appointment process.
James T. Case (Mar 28 2020 at 22:48):
@Grahame Grieve , just a question on terminology of terminology. Is there a perceived difference between a virtual appointment and a "telemedicine appointment"? From an encounter perspective, we unfortunately have a mixture of encounters and consultations of various types. Do you consider these different? We are certainly interested in updating our terminology to better support telehealth in light of the major move in this area.
Grahame Grieve (Mar 29 2020 at 09:48):
hey @James T. Case I do not know. I think that most people would think that 'telemedicine' appointment was different to a 'telephone encounter' but I don't know whether they'd automatically include a video consultation in that... I suspect it might differ between countries
Grahame Grieve (Mar 29 2020 at 09:49):
I can tell you that Australian GPs are focused on the phone, and not seeing the patient if they don't have to (easier to multi-task, I expect), and only switch to video when there's something specific to look at.
They, then, would think of all this as telemedicine.
Grahame Grieve (Mar 29 2020 at 09:50):
I don't know whether a clarification of "Telephone Encounter" is possible or appropriate.
Frank Oemig (Mar 30 2020 at 12:14):
Ad hoc I would say that 'virtual' and 'tele-something' are the same, although we may distinguish by phone, computer, w/ or w/o video etc...
Patrick McLaughlin (NLM) (Apr 01 2020 at 14:13):
The April 2020 release of RxNorm will contain the ingredient (IN): remdesivir (RXCUI 2284718), and the Semantic Clinical Drug (SCD): remdesivir 100 MG Injection (RXCUI 2284960).
https://www.nlm.nih.gov/pubs/techbull/ma20/ma20_rx_norm_investigational_drugs.html
Last updated: Apr 12 2022 at 19:14 UTC
