Vojtech Huser

Forum Replies Created

Viewing 15 posts - 1 through 15 (of 15 total)
  • Author
    Posts
  • in reply to: SARS-CoV-2 codes discussion #435559
    Vojtech Huser
    Participant

    During OHDSI COVID19 study-a-thon – we found out that LOINC codes help us only half the way. It is good that we have a code for RNA test, but the result of that test can still vary and be coded as ‘Detected’ or ‘Positive’ or ‘Not Detected’ or ‘Negative’. For qualitative tests, recommending via a nudge (not requiring) some possible result values (just like example units) may be a way to do the last mile of standardization.

     

    See the details about coded values as results here: https://forums.ohdsi.org/t/covid19-etl-help-for-converting-your-data-into-omop/10269

    in reply to: SARS-CoV-2 codes discussion #435333
    Vojtech Huser
    Participant

    This wikipedia page has gene targets. Generic code that covers them all (or separate code for each line in the table) may be needed.
    <p style=”box-sizing: border-box; font-family: Arial, Helvetica, sans-serif; line-height: 24px; font-stretch: normal; color: #3c4245;”><strong style=”box-sizing: border-box;”>Summary table of available protocols</p>

    <strong style=”box-sizing: border-box;”>Country <strong style=”box-sizing: border-box;”>Institute <strong style=”box-sizing: border-box;”>Gene targets
    China China CDC ORF1ab and N
    Germany Charité RdRP, E, N
    Hong Kong SAR HKU ORF1b-nsp14, N
    Japan National Institute of Infectious Diseases, Department of Virology III Pancorona and multiple targets, Spike protein
    Thailand National Institute of Health N
    US US CDC Three targets in N gene
    France Institut Pasteur, Paris Two targets in RdRP

     

    <a href=”https://en.wikipedia.org/wiki/COVID-19_testing#Antibody”>https://en.wikipedia.org/wiki/COVID-19_testing#Antibody</a&gt;

     

    in reply to: SARS-CoV-2 codes discussion #435329
    Vojtech Huser
    Participant

    The use case is to create as few as possible OMOP codes outside LOINC.

    OMOP CDM uses LOINC and embraces it.

    Recently, there are efforts to better capture LOINC parts in OMOP Vocabulary. See here https://forums.ohdsi.org/t/improve-loinc-representation-in-omop-vocabulary-athena-link-parts-to-lab-tests/7925

     

    Knowing the LPs (including temporary ones) is crucial for following correctly LOINC concept model.

    LOINC will at some point adopt some framework for postcoordination.

     

    As terminologist, I have few thoughts on how from components one could create good classification codes.

    E.g., test from sputum for SARS-CoV-2, test from sputum for SARS-CoV-2 RNA,

     

    The LOINC search supports such nice searchers (unfortunately not for temporary codes). e.g. https://search.loinc.org/searchLOINC/search.zul?query=hiv+1+RNA+Methodless%3Atrue

    search is: ‘hiv 1 RNA Methodless:true’

     

    The release of those codes was coordinated with SNOMED CT (per here http://www.snomed.org/news-and-events/articles/changes-coronavirus-descriptions ) which is great. I like that LOINC has temp codes published and can respond more flexibly. No temp code page for SNOMED CT 🙁

     

    I would also like to propose a LOINC group for SARS-CoV-2 tests. (and have concept of temp groups (inbetween releases).

    I would like to propose at least one (or several) methodless codes for SARS-CoV-2

    In fact there should be a policy for methodless codes accross all of LOINC lab content. (and it should be consistent) (or some postcoordination grammar (expression language) where my expression is deemed equivalent to an existing pre-coordinated code.)

     

    I would also like to see parent-child hierarchy between LP for ‘SARS-CoV-2 RNA’ and LP ‘SARS coronavirus 2 N gene’

    And be able to search for any component that is child of ‘SARS-CoV-2’ or ‘SARS-CoV-2 RNA’

     

    (that in fact assumes temp codes do show up in <b>the </b>LOINC search page.)

     

    p.s.

    By the way, upgrading this forum to some newer platform would be great. (allow ctr+V to insert image in my post)

     

    in reply to: Fully specified names are not escaped #429459
    Vojtech Huser
    Participant

    Why are you not using the native .zip file from LOINC.org ?

    Vojtech Huser
    Participant

    I am also very interested in LOINC codes for HIV tests. (and would be keen to see answer to your question)

    in reply to: LOINC and LabCorp and Quest #24228
    Vojtech Huser
    Participant

    thans for reply

    in reply to: VSAC #21390
    Vojtech Huser
    Participant

    will it be included in VSAC now when it is final?

    in reply to: Call for Feedback on LOINC Groups #18160
    Vojtech Huser
    Participant

    The new grouping of loinc codes is a great new feature.
    Given it is in alpha:
    What is the roadmap for it? Beta by October 2017?
    Full production in Jan 2018?

    Vojtech Huser
    Participant

    I like very much the new policy of creating generic terms where specific exists.
    I also feel like my forum post made a difference. Very encouraging.

    in reply to: Most Accurate Document Class #16554
    Vojtech Huser
    Participant

    I think these column are important for billing. Some are used as claims attachment.

    So I would ignore this classification into (DOC.CLINRPT, DOC.MISC, ATTACH.GENERAL and ATTACH.CLNRPT) and if a concept is a fit – why not use it.

    Vojtech Huser
    Participant

    Our mapping efferts advanced and we have a good mapping of top 80% of document types (2 mappers) and a draft mapping for all 174 distinct document types in our IDR.

    Few things we are finding out:
    – legacy doc. types and active doc types (mapping old to new using LOINC DO)
    – some local doc. types could be renamed to fit better
    – some new LOINC codes can be created and we will submit those
    – LOINC part codes vs. LOINC codes and the higher level hierarchy – sometimes it would be nice to map our doc. type to a class concept (parent), e.g., like for example lab panel

    here are the new concepts which might be interesting:

    Nursing doc: Vascular Access Device Observation Document
    Nursing doc: Skin and Hygiene Observation Document

    I downloaded the template for submission (Basic Template – MS Excel (.xls))
    but the file is more optimized for lab LOINC.
    What template should I use for clinicalLOINC DO domain?

    see the fiels here:
    REFERENCE_ID NAME ORGANIZATION PHONE FAX EMAIL ORG_SOURCE_CARE_ORG PROJECT_DESCRIPTION LOCAL_BATTERY_CODE LOCAL_BATTERY_DESCRIPTION LOCAL_TEST_CODE LOCAL_TEST_DESCRIPTION REFERRAL_LAB_CODE REFERRAL_LAB TEST_DESCRIPTION LOCAL_TEST_UNITS TEST_INSTRUMENT TEST_INSTRUMENT_MODEL TEST_REAGENT_KIT SIMILAR_LOINC UNITS EXAMPLE_ANSWERS ANSWER_COMMENTS EXAMPLE_REPORT GENERAL_COMMENTS REFERENCE_INFO NORMAL_RANGE ANALYTE DIVISOR SUFFIX CHALLENGE ADJUSTMENT COUNT PROPERTY TIME_ASPCT TIME_MOD SYSTEM SUPER_SYS SCALE_TYP METHOD_TYP

    Our excel file would have these fields:
    LOINC_NUM COMPONENT PROPERTY TIME_ASPCT SYSTEM SCALE_TYP METHOD_TYP

    in reply to: Names of Clinical Reports #16475
    Vojtech Huser
    Participant

    I am pleased to report that Well Child Visit document type is now in LOINC !

    Is there a list of how to interpret the source acronym.

    for example what these mean:

    PHS
    UPMC
    UPMC
    DJV.MC
    DJV.MC

    in reply to: Proposed harmonization of existing values to new #16431
    Vojtech Huser
    Participant

    Nice work.
    Is there a way to see which terms have changed because of this harmonization.

    in reply to: Definitions for DocOnt axis values #16433
    Vojtech Huser
    Participant

    At Marshfield Clinic, we also have a document type of “consent” and have the same problem.
    Has there been a consent term suggested to LOINC (or created)

    Vojtech Huser

    in reply to: Names of Clinical Reports #16474
    Vojtech Huser
    Participant

    Daniel,

    Thank you for reply. I will try to follow up on this with the LOINC team.

    Vojtech

Viewing 15 posts - 1 through 15 (of 15 total)