Welcome to the discussion forum for LOINC Groups. We’d like to hear your feedback on the file format and the LOINC Groups that have been created. Additionally, we’d love to know how you intend to use this artifact and any suggestions for new LOINC Groups.
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?
Since this is a new space for LOINC, the LOINC team is very interested in community feedback, so we will be organizing methods to do so over the course of the next several months. The current plan is to release another version (hopefully Beta, but perhaps Alpha 2 depending on the feedback) with the December 2017 LOINC release.
New “human friendly” view of LOINC Group File posted
Our colleagues at NLM thought it’d be nice to explore the content of the LOINC Group file in an Excel spreadsheet with the subtotal function showing the counts of each group.
We’ve posted this spreadsheet view here:
LOINC 2.61 GroupFile Alpha 1 grouped with counts
requires a free loinc.org login
With this format you can easily expand and collapse the groups in a tree-like structure.
Notes:
This spreadsheet has the columns for parent group and other fields hidden.
Also, don’t try to import this spreadsheet into a database - you’ll get the subtotal rows too. Better to use the main release format: LOINC Group File, Alpha 1
Do you have groups for an analyte regardless of unit of measure?
Very nice job, once LOINC GROUP is completed, it could be used in Thai’s medical reimbursement system in order to drive all hospitals to use LOINC in HIS.
Good to see this work. Here at Wolters Kluwer Health we are evaluating its value in helping enhance our CDS capability. I hope to have the chance to engage with others as we determine if this grouping can help us solve our issues with the flat, non-hierarchical nature of LOINC.
Hi Charles -
As you probably know, LOINC codes are not specific to a particular unit of measure. The Property does constrain which units are allowable for a given LOINC term. For example, and mass/volume unit could be used with a LOINC term that has a Property of MCnc. The units fields in the LOINC table are all meant as “examples”; they are not normative
We are making a few high level groups for LOINC codes that collect terms across Properties. One area of focus initially are groups that collect MCnc and SCnc terms for the same analyte. With the molecular weight, such a group would allow equivalence.
Take a look at the ParentGroup: LG55-6 MassMolConcSAME:Comp|Tm|Sys|MethProp:MCncORSCnc
It forms Groups that pull together measures of the same analyte with either MCnc or SCnc:
LG10153-1 Bilirubin|Pt|Urine|Test strip|MassMolEquiv|584.662 g/mole
20505-4 Bilirubin.total [Mass/volume] in Urine by Test strip
41016-7 Bilirubin.total [Moles/volume] in Urine by Test strip
I’ve got a question for our public health colleagues using LOINC. Would it be useful to have groups that pull together tests on isolates from “normally sterile sites”?
If so, is there an agreed list of such specimen sites? (I see, for example, one from MN: http://www.health.state.mn.us/divs/idepc/dtopics/invbacterial/sterile.html)
Amazing job! Maybe it could facilitate the development of common code subsets like Top2K+ in other countries/regions etc.
Multiple levels of granularity may be needed for various use cases.
We’d love to receive your feedback! In this file, you will find new groups. Particularly, we would be interested in your feedback on the following groups: a large group that rolls up six classes, a group for microbiology antibody types/method/property, reportable conditions, radiology, and document ontology.
What you won’t see in the attached file are the LG codes, but do not fear! We have been working on ways to persist the codes in our database and the December release will likely contain permanent LG codes. Since they are not ready for prime time, we removed them from this file. You also will not sure description information, which will be in final artifacts.
Community guidance is critical for the ultimate success of the new LOINC Groups so any thoughts you have are appreciated.Click here to download the LOINC Groups file.
LOINC Groups might be used to create LOINC value sets.
Typical use cases and examples would facilitate understanding of LOINC Groups and their use/value.
Visualization of the hierarchical structure of LOINC Groups would also be helpful to quickly/briefly master LOINC Groups.
Further categorization/classification of LOINC Groups would facilitate navigation of LOINC Groups.
Normally sterile sites - as per their classification is a subset of LOINC attribute “System/ specimen”. So are you proposing to come up with group of LOINC attributes - System/ Specimen?
It is not a bad idea, but then, if we go that route to classify groups, then it is endless list.
I though, using FILTER function in Excel - to get LOINC codes by Specimen is far more easy.
What do others say?
Btw - the idea of LOINC groups is great. Thumbs up. It will make coding a lot easier.
Hello,
Thank you for all you (the greater LOINC you) do.
Here’s the good news. When I discovered the LOINC_2.61_GroupFile_Alpha_1_grouped_with_counts download file with the Loinc Groups for nursing specific documentation, e.g. diastolic blood pressure, and body weight, I felt like I’d won the LOINC lottery. For the first time I could see all the codes used in one place. Although I had to filter out the lab Loinc Groups to only see the nursing specific content, it was worth the time.
Now the bad news. With the release of Loinc 2.64, 6/15/2018, not only do I not find a document labeled LOINC 2.64 GroupFile grouped with counts, as I would expect to find, so that I can see what I found with LOINC_2.61_GroupFile_Alpha_1_grouped_with_counts, but by the time I find the Loinc Groups located in a file named GroupLoincTerms, I discovered that the Loinc Group ID number had changed. How did I find this out you are wondering? When I searched on the GroupId column for LG10304-0, Body height|Len|Pt|^Patient (from the LOINC_2.61_GroupFile_Alpha_1_grouped_with_counts download file, there was no match. However, when I searched on Body height, there they were, with a different GroupId.
I am wondering why the Loinc Group Id changed from one version to the next? I am seeking guidance on how a nurse, in the absence of consistent Loinc Group Ids, can efficiently utilize Loinc to view and consume clinical observation codes without an obvious LOINC provided way? Loinc Panels are interesting to look at but do not provide a user friendly or consumable organization structure for observation codes used for clinical concepts.
Thank you in advance for your time and attention to my inquiry.
Rebecca
Hi Rebecca,
Thank you for your interest and comments about the LOINC Group work.
I completely agree with you that having the same ID over time is essential, and starting with the December 2017 Group release, the Group IDs (LG numbers) have been stable, so that the LG numbers for a Group in the December release is the same as the LG number for that Group in the new June (2.64) release.
The file you originally looked at (LOINC_2.61_GroupFile_Alpha_1_grouped_with_counts) was not part of our official release package, though it did contain content from our initial alpha release. It was created by our colleagues at NLM and posted by Dan Vreeman in this same thread on July 5, 2017. Because the counts file wasn’t part of our official release, it did not come with the Readme and Release notes, which specifically said that users should expect the contents of the file to change. Given that it was the first release of the Group work, we did not know what feedback we would get or how the structure or contents of the file would change, so we wanted to be able to change any part of it for the next release.
For the December 2017 release, we significantly changed the structure of the Group release artifact (from one CSV to several) in order to decrease data redundancy and make the contents more computable. In the Release notes for that release, we said that starting with that version, the LG numbers would persist.
The version of the file with the counts is not part of our official release package because given its formatting, subtotal functions, and hidden rows, it cannot easily be uploaded into a database or used for data analysis, and similarly, creating the file is a manual effort.
I hope this information helps, and be assured that going forward (starting with December 2017), the LG numbers are stable.
Please let me know if you have any questions.
Hi Swapna,
Thank you very much for your prompt and helpful reply.
I was thinking that was probably the case about the GroupID. I did notice they were the same for 2.63 - 2.64. That document for 2.61 was such a great view of the Loinc groups. I understand not having the resources or the need to create it for each release.
I’m wondering what your response is to the second part of my inquiry about what is the best way to view and consume the data for clinical observables? What, if any, of the downloadable content is the most useful for the clinical use case?
Also, I was wondering why the decision was made to exclude challenge codes from the groups? Or body site codes? As a clinician, I would want to view and consume all of the codes for documenting of vital signs, say, for example, systolic blood pressure-sitting or Ulnar artery - left Systolic blood pressure.
Is there an efficient way to view and aggregate clinical observations, specifically, physiologic (e.g. ejection fraction) and physical (e.g. measurements) observations? I know that there are the Panels in RELMA but the codes are not viewable with that function. Perhaps a component to filter on? Being a nurse, I am not interested in the Lab codes.
Again, thank you so much for your time and attention to my inquiry.
Regards,
Rebecca
Hi Rebecca,
Regarding the best way to view and consume clinical observables data, that will depend on your local implementation of the Groups and your EHR system. And please keep in mind that the LOINC Groups must be reviewed and validated for your specific use case.
Looking at the Parent Groups should help you decide which ones might be useful for your particular use case because the Parent Group name will tell you whether the individual Groups are related to laboratory, radiology, vital signs, etc.
We were creating Groups based on our clinical experience and feedback we have received from the LOINC Community over time. It is still a work in progress, and we haven’t deliberately excluded any particular types of LOINC codes, such as challenge terms, except maybe from specific existing Group definitions where including the challenge terms did not seem useful for that particular Group.
If there are Groups that you think would be useful, such as a broader vital signs grouping, please submit your ideas through our Groups Community submission portal (https://loinc.org/groups/community/).
You can search by any of the 6 major LOINC Parts (as well as many other fields) in both RELMA and search.loinc.org. So for example, if you are looking for ejection fraction codes, use Component:“Ejection fraction”. This search will return radiology codes for cardiac studies as well as ejection fraction measurements. If you want to restrict by the measurements, add the scale to the search, e.g., Component:“Ejection fraction” Scale:Qn.
If in general you want to exclude all laboratory terms, you can add -Type:1 to all of your searches. The RELMA and online search help both have a lot of information about search strategies, including which fields you can search on and how to include and exclude terms from your search.
Thanks again for your interest in the LOINC Groups!
Hello, is the LOINC group file still available to download?
Hi Anum,
Yes, the file is still available as part of our complete LOINC file, available at https://loinc.org/downloads/. The Group file remains in Beta. You can also search Groups at https://loinc.org/groups/.