Stream: Orders and Observation WG
Topic: DiagnosticReport.category
Bob Milius (Aug 27 2019 at 21:10):
I'm interested in the appropriate use of DiagnosticReport.category
. DR allows for multiple categories. The GenomicsReporting IG requires one of those to be Genetics. Is it appropriate to use it to describe the gene family/group that the DG is about? eg. using HGNC gene group for HLA typing? Is this an appropriate use for category? e.g.,
"category": [ {"coding": [ { "system": "http://terminology.hl7.org/CodeSystem/v2-0074", "code": "GE", "display": "Genetics" } ]}, {"coding": [ { "system": "http://terminology.hl7.org/CodeSystem/v2-0074", "code": "LAB", "display": "Laboratory" } ]}, {"coding": [ { "system": "https://www.genenames.org/data/genegroup", "code": "588", "display": "Histocompatibility complex (HLA)" } ]} ]
Rob Hausam (Aug 28 2019 at 03:32):
What would you be using for the value of 'code' in this case? The 'category' element would generally be used to provide "tags" that can be used to simplify searching and determining how/where to display the data in your app, etc., but it can't be used to provide additional context for what is being represented in 'code' - the value of 'code' is what the DiagnosticReport is about.
Bob Milius (Aug 28 2019 at 14:14):
That's another thing. The Genomics Reporting IG DG has code.coding constrained to include "Master HL7 genetic variant reporting panel". I thought to add the specific genes of the HLA gene family which are found in the observations, as additional codings since they are subsumed by the panel. So category would say this an HLA gene family genetic laboratory report, and the codings would indicate which HLA genes are being reported in the observations. Are these appropriate uses of category and code?
"code": {"coding": [ { "system": "http://loinc.org", "code": "81247-9", "display": "Master HL7 genetic variant reporting panel" }, { "system": "http://loinc.org", "code": "13298-5", "display": "HLA-A [Type]" }, { "system": "http://loinc.org", "code": "13299-3", "display": "HLA-B [Type]" }, { "system": "http://loinc.org", "code": "13302-5", "display": "HLA-C [Type]" }, { "system": "http://loinc.org", "code": "57298-2", "display": "HLA-DRB1 [Type]" }, { "system": "http://loinc.org", "code": "53938-7", "display": "HLA-DQB1 [Type]" } ]}
Bob Milius (Aug 28 2019 at 14:25):
basically, I want to be able to search for all DiagnosticReports that pertain to a gene family (e.g., HLA), and also search for DRs that are about specific genes (e.g., HLA-A) without having to dig into the observation results referenced in the DR.
Bob Milius (Aug 28 2019 at 14:43):
Alternatively, I could just use DiagnosticReport.code to include gene family and genes as identified by genenames.org (both subsumed by "Master HL7 genetic variant reporting panel"), e.g.,
"code": {"coding": [ { "system": "http://loinc.org", "code": "81247-9", "display": "Master HL7 genetic variant reporting panel" }, { "system": "https://www.genenames.org/data/genegroup", "code": "588", "display": "Histocompatibility complex (HLA)" }, { "system": "https://www.genenames.org", "code": "HGNC:4931", "display": "HLA-A" }, { "system": "https://www.genenames.org", "code": "HGNC:4932", "display": "HLA-B" }, { "system": "https://www.genenames.org", "code": "HGNC:4933", "display": "HLA-C" }, { "system": "https://www.genenames.org", "code": "HGNC:4948", "display": "HLA-DRB1" }, { "system": "https://www.genenames.org", "code": "HGNC:4944", "display": "HLA-DQB1" } ]}
Bob Milius (Aug 28 2019 at 20:46):
The more I think about it, the more I think this use of DR.code is wrong. While the individual HLA genes may be subsumed by "Master HL7 genetic variant reporting panel" and the HLA gene family/group, each of the HLA genes are not identical concepts with each other. If only DiagnosticReport had components...
Eric Haas (Aug 28 2019 at 20:50):
DiagnosticReport doesn't nest but there are extensions to point to do this.
Rob Hausam (Aug 28 2019 at 21:26):
I agree with you that this use of DR.code is wrong. But I think there are ways within the existing mechanisms to structure the reporting to achieve what you want. I'm not sure that you would need DR components to do it. If I understand your example and use case well enough (not certain about that), I think you should be able to do it with higher level observations (one or more) for the panels as DR.result and then the appropriate use of nested observations and/or components beneath that for the further details. It would be worth exploring to see how that works (or doesn't).
Bob Milius (Aug 28 2019 at 21:48):
Thanks. We already have a "Region Studied" observation profile in the Genomics Reporting IG that can hold this gene level information, but I was just trying to find way of looking at just the DR at a high level without looking into observations (assuming the DR is a summary of all of the observations). But this may be enough if I can do the following, which basically says, "this is a genetic reporting panel about the HLA gene family." Here I'm asserting that "Histocompatibility complex (HLA)" is subsumed by the ""Master HL7 genetic variant reporting panel."
{ "category": [ {"coding": [ { "system": "http://terminology.hl7.org/CodeSystem/v2-0074", "code": "GE", "display": "Genetics" } ]}, {"coding": [ { "system": "http://terminology.hl7.org/CodeSystem/v2-0074", "code": "LAB", "display": "Laboratory" } ]} ], "code": {"coding": [ { "system": "http://loinc.org", "code": "81247-9", "display": "Master HL7 genetic variant reporting panel" }, { "system": "https://www.genenames.org/data/genegroup", "code": "588", "display": "Histocompatibility complex (HLA)" } ]} }
Grahame Grieve (Aug 28 2019 at 23:02):
I would have expected that to the code
Last updated: Apr 12 2022 at 19:14 UTC