Condition mapping improvement using SNOMED Extension proposal

[MPhilofsky (Melanie Philofsky)]

@Dymshyts,

I’m curious on your process to

To add SNOMED-like concepts with SNOMED-like relationships and hierarchythat allows us to make accurate condition mappings.

Do you use any tools? Or is this a manual process? Have you assessed SOLOR? From my understanding, this open source tool will eliminate the

Case 1. Mapping to a more general concept

and

Case 3. Mapping to a same-meaning concept but with different hierarchy (ICD10CM and SNOMED uses different principles for hierarchy sometimes)

by using description logic and logical expressions to extend the vocabulary. Stanford Protege might also be of use.

Melanie

[abedtash_hamed (Hamed Abedtash)]

@MPhilofsky I’ve recently heard about SOLOR. Do you know if it has the mappings between SNOMED CT and ICDs? Or, it is just about SNOMED, LOINC, and RxNorm?

[Christian_Reich]

I’m seeing the effect in few of our studies.

Can you show us those examples? We will need them to make the case for adding SNOMED Extension concepts. It’s a departure from our strong principle to not create (non-administrative) Concepts.

[Dymshyts (Dmytry Dymshyts)]

Do you know if it has the mappings between SNOMED CT and ICDs? Or, it is just about SNOMED, LOINC, and RxNorm?

according to their web-site info:
“Yes, SOLOR does not map one terminology system to another.”

Do you use any tools? Or is this a manual process?

it’s SQL-based algorithm, but, yes for most of the mappings decision was made manually, because it’s need to be clinically evaluated.

Have you assessed SOLOR? From my understanding, this open source tool will eliminate the

Have you try to run it? Do you know how to use the tool?
The same question about

Stanford Protege

There’s a lot of documentation, so some guidance from someone who used it would be useful

[MPhilofsky (Melanie Philofsky)]

@abedtash_hamed and @Dymshyts,

Do you know if it has the mappings between SNOMED CT and ICDs? Or, it is just about SNOMED, LOINC, and RxNorm?

SOLOR integrates SNOMED, LOINC, and RxNorm into one common terminology model (not sure if terminology model is the correct term). And can be used to extend the terminology. Something that is being proposed in this thread. On paper, this looks to be a good tool to support the extensions. Also, SOLOR is open source, if there is a community behind SOLOR with the same ICD-SNOMED problem, then maybe all the extension work won’t have to fall on the OHDSI community.

you’ll miss significant number of records (ultimately patients) in the cohort and analysis if standard concepts (SNOMED) are the starting point to prepare code lists

We also have concerns about using SNOMED. We have not started using the OMOP CDM, so we have not done an assessment on the ICD to SNOMED terminology mismatch and subsequent patient additions to cohorts without the specified ICD codes. But the reality is we live in the American extension of the ICD world.

Have you try to run it? Do you know how to use the tool?

I haven’t tried SOLOR or Protege.

[MPhilofsky (Melanie Philofsky)]

One more thing

There’s still the problem of concepts, concept relationships, hierarchies becoming deprecated.

@Christian_Reich

The problem is that we control neither the content of these Source Concepts, nor their life cycle.

We could use the extensions to keep the deprecated codes and relationships alive.

[abedtash_hamed (Hamed Abedtash)]

Can you show us those examples? We will need them to make the case for adding SNOMED Extension concepts. It’s a departure from our strong principle to not create (non-administrative) Concepts.

Sure, I’ll share the results with full comparisons.

[Dymshyts (Dmytry Dymshyts)]

We could use the extensions to keep the deprecated codes and relationships alive.

there is another proposal

Proposal to keep outdated standard concepts active and standard

We reviewed the logic of standards concepts deprecation. So far when the source removes the concept from list of the active concepts, we deprecate it, and it doesn't matter what is the reason of deprecation: incorrect concept - duplicate, having wrong description (e.g. wrong drug dosage), some weird classification branch, etc. correct, but outdated for the other reasons for example CPT4 in the last release makes "76645 | Ultrasound of breasts" concept outdated (maybe it's not a billing code anym…

in a few words: not to deprecate the concepts that become obsolete according to the source. Use deprecation if the concept was added mistakenly.

[Christian_Reich]

@Dymshyts:

For the RxNorm drugs, since we cannot have duplicates there, should we add that we will run the MapDrugVocab script, mapping deprecated to new ones (even if to higher granularity concept)?

[Dymshyts (Dmytry Dymshyts)]

Wrong panel.
but anyway:) it’s a good idea. let me analyse the deprecated RxNorm concepts