search

OHDSI / Vocabulary-v5.0

Build process for the OHDSI Standardized Vocabularies. Currently not available as independent release.
The Unlicense
206 stars 73 forks source link

Proposal to create cancer modifiers for 'hormone sensitive malignancy', 'hormone resistant malignancy' terms. #995

Open dimshitc opened 2 months ago

dimshitc commented 2 months ago

ICD10CM concepts Z19.1|Hormone sensitive malignancy status, Z19.2|Hormone resistant malignancy status don't have a proper target concept in OMOP, they are just maped 1340204|History of event + 443392|Malignant neoplastic disease.

Hormone sensitivity is clearly a cancer modifier, so it would be good to create such concepts in Cancer modifier vocabulary and map these source concepts respectively. @vladkorsik

vladkorsik commented 2 months ago

Hi @dimshitc ,

The concept of hormone sensitivity is a complex construct that may reflect either the nature of the tumor before the start of treatment (predicting resistance to endocrine therapy) or the fact of established resistance. From coding guides, I assume the latter case is implied, but real-life application and practice are far less straightforward.

Simply adding a new concept may not suffice, as ideally it requires a specific class of concepts. Making changes to existing concept classes could provoke additional discussions and necessitate formalization of approaches to the class. I particularly highlight this in light of the ongoing discussions within the working group, both regarding Staging/Grading concept_class_id (new concept classifiers vs. division of classes) and the use of the Metastasis class (issues with modeling and default ETL storage).

Furthermore, even if the proposed solution is supported, it's essential to note that the "Cancer modifier" Vocabulry is currently unsupported by the Vocabulry Team. The whole team is ready to help accept and improve community contributions if they happen.

Let me tag @Alexdavv @TinyRickC137 @cgreich @golozara to make the discussion visible.

dimshitc commented 2 months ago

thanks, @vladkorsik
Taking into account your thoughts, how would you recommend to map the aforementioned codes?

Z19.1|Hormone sensitive malignancy status, Z19.2|Hormone resistant malignancy status

cgreich commented 3 weeks ago

This is a good one. I don't have a good idea. Because the definition of that thing is wobbly, as @vladkorsik said. One solution would be to ignore it. If you need a phenotype with that thing just take the malignancy, find the hormone treatment and the discontinuation of the hormone treatment. That is actually the reason the name will be assigned: Hormones stopped working.