gyeol99
August 23, 2019, 1:03am
1
Hi @Christian_Reich ,
I hope I’ve mentioned the right person.
The current code for building the drug_era table only references RxNorm.
WHERE c.vocabulary_id = 8 ---8 selects RxNorm from the vocabulary_id
This results in RxNorm Extension-unique ingredients to be left out in the final drug_era table. Shouldn’t we be adding another vocabulary_id to include RxNorm Extension-unique ingredients as well?
WHERE c.vocabulary_id in (8, 00) ---8 selects RxNorm and 00 selects RxNorm Extension from the vocabulary_id
Thanks in advance!
Song
@gyeol99 :
This is very old code. From years before we had RxNorm Extension. Vocabulary_ids were still numeric. You need to move on from Version 4.
If you have to use V4 for some reason, include vocabulary_id 82: WHERE c.vocabulary_id in (8, 82).
gyeol99
August 23, 2019, 3:46am
3
@Christian_Reich Thanks for your quick response.
Just one more question: Have we as a community decided on what to do with the stockpiling issue? Or is it still up to data owners?
gyeol99
August 23, 2019, 4:58am
4
@Vojtech_Huser :
The settings for DRUG_ERA are set. 30 days persistence and no stockpiling. If you want to replace that with ingredient-specific settings - please propose to the CDM WG. Of course you would have to create or adopt such a heuristic.
The persistence window in ATLAS and its ability to consolidate cohort records has nothing to do with the ERAs. It’s up to you what you define in your cohort.
Never mind. Found the answer I was looking for. No stockpiling!
