search

opensafely / ics-research

This is the code and configuration for our paper, Inhaled corticosteroid use and risk COVID-19 related death among 966,461 patients with COPD or asthma
https://opensafely.org
0 stars 1 forks source link

*DISEASE* Other conditions requiring oral steroids #48

Closed HenryDrysdale closed 4 years ago

HenryDrysdale commented 4 years ago

For asthma exacerbations, we need a list of conditions to check for on the day of oral steroid prescribing, such as RA, IBD and others

ianjdouglas commented 4 years ago

List used by Bloom et al https://www.atsjournals.org/doi/pdf/10.1164/rccm.201808-1516OC

steroid_disease.xlsx

CarolineMorton commented 4 years ago

Hi

Thanks for this. From your document, we have a number of these covariates already as part of other lists already on https://codelists.opensafely.org/

We do not have a current code list for TA / GCA or other steroid disease. We opted not to have automimmune in the risk factors study due to the issues of mapping a lot of eponymous disease (With v low incidence) into CTV3. We might want to do the same here pramatically.

ianjdouglas commented 4 years ago

Thanks @CarolineMorton - sounds a good pragmatic approach and likely to capture most. Is it relatively painfree to extract the relevant codes from bigger lists? e.g. ILD from other chronic resp disease etc?

CarolineMorton commented 4 years ago

Yes, you will need to download them from the codelists on opensafely linked above, and then manually sort. @annaschultze did this a lot to remove the heart failure codes from chronic heart disease and copd from chronic resp disease and im sure would be able to help if needed. I think pulling into excel and then running filters with contain statements, should pick up most things.

CarolineMorton commented 4 years ago

By the way, I should mentioned TA / GCA codes if they are in Read 2 (and I htink they are) will need to be mapped over to CTV3 by adding in QOF and Snowmed code as per the previous procedure. If you let me have them as the first priority, I can get to Chris (it takes about 24 hours for mapping).

ianjdouglas commented 4 years ago

If we're just extracting them from the OpenSAFELY code lists, won't they already be mapped?

CarolineMorton commented 4 years ago

Yes the ones with codelist are already mapped but GCA and TA are not in any codelists and need their own new one.

ianjdouglas commented 4 years ago

@hjforbes - would you be able to have a look at this by any chance? In the spreadsheet there are only 3 codes for giant cell or temporal arteritis so hopefully not a major undertaking

hjforbes commented 4 years ago

Sure, I'll do it now

hjforbes commented 4 years ago

GCA/TA: I found a few extra Read codes. @CarolineMorton can you give the Read codes a quick glance (there's only 7 of them), then I think it's ready for mapping.

STEP 1: READ v2 codes - READ_v2_GCA_TA.xlsx STEP 2: no QOF codes. STEP 3: snomed codes - snomed-gca_ta.xlsx

CarolineMorton commented 4 years ago

All looks fine, I will send for mapping now.

HenryDrysdale commented 4 years ago

Hi all,

@CarolineMorton @ianjdouglas @hjforbes

I have created separate code lists for RA, IBD, MS, ILD and SLE by extracting from existing lists. I've uploaded all except ILD to codelists.opensafely.org. A few points to note:

ILD

RA

IBD

MS

SLE

ianjdouglas commented 4 years ago

Thanks v much @HenryDrysdale and @hjforbes. My view would be to mirror the list used in Bloom as it's referenceable. I can see there could be room for further work here but to be pragmatic for now I'd keep to these.

HenryDrysdale commented 4 years ago

Ok thanks @ianjdouglas. I've created the ILD as per Bloom, and uploaded to codelists.opensafely.org:

https://codelists.opensafely.org/codelist/opensafely/interstitial-lung-disease/

hjforbes commented 4 years ago

Final reviewed codes for GCA/TA - @CarolineMorton ready for sign off CTV3_GCA_Raw_HF_checked.xlsx

CarolineMorton commented 4 years ago

Yes very happy to sign this off!