Closed MaximMoinat closed 3 years ago
MaximMoinat
commented
3 years ago Decided last week to map these tables. It contains the same and more records than the original gp_clinical and gp_prescription tables (the new is a superset of the old). The logic should be that if the GP Covid tables are present, map them. If not, map the original gp_clincial and gp_prescription tables (or make a toggle).
More information from Spiros is still needed.
MaximMoinat
commented
3 years ago Estimations made, to be confirmed whether this can be added to the transformations
spiros
commented
3 years ago yep, just to confirm, I am working on producing some frequency counts for the local EMIS and TPP codes and a proposal of what to map and what to drop
On Mon, Feb 8, 2021 at 10:36 AM Maxim Moinat notifications@github.com wrote:
Decided last week to map these tables. It contains the same and more records than the original gp_clinical and gp_prescription tables. The logic should be that if the GP Covid tables are present, map them. If not, map the original gp_clincial and gp_prescription tables (or make a toggle).
More information from Spiros is still needed.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/EHDEN/ETL-UK-Biobank/issues/216#issuecomment-775009037, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAFZZF2ZYOFAN4SYJJBHK3S56WAZANCNFSM4XESQNSA .
spiros
commented
3 years ago I've had a closer look at this and my opinion is to ignore EMIS/TPP local codes entirely with some exceptions (see below).
These codes are almost exclusively administrative codes for example, requesting a test or referring to a particular clinical in hospital or giving smoking advice. For example, in EMIS, there are a handful of codes which could have some data attached to them but they are so rarely (i.e. 44K times out of 270M records -> ~0.01%) used that it's not definitely not worth the effort of mapping them.
The only codes I think we should map are the ones related to COVID-19 (where possible) - see list below:
| Local code | Number of records | Local code definition |
|---|---|---|
| EMISNQTE31 | 700 | Tested for 2019-nCoV (novel coronavirus) infection |
| EMISNQSU106 | 321 | Suspected 2019-nCoV (novel coronavirus) infection |
| EMISNQEX59 | 230 | Excluded 2019-nCoV (novel coronavirus) infection |
| EMISNQEM35 | 146 | EMIS COVID-19 care pathway template entry |
| EMISNQCO303 | 74 | Confirmed 2019-nCoV (novel coronavirus) infection |
| EMISNQEX58 | 3 | Exposure to 2019-nCoV (novel coronavirus) infection |
| Y228a | 12583 | High risk category for developing complication from coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 infection (finding) |
| Y22b6 | 10163 | Provision of advice, assessment or treatment limited due to coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 pandemic (situation) |
| Y228b | 4913 | Moderate risk category for developing complication from coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 infection (finding) |
| Y212d | 2120 | Advice given about severe acute respiratory syndrome coronavirus 2 infection (situation) |
| Y212c | 1655 | Advice given about severe acute respiratory syndrome coronavirus 2 by telephone (situation) |
| Y211b | 1533 | Telephone consultation for suspected severe acute respiratory syndrome coronavirus 2 (procedure) |
| Y212e | 1065 | Educated about severe acute respiratory syndrome coronavirus 2 infection (situation) |
| Y20cf | 930 | Suspected coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 (situation) |
| Y228c | 774 | Low risk category for developing complication from coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 infection (finding) |
| Y20d2 | 160 | Severe acute respiratory syndrome coronavirus 2 not detected (finding) |
| Y20fa | 113 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 (disorder) |
| Y20ce | 71 | Exposure to severe acute respiratory syndrome coronavirus 2 infection (event) |
| Y20d1 | 63 | Severe acute respiratory syndrome coronavirus 2 detected (finding) |
| Y228d | 61 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 confirmed by laboratory test (situation) |
| Y22a2 | 37 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 excluded by laboratory test (situation) |
| Y210a | 21 | Pneumonia caused by severe acute respiratory syndrome coronavirus 2 (disorder) |
| Y228e | 20 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 confirmed using clinical diagnostic criteria (situation) |
| Y229d | 19 | Swab for severe acute respiratory syndrome coronavirus 2 taken by healthcare professional (situation) |
| Y22aa | 19 | Signposting to CHMS (COVID-19 Home Management Service) (procedure) |
| Y229e | 17 | Close exposure to severe acute respiratory syndrome coronavirus 2 infection (event) |
| Y22a1 | 13 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 excluded (situation) |
| Y23f4 | 12 | SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) detection result negative |
| Y210e | 12 | Detection of severe acute respiratory syndrome coronavirus 2 using polymerase chain reaction technique (procedure) |
| Y22a4 | 9 | Assessment using coronavirus disease 19 severity scale (procedure) |
| Y210b | 8 | Infection of upper respiratory tract caused by severe acute respiratory syndrome coronavirus 2 (disorder) |
| Y22a3 | 7 | Coronavirus disease 19 caused by severe acute respiratory syndrome coronavirus 2 excluded using clinical diagnostic criteria (situation) |
| Y22a0 | 6 | Taking of swab for severe acute respiratory syndrome coronavirus 2 (procedure) |
| Y0fc2 | 5 | Client provided with any other recovery support elements |
| Y23f7 | 3 | SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) detection result positive |
| Y20fd | 3 | High priority for severe acute respiratory syndrome coronavirus 2 vaccination (finding) |
| Y229a | 3 | Self-taking of swab for severe acute respiratory syndrome coronavirus 2 offered (situation) |
| Y229b | 2 | Swab for severe acute respiratory syndrome coronavirus 2 taken by subject (situation) |
| Y0ad9 | 2 | Currently Residing: Specialist rehabilitation/recovery |
| Y22a5 | 1 | Coronavirus disease 19 severity scale (assessment scale) |
| Y212a | 1 | Severe acute respiratory syndrome coronavirus 2 vaccination not indicated (situation) |
| Y213a | 1 | Antigen of 2019-nCoV (novel coronavirus) |
| Y211a | 1 | Measurement of severe acute respiratory syndrome coronavirus 2 antigen (procedure) |
| Y210c | 1 | Severe acute respiratory syndrome coronavirus 2 serology (observable entity) |
| Y229c | 1 | Coronavirus disease 19 severity score (observable entity) |
MaximMoinat
commented
3 years ago Thanks @spiros. I do have a question. For both EMIS and TPP there is a mix of local and generic codes in the code column. e.g. in TPP there are READ3 codes and the local Yxxxx codes. We will of course map the READ codes as before.
For EMIS, there are codes starting with EMIS and codes that are a string of numbers (e.g. 1091811000000102). Do the latter have some direct mapping to READ or SNOMED we can use?
I have put the mapping of the local codes in a separate issue.
Four new tables, very similar to the already mapped gp_clinical and gp_prescription.