REVMA Prague

[SimonKonar]

✅ Bridgehead deployed
Script for data harmonization (https://github.com/BBMRI-cz/fhir-module) and uploading is updated, curretly is getting review from @RadovanTomik, so the changes can be merged, and fhir-module can be deployed at the REVMA

[SimonKonar]

Changes have been merged, deployment to at REVMA is needed

[SimonKonar]

While preparing data for REVMA, few questions have risen:
There is a problem with data quality regarding the REVMA.

• They insist on having the collections inside Directory based on the condition. However, inside their system, the data is not in good quality. They have internal names for diagnoses, and respective MKN-10 code (not even ICD-10). However, multiple different diagnoses each have same MKN-10 code as seen in the image below.
  

• There is code M01.20 which exists in MKN-10 code, but does not exist in ICD-10. However, in ICD-10, there is code A69.2, which holds the same description for the diagnoses. But ms. Stiburkova at REVMA said they want to use the MKN-10 code.

• Most common samples inside the REVMA export have double diagnosis (as seen in the image above). However, these code mutually exclude themselves (M0580 - Other rheumatoid arthritis with rheumatoid factor of unspecified site ; M0600 - Rheumatoid arthritis without rheumatoid factor, unspecified site)

So the question arrives, How much should we meddle with affairs of their quality of data ? Are they supposed to improve the quality of data? Would htey even be willing to? Is it okay to upload these "unclean" data inside the Bridgehead & Directory ?

[SimonKonar]

After call with IT specialist from REVMA, The following was agreed upon:

• The codes can be changed - there can be a new column called ICD-10 with respective icd-10 codes, which will ease up the data uploading to the BH.
• They will talk about the issues with the same codes, for different diagnosis (in REVMA, they differentiate the samples based on the TDI_NAME , they dont use the codes as an identification of various diagnoses)