Update of MedDRA

Prerequisites

• Schema DevV5 with copies of tables concept, concept_relationship and concept_synonym from ProdV5, fully indexed.
• SNOMED must be loaded first.
• Working directory is MedDRA.

Sequence of actions

Source filling

1. Run create_source_tables.sql
2. Download the current Meddra from https://www.meddra.org/software-packages (english) and "SNOMED CT – MedDRA Mapping Release Package"
3. From MedDRA_xy_z_English.zip extract files from the folder "MedAscii":

• hlgt_hlt.asc
• hlt.asc
• hlt_pt.asc
• hlgt.asc
• llt.asc
• mdhier.asc
• pt.asc
• soc.asc
• soc_hlgt.asc

4. From "SNOMED CT - MedDRA Mapping Release Package DD MONTH YYYY.zip" extract *.xlsx file and rename to meddra_mappings.xlsx
5. Run in devv5 (with fresh vocabulary date and version):

SELECT sources.load_input_tables('MedDRA',TO_DATE('20160901','YYYYMMDD'),'MedDRA version 19.1')

Filling stage and basic tables

6. Perform manual work described in the readme.md file in the 'manual_work' folder.
7. Run load_stage.sql
8. Run check_stage_tables function (should retrieve NULL):

SELECT * FROM qa_tests.check_stage_tables ();

9. Run generic_update:

DO $_$
BEGIN
	PERFORM devv5.GenericUpdate();
END $_$;

10. Run basic tables check (should retrieve NULL):

SELECT * FROM qa_tests.get_checks();

11. Clean cash:

SELECT * FROM qa_tests.purge_cache();

12. Run scripts to get summary, and interpret the results:

SELECT DISTINCT * FROM qa_tests.get_summary('concept');

SELECT DISTINCT * FROM qa_tests.get_summary('concept_relationship');

13. Run scripts to collect statistics, and interpret the results:

SELECT DISTINCT * FROM qa_tests.get_domain_changes();

SELECT DISTINCT * FROM qa_tests.get_newly_concepts();

SELECT DISTINCT * FROM qa_tests.get_standard_concept_changes();

SELECT DISTINCT * FROM qa_tests.get_newly_concepts_standard_concept_status();

SELECT DISTINCT * FROM qa_tests.get_changes_concept_mapping();

14. Run manual_checks_after_generic.sql, and interpret the results.