Spike: Combined 2016/19 ICD 10 edition

[Jongmassey]

No description provided.

[Jongmassey]

combined.zip

Imported in a Django shell thusly:

from datetime import datetime
valid_from = datetime.fromisoformat("2026-01-01")
from coding_systems.icd10.import_data import import_data
filename = "/home/jongmassey/opencodelists-data/coding_systems/icd10/combined.zip"
import_data(filename,"test",valid_from, check_compatibility=False)

[Jongmassey]

Slack thread with screenshots

[Jongmassey]

General thinking with the new model for this is:

• A "release" of ICD 10 is a release of our conception of the merging of the editions discussed. If one of the underlying editions is updated, we'd need to cut a new release
• The association between a concept and an edition is stored such that "what edition(s) is this concept present in?" can be presented to the user
• Where a concept's description is different between editions, this can be used as a synonym
• This includes modified codes and details of the modifiers (for 2016 anyway, 2019 needs more work as they're not included in the flat file and the about of ClaML fandango required to extract them from the XML was more than what I had time for in this spike). We could either just let the existing hierarchical presentation of the codes tell the story or maybe display them differently in a more meaningful manner
• Rubrics are included, they could be displayed in "more info" dialogue as often have important info, and could be used in term searching (i.e. users' search terms might be in the explanatory notes for a concept, but not its main description)

There's a bunch more polish to be done and thought to be given about what's the best way of solving this problem (not to mention the question of what to do about pre-existing codelists) but hopefully this starts some conversations.