Isaric followup #175
Isaric followup #175
Conversation
isaric/parsers/isaric-tier1.toml
Outdated
| # required fields - "visit_id", "subject_id", "country_iso3", "start_date", "outcome", "date_outcome" | ||
|
|
||
| [visit] | ||
| outcome = "recovered" |
There was a problem hiding this comment.
@sadiekelly how are long COVID patients (if present in these studies) assigned outcome?
There was a problem hiding this comment.
@abhidg outcome can be discharged for those patients who have a discharge date given in flw_first_disch. Other patients were not admitted. Do we want to consider the survey completion as a 'visit', and find a relevant term that can be omop-ed for these (follow up visit completed - not sure what is available, I can look)? date of visit is then the date of survey completion, but that would likely be all that could be populated
|
Summary of the missing optional fields for the isairc-followup-fr parser:
|
There was a problem hiding this comment.
hi @pipliggins here's my review of the french follow up parser. My main comments are around integration of this follow up data with the acute data collected in the main CORE database, which can be linked by subjid. So thinking whether some data collected here at follow up can be integrated with and used to upd
French inserm follow up parser review 2023-10-10.xlsx
ate the subject table e.g. where the patient has died. It might be easier to discuss more fully, let me know.
There was a problem hiding this comment.
Thanks @sadiekelly - I've made the edits required. Re linking the data; this is something we'll have to do on the database side, after the data is initially parsed, rather that within the parsers themselves. So I've created a 'subject' and 'visit' table for the follow-up data, but when we come to merge the data based on the subject ID's, we can write a hierarchy rule that allows the followup data to overwrite the acute dataset where necessary. Hope that makes sense!
There was a problem hiding this comment.
hi @pipliggins here's my review of the tier 1 freestanding parser. This and the brazil follow up are almost identical and share very similar points for review/discussion, so I'd suggest resolving them and updating both at the same time.
Tier 1 freestanding parser review 2023-10-10.xlsx
There was a problem hiding this comment.
@pipliggins I've realised the tier1 parser should also fit the initial/ongoing surveys - I'll review this parser against those now, apologies for missing this!
There was a problem hiding this comment.
@sadiekelly no worries - I've just done the Brazil parser, so if you're happy with those changes I'll apply the same logic here where required
There was a problem hiding this comment.
@pipliggins here's a (very short) review when comparing Tier1 parser to Initial follow up. I haven't duplicated anything that is already listed in the Tier 1 parser review, so just a couple of points here
Initial follow up parser review 2023-10-10.xlsx
There was a problem hiding this comment.
@pipliggins and here's a few comments regarding the ongoing follow up - thanks!
Ongoing follow up parser review 2023-10-12.xlsx
Contains most of the required changes + notes for possible further edits Untested
|
Summary of the missing optional fields for the isairc-followup-fr parser:
|
2 parsers included, as the French followup data follows a very different naming convention to the others.
The visit table should be removed/commented out for the Initial*/Ongoing* data tables using the isaric-tier1 parser, as they contain follow-up data only.