Loris: [SQL] `candidate.sex` column type does not match `tarchive.PatientSex` column

[johnsaigle]

The candidate.sex column is an enum with Male and Female as possible values.

The tarchive.PatientSex column is a varchar(255). Raisinbread contains values M, F, and O.

The second case implies an "Other" value for sex which may have impact on study ontologies.

These columns should be made consistent.

[maltheism]

Just my thoughts but I think there is actual use for a third option in research studies. There are people that are intersex and hermaphrodites while rare as well would not fit in the M or F category.

[johnsaigle]

For sure, and it also covers cases when the value is unknown or could not have been collected for whatever reason. Not sure of the needs of projects though.

[maltheism]

@ridz1208 @driusan thoughts?

[johnsaigle]

I put it in the LORIS meeting agenda so we can discuss it as a group. I feel like we'll need the input of people closer to the research process. It's more of a research-based decision than a software architecture decision.

[johnsaigle]

The result of the discussion is to add an Other option to the enum in the Candidate table.