Sex assigned at birth separate from Gender

[alix-lacoste]

Hello,

I work in a genetics company where "sex assigned at birth" is an important attribute of a patient, as it is related to a person's DNA. I believe OMOP should distinguish between "gender" and the concept "sex assigned at birth".

"Gender" is socially constructed. A person can have a gender = male and sex_assigned_at_birth = female. Both are important attributes to take into consideration for healthcare.

Gender is defined by the World Health Organization as "the characteristics of women, men, girls and boys that are socially constructed. This includes norms, behaviours and roles associated with being a woman, man, girl or boy, as well as relationships with each other. As a social construct, gender varies from society to society and can change over time." There are other genders, such as "non-binary" not named in this definition. However, the point is that gender can change over time, while "sex assigned at birth", which is usually, but not always, equivalent to the types of sex chromosomes a person has, does not change over time.

Not that I prefer not use the term "biological sex" as it is misleading for individuals that are intersex or have had surgery that modifies their sexual organs. "Biological sex" does not have a strict definition, unlike "sex assigned at birth".

Conflating gender and sex assigned at birth could significantly hurt research and patient care.

[ablack3]

I think that "gender_concept_id" is defined as "biological sex at birth" which is meant to be equivalent to "sex assigned at birth"
https://ohdsi.github.io/CommonDataModel/cdm54.html#PERSON

The wording is unfortunate but changing a column name in the CDM specification is really hard to do because it would mean that every OMOP CDM would need to be changed. It could only happen on a major release which is possibly a long time from now. Updating the documentation on the other hand is relatively painless. Would a documentation update sufficiently clarify things?

[alix-lacoste]

Thank you for the answer. I had not seen the definition for the "gender_concept_id before so thanks for linking to it."

I do not see a need to change the definition at this time. However, the name ("gender") is very misleading. So I think the column name would need to change in a future release. It would also be appropriate to use an optional "gender" column, to reflect how individuals self-identify.

[clairblacketer]

@alix-lacoste this has been an ongoing discussion in the community for a few years. We understand that the wording is not appropriate to convey the meaning behind it but, as @ablack3 details, every single one of our packages and pieces of software refers to this column. It is on a future roadmap to address this issue but it will be an incredible lift that has to be done by a volunteer crowd. We have done our best in the meantime to update the documentation to reflect our intentions and be as unambiguous as possible.

[alix-lacoste]

@clairblacketer thank you for the additional changes. I look forward to updates in the future.