Creating a Cohort's Complement Set

[zhehaowUNC]

is there any way to create the complement of an existing cohort or at least get the id's of the patients that are not in that cohort by calling webapi? if not, what would be a possible approach to this?

[chrisknoll]

No we try to avoid revealing personally identifying information to clients. I would resort to querying direclty into your cdm/results scheama where you look at the results.cohort table for the people in the cohort an then your cdm.person table to identify who exists in your data.

[zhehaowUNC]

this is my solution locally when I do have direct access to the postgresql database. However I'm trying to see if I can actualize this functionality in Atlas online connected to real patient databases, which I believe I don't have access to? I'm trying to compare the patients in one cohort vs those in its complement cohort and extract a subset of patients from the complement cohort that are similar in certain ways to those in the original cohort. Then my goal is to provide stats on the three sets similar to how cohort characterization provides insight on a single cohort. I'm not attemping to directly reveal individual patient info to viewers.