OHDSI / PhenotypeLibrary

A repository to store, organize and maintain the content of the OHDSI Phenotype library. OHDSI Forum post https://forums.ohdsi.org/t/ohdsi-phenotype-library-announcements/16910
https://ohdsi.github.io/PhenotypeLibrary/
Other
41 stars 16 forks source link

Cohort exit criteria #89

Open siir-su opened 4 months ago

siir-su commented 4 months ago

I have observed that most definitions have the cohort exit crtierion of "fixed duration relative to initial event" instead of "end of continous observation", which is the default on Atlas UI. I also did not seem to find any indication of this criterion being related to chronic vs acute conditions. Is there a certain reason why we prefer assigining a fixed duration relative to index event in cohort definitions? This subsequently results in more 'records' overall for the cohort of interest, as patients can qualify later points in time without being collapsed under their entire observation period. On the other hand, it is generally counterintiutive to define a cohort exit - since we would generally want to follow patient-level outcomes until the end of continuous observation or censoring.

It would be great if you could inform on whether this is a best practice definition to define cohort exit at the end of cohort entry date.