OHDSI Home | Forums | Wiki | Github

Creating a cancer cohort for reasearch purporse

Researchers
,
#1

Hi, I currently have access to our institution’s EHR records which is OMOP derived. I aim to create a cancer cohort out of this to make it viable for research purposes. Is there any documentation or code in SQL that I can look up in order to identify cancer-diagnosed patients, demographics, survey related data?

I can identify that these data variables exist but have been having a hard time mapping them together in order to create a clean table with all the pateint_id records. i would really appreciate any guidance I can get as I am just getting started.

Thank you in advance!

#2

Hi @srchintala ! It seems that your first step would be to create a cohort definition. How to create a cohort definition is described in the Book of OHDSI. If you don’t have ATLAS installed, you can nowadays also use the Capr R package.

Once you have defined and generated your cohort, there are many OHDSI tools you can apply to it. A second step might be to run CohortDiagnostics on your cohort.

Note that in OHDSI we typically do not ‘download a cohort’. We keep a list of patient IDs with start and end dates in a cohort table on the server where the full CDM data also resides, and with that we can perform many operations.

#3

Thank you @schuemie ! I will make sure to look into this!

t