Define a simple ICD case cohort in Atlas
This example shows how to build a simple case cohort using the Finnish-specific H81 ICD10fi codes.
We will first use Cohort Entry Events to include everyone in the cohort and then use Inclusion Criteria to exclude people who are not in our Concept Set. In other words, we will use the Inclusion Criteria to only include patents with the Finnish-specific H81 ICD10fi codes into the case cohort.
Make sure to read the Using standard and non-standard codes in Atlas Cohort Definitions section from Standard and non-standard codes to get more information on how to define cohorts with standard or non-standard codes. During this example. keep in mind that the H81 ICD10fi code is a non-standard code.
Step 1: Create an empty Cohort
Select the Cohort Definitions option on the left and click the New Cohort button on the right.
Name the Cohort using a descriptive name followed by your initials (e.g. Test_Cohort_Cases_MN).
Step 2: Define Cohort Entry Events
Include everyone and every event in the cohort by clicking the Add Initial Event button in the Cohort Entry Events section and by selecting the Add Visit Occurrence option from the list:
Select a visit occurrence of "Any Visit" to include all visits in the data.
Step 3: Define Cohort Inclusion Criteria
We will now use Inclusion Criteria to only include patients with the H81 ICD10fi codes into the cohort.
Click New inclusion criteria button in the Inclusion Criteria section and enter the criteria name (e.g. TestPhenotype) and an inclusion rule description:
Click Add criteria to group button and select Add Condition Occurrence:
Note the "at least 1” selection for the cases cohort.
Click Add attribute button and, because we are dealing with a non-standard code, select Add Condition Source Concept:
The Concept Set name will be displayed on the right of the "Condition Souce Concept is" text:
Step 4: Generate the case cohort
Open the Generation tab and generate the cohort by clicking the Generate button next to the FinnGen data release (FinnGen CDM R11 in the screenshot below). The cohort generation will take a few minutes and the speed is not dependent on the size of the sandbox IVM.
The case cohort is now ready. This cohort can be explored within Atlas and used in Cohort Operations and Trajectory Visualization tool. The cohort may also be exported from Atlas for further analyses like those described in the General Workflows for the most common analyses.
Next steps
Last updated
