Hi, @pandamiao,
You can think of the ‘restrict initial events’ to an Inclusion rule that is applied before the other inclusion rules, for purposes of reducing your base population to a smaller subset, before applying inclusion rules.
You could take the criteria from ‘Restrict initial events’ and put it into a first inclusion rule, and get the same final cohort. This is the difference:
Cohort Entry Event: Diagnosed with Hypertension; 1 million people
Inclusion Rule 1: Type 2 Diabetes any time prior: 425,000 poepole
Inclusion Rule 2: No Kidney Disease: 920,000 people
Final Cohort: 350,000 People (those with T2DM and No kidney Disease)
This would show the ‘inclusion impact report’ as you started with 1 million people and then show how 425,000 had the T2DM, and 920,000 people. But what if your base population really should be those with Hypertension and T2DM prior? You’d want to restrict those initial events:
Cohort Entry Event: Diagnosed with Hypertension; 1 million people
Restrict Initial Events: Type 2 Diabetes any time prior: 425,000 poepole
Inclusion Rule 1: No Kidney Disease: 350,000 people
Final Cohort: 350,000 People (those with T2DM and No kidney Disease)
The ‘inclusion impact report’ will now start with 425,000 poeple (those diagnosed with hypertension with prior T2DM). And the inclusion rule will be applied to those events that came from cohort entry events after applying the initial event restriction.
As you can see, the logic of the cohort definition resulted in the same people/episodes, but the ability to restrict those initial events before applying inclusion criteria lets you have a better picture about how your inclusion criteria are being applied to a more specific population.
-Chris
