You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We'd just need to detail the desired functionality. Your example where you'd want to exit the cohort based on criteria is what censoring events does. Exiting on end of observation period + censoring events = exit cohort at a censor event. The reason why censoring events was provided separately from the other 'exit strategies' was that you may want to do end ob observation + censoring, or fixed duration + censoring, or continuous drug exposure + censoring.
I think that one new type of exit would be an exit based on era-fiying any set of domain criteria (the continuous drug exposure exit strategy assumes only drug exposure domain , with drug-exposure specific parameters such as persistence window and gap days). So, this is something I think has a clear objective and is something that would fit under a new 'exit strategy'.
For your use case, we'd just need to understand 'is it a new exit strategy we're thinning of?' and 'how should this exit strategy work?'
As @pbr6cornell outlined here: https://forums.ohdsi.org/t/limitations-of-atlas-cohort-definitions-tool/15752/2, we often need to have the cohort exit based on something other than end of obs period, continuous use of a drug (based on era), or a fixed duration.
I'm wondering if we can make this section more flexible to allow for a full set of criteria events and attributes, much like we have in the censoring.
The text was updated successfully, but these errors were encountered: