Cohort Characterization design and resuts extensions (incidence vs prevalence, source codes, post/prior/around index date analysis)

[alex-odysseus]

Business Requirements

• in the result tables when appropriate it should be possible to split the whole analysis time frame by years and display the prevalence counts instead of incidence for the whole time frame
• when the source codes are essential in an analysis there is no way to see their counts fine grained for each source code in particular
• there should be a way to perform the "post index date analysis" -- a time window after the index date -- events of interest
• there should be a way to perform the "time frame around index date analysis" -- a time window around the index date -- events of interest
• there should be a way to perform the "prior index date analysis" -- events of interest (grouped in categories) --- how many of those were found quantifying the categories and not only generating the proportion of patients

Design Ideas and Functional Requirements

To be defined

Open Questions

• how to collapse related result entries defining a basis, for instance, there is no interest in seeing all variations of diabetes Concepts but to have them in one row mentioning that "diabetes" should be the basis

@mrechkem @Sathyaraos