Time distribution diagnostics

[gowthamrao]

I think it would be useful to describe the subject and event counts by time distributions binned in 30 days e.g.

0 days, < 30 days, < 60 days, < 90 days, < 120 days.. >= 360 days

• Time in days between a person’s observation start date in the data source and their cohort start date;
• Time in days between a person’s cohort start date and cohort end date;
• Time in days between a person’s cohort start date and the last date of continuous observation in the data source

This would help answer questions like

• how many subjects/events had 0 days
• how many had 30 days