I have a number of questions regarding background incidence rates. For context, I am working with a user-defined grid with a number of census block groups (CBGs) within a 4-county area.
Is it recommended that I multiply the county-specific background incidence rates within BenMAP by the ratio of my census block group population to the county population? This will create incidence rates in the 1E-06 range, in my case.
Is there a limit to the number of significant digits you can input as background incidence values? When I loaded my user-defined incidence rates based on incidence*(CBG pop/County pop), all of my resulting point estimates were zero.
If I choose NOT to apportion my background incidence rates by my user-defined grid population, how does BenMAP apportion the county-specific incidence rates to each of my grid domains? I ask because I tried this and got nonzero point estimates, tried to back-calculate the background incidence rate, and found it relatively unclear what BenMAP had done to get these incidence rates (most of them are somewhere between 30-70% of the county-specific incidence rates).
Are the units of the incidence rate datasets within BenMAP #cases/people? Is this based on #cases/100,000 people?