st: Analysis of SIR or HR with timevarying covariates and late entry big data

[roland andersson <rolandersson@gmail.com>]

I want to compare the incidence rate of a disease in 1st and 2nd
generation immigrants with that in the native population. I have
individual population data on total 11,000,000 individuals gathered
during a 20 year period. The data is year of birth, country of birth
of the individual and his/her parents, sex, date of immigration and
emigration, death and operation for the disease. You can only have the
disease once. The disease varies strongly with age and has been
decreasing over the study period.

So this is time to event data with late entry. As the disease is
strongly related to age I think to use age as analysis time. If I use
Cox-regression I do not know how to adjust for the decreasing
incidence over time.

I think I can do stsplit on year of birth and analysis time, but I am
afraid i will run into memory problem.

May I use stir and adjust for year of birth?

Help!
Roland Andersson
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/