doubly censored multilevel survival anlysis - is it possible?

[janne]

Hi

I have a cohort of diabetes-patients from 2006 (but no data on how long they have lived with the disease prior to 2006). We want to follow upp these patients with regards to occurrence of first time cardiovascular diseases. We want to look at both differences between areas and differences between income-groups.

Is there such thing as a doubly-censored multilevel survival analysis? Or is there any other way to solve the problem?

Kind regards

Janne