st: how to make sttocc suitable for case-cohort?

["Simons Colinda (EPID)" <colinda.simons@maastrichtuniversity.nl>]

Dear Statalist,

I have been using the sttocc command to generate a nested case-control study from a cohort study. 
However, I do not have full cohort data in my dataset, only case-cohort data.
This means that I have my subcohort, which is representative of the whole cohort because it was selected randomly at baseline, and I have incident cases, which were enumerated for the entire cohort.
I now want the sttocc command to sample controls from the subcohort only because the subcohort is representative of the whole cohort. 
Some cases are in the subcohort. These can be selected as a control if their not yet a case when another case becomes a case. This is how it should be, I think. 

There does not seem to be an option in sttocc to specificy that it can only select controls from a subgroup (the subcohort) instead of the whole dataset (which also contains cases from outside the subcohort).

Does anyone have any ideas on how to sample controls using sttocc from a subset of the data?

I realize that sampling controls from a random sample of the cohort might not be ideal, but this is all I can do.

Any thoughts would be greatly appreciated!

Thanks,
Colinda Simons, PhD

Maastricht University
Dept. of Epidemiology
The Netherlands

*
*   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/