Clustering IDs

[RebN]

Hi

I am working on analysis for a cohort of patients and have a reference cohort as a comparison. 10 references were matched on various characteristics (age, race, etc). I am trying to find a simple way to "pick" which reference members are in a given analysis based on whether their corresponding case is in the analysis. There is a clustrid variable which is a unique identifier for a group of 1 patient and 10 referents.

For example, if I am restricting the analysis to only 75% of the patients, I only want to select those referents whose patient is part of the 75%.

I have been doing this a long way by creating a series of data steps with variables indicating whether the case is in the analysis, but this is very time-consuming.  Is there a simple way tell SAS how to select the referents?

Thank you in advance

[ChrisNZ]

I am afraid you got no reply because the question is not the clearest.