Strata statement in surveyfreq for case-cohort data

[kpberger]

I have a case-cohort study that I need frequency cross-tabs for. I am running surveyfreq with weights because I would like to approximate the source population. Do I include an indicator for being sampled into the subcohort vs case group in the strata statement? When I do it lowers the p value around a hundred-fold.

[SAS_Rob]

The use of the STRATA statement would depend upon what your survey design is.    The STRATA statement is meant to identify strata from a complex sampling design and not to run some sort of a conditional model/analysis that is usually associated with a case/control study.

If you have a case control study based on a weighted random sample then you would want to include the indicator on the TABLES statement only and not the STRATA statement.

[kpberger]

As a simpler hypothetical, what if I have a stratified random sample, and the stratification variables are sex and age. In this study, we oversampled girls and younger ages so we would have enough N to look at age-related differences in a binary outcome. I sampled with known probability so I also have weights. If I need a crosstab of age by outcome, do I include both sex and age in the strata statement of the surveyfreq? Why won't surveyfreq let me add age to the strata statement if it is also in the Tables statement.

[Season]

Actually, the STRATA statement in the SURVEY procedures are intended to house the strata used in the sampling design, not the croostabs. For your problem, it is appropriate to use syntax like gender*outcome or age group*gender*outcome like you might frequently do in the FREQ procedure.