Log binomial regression with complex survey data and potential collinearity?

[SAS93]

I'm using complex survey data (PRAMS) to look at "risk" of a binary outcome using log binomial regression. 

I'm not too familiar with LBR, but I've read that the only real difference is the link function, so I considered setting up a PROC SURVEYLOGISTIC code block & specifying LINK = LOG. These models are planned to be stratified (DOMAIN) by race/ethnicity.

The only issue is that I'm concerned about potential multicollinearity. I used CORRB to check the standard, weighted logistic model and didn't find any high correlation between the main exposure variable and the other control variables, but is this accurate of a method to be able to ignore that potential issue? 

I guess I'm trying to figure out how (or if I can) use log binomial regression with complex survey data, while also trying to account for potential multicollinearity. 

The SAS resources all usually point back to using PROC GENMOD to estimate log binomial models, but you can't use that with complex survey data. And regression-type models make it a bit difficult to look at collinearity.

Which battle should I choose?  

[SteveDenham]

Well, collinearity doesn't depend on the dependent variable.  Consequently, if you want to look at collinearity diagnostics you can use PROC REG.  If you have categorical variables, you'll likely have to run GLMMOD first to set up all the dummy variables.

If you have IML, you could manipulate your design matrix from SURVEYLOGISTIC to accomplish the eigenvalue based tests for collinearity.  Check @Rick_SAS 's blog for this topic.

SteveDenham.