# DFBeta Measures for Subsets of Variables

I need to calculate the DFBETA statistics for a logistic regression. The number of columns in the $$X$$ matrix is $$3000$$ and the number of records is in the millions. I've been using R's dfbetas() function which this takes a very long time. I'm primarily interested in $$10$$ variables so is it possible to quickly calculate the DFBETAs for that subset only, instead of for the whole $$X$$ matrix?