http://goo.gl/oCzlTn
An old study from from 1954 to 1964 presented data data in England on paraplegics. They said, "The results where no catheterisation or only intermittent catheterisation had been used by the referring hospital are superior to those where a Gibbon catheter was used and infinitely superior to those where a Foley catheter was used."
They did not use p-values (maybe that was not the standard in those days?), but if you run the math based on the data in the article, you find that the rate of bacteriuria is 0% in the no catheter group, 7.5% in the intermittent catheter group, and 60.9% in the Foley catheter group!
In a more recent study, a group at Brown University (Rhode Island Hospital) did a review of the trauma registry at their Level 1 trauma center from 2003 to 2008, which included over 5,700 patients. They found that after controlling for other factors like injury severity, diabetes, age, etc, patients with a UTI (who comprised 11.9% of the patients) had an in-hospital mortality of 9.6%, significantly higher than those patients without a UTI (3.5%, p < 0.001).