Multivariate analysis for predictors of antifungal treatment of funguria in hospitalized patients.

摘要

This study describes the association between fifteen predictors (age, hospital site, underlying host disease, physician treating type and so on) and treatments of fungal urinary tract infection (antibiotic treatment and treatment with urinary catheters removal or change). Utilizing bivariate analysis and multivariate analysis on data from a prospective observational study of ten U.S. medical centers of hospitalized patients with a urine culture positive for yeast, we found that antibiotic treatment was significantly associated with eight independent predictors (site, fever, pyuria, presence of urinary device, yeast, urinary colony, physician treating type and treatment with urinary catheter removal or change) and treatment with catheter removal or change was significantly related with five predictors (site, diabetes, presence of urinary device, urinary colony and receipt of antibiotic treatment) in multivariate models. Training medical staff in specific hospitals with the significance of limiting unnecessary use of antibiotics and considering urinary catheter removal or change is highly recommended.