Data Analysis for a Clinical Trial of the Management of Urinary Tract Infections in Residential Long-Term Care Facilities

Abstract

The main object of the research is to analyze the effect of the clinical intervention algorithms proposed for reducing antibiotic use for older adults in long-term care facilities (LTCFs) by managing urinary tract infections (UTIs). 20 paired nursing homes were enrolled in the 12-month study. Within each pair, one was randomized to use of the intervention algorithms and the other to use of regular management. Cluster-level paired t-tests (unweighted and weighted) and regression analyses (unweighted and weighted) were used in the analysis of the data. Paired t-tests show that the algorithms did not significantly reduce the antibiotic use, the number of urine cultures or the antibiotic use for urinary infections in most months. However, they did reduce the proportion of antibiotic use for urinary infections significantly in most months. Regression analysis indicates that the difference between the control group and intervention group has no significant increasing or decreasing trend with time (month). And the algorithms significantly reduced the antibiotic use for urinary infections, number of cultures and the proportions through the 12-month study. The analyses reached a similar conclusion using nonparametric methods and weighted analysis.