Epidemiology and economics of adult patients hospitalized with urinary tract infections

Abstract

Abstract Background: Urinary tract infections (UTI) are among the most common bacterial diseases worldwide, with significant clinical and economic burden. Surveillance of pathogen epidemiology and risk factors for resistant pathogens in the hospital setting may improve the management of UTI. Objective: To evaluate microbiology and antimicrobial susceptibility of UTI pathogens, with associated costs, in hospitalized patients. Methods: Patients diagnosed with UTI between July and September 2013 were retrospectively screened for clinical symptoms and treatment within 24 hours of admission, then categorized into groups: community acquired (Group 1); recent healthcare exposure (Group 2); or a history of identification of an extended-spectrum beta lactamase (ESBL)-producing organism (Group 3). Clinical, epidemiological, and financial data were compared between groups. Results: From 308 included patients, a total of 216 pathogens were identified. Escherichia coli was most commonly identified pathogen, but frequencies differed between groups (p = 0.002), as did those of ESBL-producing pathogens (p < 0.001) and Pseudomonas aeruginosa (p = 0.005). Appropriate empirical therapy also differed between groups (p = 0.003). Length of stay was longer for healthcare associated UTI with inappropriate empirical therapy (5.2 versus 6.3 days, p = 0.016). Increased cost was associated with factors other than antimicrobial costs. Intensive care unit (ICU) stay (p < 0.001), care facility at discharge (p = 0.001), Foley catheter (FC) present on admission (p = 0.002), and Charlson comorbidity index (CCI) (p = 0.017) predicted increased cost overall, while ICU stay (p < 0.001), time to appropriate therapy (p < 0.001), and CCI (p = 0.015) predicted higher cost in patients with pathogens identified. Conclusions: Changes in antimicrobial susceptibility are evident with exposure to healthcare, the presence of a FC, and a history of resistant pathogens. Risk-based empirical prescribing and rapid de-escalation may improve care and reduce costs.