Search for a command to run...
Abstract Background Inappropriately ordered and or collected urine cultures have been associated with high rates of urine culture contamination and associated downstream consequences, such as over-prescribing of antimicrobial (antibiotics or antifungal) medications. This study utilized a large sample of hospitals with geographical representation across the United States (U.S.) to quantify the frequency of urine culture contamination and antimicrobial utilization. Methods This multicenter, retrospective, de-identified cohort analysis included urine specimens from hospitalized adults aged 18 years or older from 159 U.S. facilities in the BD Insights Research Database (Becton Dickinson and Company, Franklin Lakes, NJ) from 2018 through 2022. The facilities included were diverse and included both small and large facilities with and without teaching status. Contaminated urine cultures were identified using a proprietary algorithm that considered mixed flora, organism type, and used colony thresholds for various organisms. To evaluate potential over-prescribing of antimicrobial medications, a subgroup analysis was conducted in patients with contaminated urine cultures without positive cultures from any other specimens (specifically, blood, respiratory, and wound) within 10 days of the urine collection date (-2 days and +7 days). This time-period was evaluated for new antimicrobial orders that were not one-time medications or prescribed as needed and the inpatient duration of these orders. To understand the differences in antimicrobial ordering by the testing of other specimen cultures, the type of antimicrobial order and the duration was estimated using frequencies and means within the subgroup and by the testing status of another culture (any culture, respiratory culture, or blood culture testing). Results A total of 1,252,986 urine cultures were included in the analysis from 1,089,069 inpatient encounters. Of these urine cultures, 16.5% met the criteria for urine specimen contamination. Among the 206,562 contaminated urine samples, a subgroup of 101,187 (49%) did not have a positive culture from any other source within 10 days and were included in the analysis evaluating antimicrobial orders. Antimicrobial orders were observed in 64.7% of the subgroup with an average duration of 2.6 inpatient days. Despite no other positive culture, antimicrobial utilization varied by the testing status of other specimens. The frequency of antimicrobial ordering was the least common in patients with a urine culture contamination without testing cultures in another specimen (52.8%) and the most common in patients with a urine culture contamination with a blood culture (83.4%). In patients with a contaminated urine specimen without testing for another culture, the 5 most common antimicrobials that accounted for 80% of the antimicrobial use were: ceftriaxone, cefazolin, levofloxacin, piperacillin/tazobactam and intravenously administered vancomycin. Conclusion Over half (65%) of patients with urine culture contamination without another source positive culture in the inpatient setting have antimicrobial orders. An assessment of the clinical care gaps associated with contamination/ false positive cultures, as well as their downstream outcomes impact may help inform quality improvement-based interventional studies that deal with very specific healthcare associated infections.