Methods: We analyzed automated electronic data in a research database of CareFusion-BD from 332 acute care hospitals (1/1/2010 – 9/30/2013). We used the urine Nosocomial Infection Marker (NIM) to identify potential UTI events. The urine NIM is a previously validated algorithm that uses microbiology and census data to identify non-duplicate positive cultures that are potential hospital UTIs. UCI was defined as number of urine specimens collected for culture / number of admissions that had any urine culture performed. Pearson correlation was used to assess the relationship between the UCI and urine NIM rate at the hospital level, stratified by younger (18-64 years) versus older (65+ years) age groups.
Results: For the 6,749,870 younger and 4,313,982 older adult admissions respectively, the proportion of admissions with urine culture performed was 18% vs. 32% (P<0.0001); the urine NIM rates were 47.7 vs. 70.3 per 1000 specimens (P<0.0001), 10.7 vs. 28.8 per 1000 admissions (P<0.0001), and 2.7 vs. 5.6 per 1000 patient-days (P<0.0001). The UCI was 1.26 vs 1.27 for the two age groups. Urine NIM rate per patient-days was positively correlated with UCI for the younger (r=0.66) and older age groups (r=0.58).
Conclusion: Higher urine NIM rates among the older population may in part be explained by a higher prevalence of urine culturing as well as a higher rate of urine culture positivity. Urine culture intensity among patients who had any urine culture performed was positively correlated with observed urine NIM rate in both age groups. Understanding the influence of patient-level factors on urine culturing practices will help guide recommendations on appropriate urine culturing, promoting more accurate measurement of urinary tract infections and improved antimicrobial use.
Y. P. Tabak,
V. Gupta, None
L. Vankeepuram, CareFusion: Employee , Salary
X. Sun, None
C. Gould, None