The UMB Digital Archive is getting an upgrade! The upgrade requires a content freeze starting 1/27/25 and is expected to last two weeks. Any new user accounts or submissions made to the Archive during this time will not be transferred to the upgraded site. Contact ArchiveHelp@hshsl.umaryland.edu for questions.
Comparing Hospital-Onset Bacteremia to Central Line Associated Bloodstream Infection as a Hospital Quality Measure
Abstract
Background Central Line Associated Bloodstream Infections (CLABSI) rates are a benchmark for hospital quality despite problems with surveillance bias and inter-observer variability and subjectivity. The rate of Hospital-onset bacteremia (HOB) may offer significant advantages over CLABSI; including being more objective and un-biased. Methods We conducted a multisite cohort study via the Society for Healthcare Epidemiology of America (SHEA) research network to examine the relationship between HOB and CLABSI rates and compare ability of each to distinguish between hospitals. Hospitals reported total CLABSIs, central line days, HOBs, patient days, and total blood cultures performed for each ICU and completed a survey relating to CLABSI reporting. Mixed-effect Poisson regression was used to evaluate HOB as a predictor for CLABSI. Standardized infection ratios (SIR) for HOB and CLABSI for medical and neonatal ICUs were calculated using the pooled mean rates of the study sample as the benchmark. Results We obtained data for 79 ICUs from 15 hospitals within the US and Canada. From January 2012 to December 2013, 627 CLABSIs, 11 024 HOB, 464 224 central line days and 959 647 ICU patient days were reported. HOB was a strong predictor of CLABSI; a change in the rate of HOB of 1 predicted a relative change of 2.2% in CLABSI rate. Standardized infection ratios for HOB and CLABSI for medical and neonatal ICUs showed large confidence intervals that overlapped with each other for the CLABSI measure with 14 of 18 (77.7%) CLABSI 95% confidence intervals containing the null value of 1, compared to only 6 of 18 (33.3%) HOB 95% confidence intervals (p-value 0.02, fisher's exact test). CLABSI reporting requires 15.8 hours of nurse time per month. Conclusions In this multicenter study, HOB rates were strongly predictive of CLABSI rates. HOB can be collected automatically saving nurse time and provide better discrimination hospital quality than CLABSI.Description
University of Maryland, Baltimore. Epidemiology and Preventive Medicine. M.S. 2014Keyword
central line-associated bloodstream infectionhospital associated infection
hospital-onset bacteremia
outcome measures in healthcare quality
standardized infection ratio
surveillance of healthcare-associated infections
Cross Infection
Outcome Assessment (Health Care)
Identifier to cite or link to this item
http://hdl.handle.net/10713/4377Related items
Showing items related by title, author, creator and subject.
-
Infection of fetal neural tissue by guinea pig cytomegalovirus: Association of cell types infected and neuropathogenesis.The specific relationship between human congenital cytomegalovirus infection, neuropathological changes such as microcephaly and calcifications and neurologic dysfunction such as mental retardation is not clear. Of special interest is how a limited number of infected cells leads to global effects such as microcephaly and mental retardation. Guinea pig cytomegalovirus (GPCMV) was used to investigate which cell types in the guinea pig fetal nervous system are infectable and the relationship between infection and associated problems. Monoclonal antibodies with specific reactivity against GPCMV proteins were made. One was found to label infected cells in vivo and in vitro and was characterized as identifying a late structural protein of GPCMV. In cells infected by GPCMV, a comparison of viral protein production was undertaken using the newly created monoclonal antibody against GPCMV and antisera against multiple GPCMV proteins. GPCMV infected cells were identified using the monoclonal antibody and anti-GPCMV antisera. The specific fetal brain cell types infected by GPCMV were identified in vitro and in vivo by anatomical localization and immunocytochemical labelling with cell specific antibodies directed against neurons, astrocytes, progenitor cells, macrophages, and endothelial cells. In fetal brain cells, fewer infected cells expressed the late structural protein recognized by the monoclonal antibody than were labelled with the polyclonal antisera. This difference was not seen in infected salivary gland or fibroblasts. In vivo infection was determined to be localized primarily to areas adjacent to the lateral ventricles. Endothelial cells and ependymal cells were determined to be infected in vivo by GPCMV. Macrophages and progenitor cells were found to be infected by GPCMV in vitro. These studies suggest that not only are certain cell types infected but the pattern of viral protein expression may differ between cell types and perhaps developmental stage. Infection of endothelial, ependymal and progenitor cells may destroy the substrate necessary for normal brain development and lead to the developmental derangements of the fetal brain known to be associated with human congenital cytomegalovirus infection.
