Abstract
Background: Community-onset (CO) bloodstream infections (BSI) are those BSI where the blood culture is drawn <48 hours from hospital admission. However, exact times of culture draw or hospital admission are not always available. We evaluated the validity of using 2- or 3- calendar day based definitions for CO-BSI by comparing to a "gold standard" 48-hour definition. Findings: Among the population-based cohort of 14,106 episodes of BSI studied, 10,543 were classified as CO based on "gold standard" 48-hour criteria. When 2-day and 3-day definitions were applied, 10,396 and 10,707 CO-BSI episodes were ascertained, respectively. All but 147 (1.4%) true CO-BSI cases were included by using the 2-day definition. When the 3-day definition was applied, all cases of CO-BSI were identified but and additional 164 (1.5%) cases of hospital-onset HO-BSI were also included. Thus the sensitivity and specificity of the 2-day definition was 98.6% and 100% and for the 3-day definition was 100% and 98.5%, respectively. Overall, only 311 (2.2%) cases were potentially miss-classifiable using either the 2- or 3-calendar day based definitions. Conclusions: Use of either a 2- or 3-day definition is highly accurate for classifying CO-BSI.
Author supplied keywords
Cite
CITATION STYLE
Laupland, K. B., Gregson, D. B., & Church, D. L. (2015). Validity of calendar day-based definitions for community-onset bloodstream infections. BMC Research Notes, 8(1). https://doi.org/10.1186/s13104-015-1051-x
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.