Every hour of delay between onset of sepsis and administration of the sepsis bundle increases sepsis mortality. Hospital-onset sepsis is known to have higher mortality than community-onset sepsis, but the differences in care between cases of hospital-onset and community-onset sepsis have not been characterized. The purpose of this study was to examine adherence to the sepsis bundle when sepsis arises in different areas of the hospital (ward, intensive care unit [ICU], or peri-operative area) compared to community-onset sepsis.
Retrospective cohort study using clinical data from 4 University of California hospitals. Admissions for sepsis from 2014-2016 were identified by diagnosis codes. Generalized linear models were used to estimate likelihood of adherence in terms of relative risk (RR). Time-to-event analysis involved Cox proportional hazards models and Kaplan-Meier curves.
Overall, the sepsis bundle was administered in accordance with guidelines in 11.1% of cases. On multivariable analysis, cases of hospital-onset sepsis were less likely to receive the sepsis bundle within the recommended time frame (adjusted RR 0.57, p < 0.001) than were cases of community-onset sepsis, including a lower likelihood of having blood cultures drawn (adjusted RR 0.75, p < 0.001), serum lactate checked (adjusted RR 0.48, p < 0.001), or broad-spectrum antibiotics administered within 3 hours (adjusted RR 0.65, p < 0.001). Among the cases of hospital-onset sepsis, those arising in the ICU were more likely than those arising on the ward or in the perioperative area to receive the sepsis bundle within the recommended time frame (RR 1.83, p = 0.002). On time-to-event analysis, hospital-onset was associated with significant delays for all 3-hour bundle components except intravenous fluids.
Developing sepsis in one of the inpatient hospital areas was associated with a delay in guideline-adherent sepsis care. The ward and perioperative area experienced the longest delays. Further research is needed to determine if these delays in care contribute to the increased mortality associated with hospital-onset sepsis.
D. Bell, None
M. Wong, None
See more of: Poster Abstract Session