Search

AbstractObjective:We compared the number of blood-culture events before and after the introduction of a blood-culture algorithm and provider feedback. Secondary objectives were the comparison of blood-culture positivity and negative safety signals before and after the intervention.Design:Prospective cohort design.Setting:Two surgical intensive care units (ICUs): general and trauma surgery and cardiothoracic surgeryPatients:Patients aged ≥18 years and admitted to the ICU at the time of the blood-culture event.Methods:We used an interrupted time series to compare rates of blood-culture events (ie, blood-culture events per 1,000 patient days) before and after the algorithm implementation with weekly provider feedback.Results:The blood-culture event rate decreased from 100 to 55 blood-culture events per 1,000 patient days in the general surgery and trauma ICU (72% reduction; incidence rate ratio [IRR], 0.38; 95% confidence interval [CI], 0.32–0.46; P< .01) and from 102 to 77 blood-culture events per 1,000 patient days in the cardiothoracic surgery ICU (55% reduction; IRR, 0.45; 95% CI, 0.39–0.52; P< .01). We did not observe any differences in average monthly antibiotic days of therapy, mortality, or readmissions between the pre- and postintervention periods.Conclusions:We implemented a blood-culture algorithm with data feedback in 2 surgical ICUs, and we observed significant decreases in the rates of blood-culture events without an increase in negative safety signals, including ICU length of stay, mortality, antibiotic use, or readmissions.