(1402) Improving Sepsis Care With a Best Practice Alert in an Oncology-Dedicated Acute Care Center

Introduction: Sepsis is a life-threatening condition that develops in response to an infection and is a leading cause of preventable deaths worldwide. Sepsis in cancer patients confers significantly higher incidence and mortality rates, readmissions, and associated costs as compared to non-cancer patients.
We implemented an on-arrival Sepsis Best Practice Alert (BPA) in an acute cancer care center, to improve identification, management, and outcomes for sepsis among cancer patients.

Methods: A single center multidisciplinary sepsis program was established after process mapping of existing practices was conducted regarding sepsis management. The program included a modified SIRS-based BPA, supported with an algorithm, order-sets, educational modules, dashboard, and a webpage of resources. Process and outcome measures were collected for patients with suspected sepsis from March 2020 through January 2021. Baseline data was collected from a cohort of patients suspected of sepsis from January 2019 through February 2020.

Results: There was a total of 28,475 patient visits (15,773 patients) from January 2019 through January 2021 (17,634 pre- and 10,841 post-implementation). Pre and post-implementation data showed: 1) sepsis mortality index decreased by 30% (1.28 to 0.89, p< 0.001), 2) observed mortality rate for sepsis without septic shock decreased from 17.2% to 10.9% (p < 0.001), 3) observed mortality rate for septic shock remained unchanged (48.6% vs. 48.3%, p=0.970), 4) decreases were observed in median [IQR] times to first antibiotic administration (82 [51-139] min vs. 53 [34-90] min, (p < 0.001), to initial blood culture collection (51 [34-82] pre vs. 36 [25-62] min, (p < 0.001), and to initial lactate collection (54 [35-97] min vs. 41 [29-65] min, (p < 0.001), and 5) hospital length of stay and thirty day readmission rates did not reach statistical significance (9.0 vs. 8.0 days, p=0.068, and 29.8% vs. 28.1%, p >0.05, respectively).

Conclusions: Implementation of a comprehensive sepsis program driven by a sepsis BPA was associated with significant improvement in median time from door to antibiotic administration, sepsis mortality index, and time to collection of initial lactate and blood cultures. The data and lessons learned from the initial improvements can inform futher quality improvement efforts.