770. Multimodal Sepsis Performance Improvement Initiative Improves Early and Appropriate Treatment, Reduces Sepsis-related Readmissions, and Reduces Overall Mortality

BACKGROUND: CMS has implemented the SEP-1 Core Measure, which mandates that hospitals implement sepsis quality improvement initiatives. At our hospital, a 900-bed tertiary hospital, a sepsis performance improvement initiative was implemented in April 2016. In this study, we analyzed patient outcomes before and after these interventions. METHODS: We studied coding data in patients with a diagnosis of Sepsis reported to CMS using a third-party performance improvement database between October, 2015 and July, 2017. The interventions included a hospital-wide education campaign about sepsis; a 24–7 electronic warning system (EWS) using SIRS criteria; a rapid response nursing team that monitors the EWS; a 24–7 mid-level provider team; a database to monitor compliance and timely treatment; and education in sepsis documentation and coding. We performed a before and after analysis of patient outcomes. RESULTS: A total of 4,102 patients were diagnosed with sepsis during the study period. 861 (21%) were diagnosed during the pre-intervention period and 3,241 (80%) were diagnosed in the post-intervention period. The overall incidence of sepsis, severe sepsis, and septic shock were 59%, 13%, and 28% consecutively. Regression analysis showed age, admission through the ED, and severity of illness as independent risk factors for increased mortality. Adjusted for these risk factors, the incidence of severe sepsis and septic was reduced by 5.3% and 6.9% in the post-intervention period, while the incidence of simple sepsis increased by 12%. In the post-intervention period, compliance with all 6 CMS mandated sepsis bundle interventions improved from 11% to 37% (P = 0.01); hospital length of stay was reduced by 1.8 days (P = 0.05); length of stay above predicted was less by 1.5 days (P = 0.05); re-admission rate was reduced by 1.6% (P = 0.05); and death from any sepsis diagnosis was reduced 4.5% (P = 0.01). Based on an average of 2000 sepsis cases at our hospital, this amounted to 90 lives saved per year. Death from severe sepsis and septic shock both were also reduced by 5% (P = 0.01) and 6.5% (P = 0.01). CONCLUSION: A multi-modal sepsis performance improvement initiative reduced the incidence of severe sepsis and septic shock, reduced hospital length of stay, reduced readmission rates, and reduced all-cause mortality. DISCLOSURES: All authors: No reported disclosures.