Quality Coordinator Aultman Hospital, Ohio, United States
Introduction: The purpose of this project was to evaluate the impact of a system-automated sepsis alert on: the number of sepsis alerts called, the percentage of patients who met severe sepsis criteria with a sepsis alert, and SEP1 bundle compliance at one suburban teaching hospital. To increase early treatment of sepsis, a “sepsis alert” system has been developed. This sepsis alert system can be more effective when combined with electronic systems, as well as the amount of time from diagnosis to antibiotic administration. When this electronic system is automated and combined with the Emergency Department command system, it has been shown to improve sepsis mortality in the highest acuity patients.
Methods: Quality improvement data collected for sepsis alerts was analyzed before and after the automated alert was implemented. This included the raw number of sepsis alerts paged per month, the percentage of patients who qualified for the sepsis core measure set that had a sepsis alert paged, and chart abstraction evaluating SEP1 compliance of sampled patients for the sepsis core measure.
Results: The goal of sepsis alerts paged in 2023 was 100 per month, this was reached with an average of 113 sepsis alerts per month. For 2024, the goal was increased to 130 alerts per month. Hospital average for quarter one and quarter two of 2024 is 172 alerts per month, with 80 (23%) of those being an automated alert. Patients who met severe sepsis criteria in 2023, 60% had a sepsis alert paged, increasing to an average of 79% after implementation. SEP1 compliance increased from 66% in 2023 to 75% after quarter two 2024. Lives saved for qualified patients was measured through expected mortality minus observed mortality. In 2023, 22 lives were estimated to have been saved and after initiation of automated sepsis alerts, 26 lives are estimated to have been saved through quarter two. Data through quarter 4 of 2024 will be presented.
Conclusions: Automated sepsis alert implementation increased the volume of early identification of those displaying signs of sepsis, improved 1-hour bundle compliance, and has potential to improve outcomes for sepsis patients. The study institution is one entity in a multi-entity healthcare system and will look to implement the automated sepsis alert through the health care delivery system.
