Presentation (CTI)

OBJECTIVES: To evaluate the causal impact of early antibiotic administration on patient outcomes in sepsis management, we followed the 2021 Surviving Sepsis Campaign guidelines, specifically focusing on the recommendation to administer antibiotics within three hours of sepsis recognition.
METHODS: Using a Double Machine Learning (DML) approach, we analyzed observational data to determine the effects of meeting this 3-hour threshold on key outcomes, including the length of hospital and ICU stays, as well as hospital and ICU mortality rates. We used the eICU Collaborative Research Database, a large multi-center critical care dataset from hospitals across the United States.
RESULTS: Although initial summary statistics suggested a substantial influence of the timing of antibacterial therapy on patient outcomes, indicating earlier treatment with reduced mortality rates and shorter hospital and ICU stays, our DML model refined these results. It showed that administering antibiotics within the 3-hour guideline primarily affected the length of hospital and ICU stays, with no significant impact on mortality rates. Hospital stay duration was reduced by an estimated -1.73 days (95% CI [-2.33, -1.13]), while ICU stay was shortened by -0.67 days (95% CI [-0.98, -0.37]).
CONCLUSIONS: Given the substantial hospital costs associated with septicemia, these findings highlighted the importance of timely treatment in improving resource utilization and patient recovery, even if survival outcomes remained unaffected.