Association of a sepsis initiative on broad spectrum antibiotic use and outcomes in an ED population

Abstract

IntroductionSepsis identification and treatment is a priority for emergency department (ED) providers and payors alike. However, aggressive metrics aimed at improving sepsis care could have unintended consequences for patients who do not have sepsis. MethodsAll ED patient visits for a one month period before and after a quality initiative to increase early antibiotic use in septic patients were included. Overall broad spectrum (BS) antibiotic use, admission rates, and mortality were compared in the 2 time periods. A more detailed chart review was performed on those who received BS antibiotics in the before and after cohorts. Patient were excluded for pregnancy, age < 18, COVID-19 infection, hospice patients, left ED against medical advice, and if antibiotics were given for prophylaxis. In BS antibiotic treated patients, we sought to determine mortality, rates of subsequent multidrug resistant (MDR) or Clostridium Difficile (CDiff) infections and rates of non-infected patients receiving BS antibiotics. ResultsThere were 7967 and 7407 ED visits in the pre- and post-implementation periods, respectively. Of those, BS antibiotics were administered in a total of 3.9% pre-implementation and 6.2% post-implementation (p ≤ 0.00001). Admission was more common in the post-implementation period, but overall mortality was unchanged (0.9% pre-implementation and 0.8% post-implementation, p = 0.41). After exclusions, 654 patients treated with BS antibiotics were included in the secondary analyses. Baseline characteristics were similar between the pre-implementation and post-implementation cohorts. There was no difference in the rate of CDiff infection or the proportion of patients receiving BS antibiotics who were not infected, but there was an increase in the post-implementation period in MDR infections after ED BS antibiotics, 0.72% vs. 0.35% of the entire ED cohorts, p = 0.0009. ConclusionsWe found that a QI sepsis initiative was associated with an increase in the proportion of patients who received BS antibiotics in the ED, and a small absolute increase in associated subsequent MDR infections, with no apparent effect on mortality in all ED patients or the subset treated with BS antibiotics. Further research is needed to assess the impact on all patients affected by aggressive sepsis protocols and initiatives, rather than only those with sepsis.