Presenting Symptoms and Antibiotic Timing Among Hospitalized Adults With Presumed Sepsis

Importance Timely antibiotics are associated with reduced sepsis mortality, particularly among patients with hypotension; however, antibiotic delays are common in practice. Presenting symptoms may influence antibiotic timing and confound the association between antibiotic timing and outcomes. Objectives To evaluate whether any presenting symptoms are associated with antibiotic timing and to assess risk-adjusted 30-day mortality by antibiotic timing after adjustment for symptoms. Design, Setting, and Participants This cohort study was conducted among adults hospitalized for community-onset presumed bacterial sepsis at 67 hospitals in the Michigan Medicine Safety Consortium from November 2020 to May 2024. Exposures Presenting symptoms; time from hospital presentation to first antibiotic administration. Main Outcomes and Measures Timely antibiotics, defined as antibiotic administration within 3 hours of hospital presentation for patients with hypotension and within 5 hours for patients without hypotension, was the primary outcome. Secondary outcomes included time to antibiotics, measured as a continuous variable, and 30-day mortality. Results Among 29 647 patients included in this analysis, the median (IQR) age was 71 (61-80) years, 14 676 (49.5%) were male, and the median (IQR) Charlson Comorbidity Index was 3 (1-5). Among 6704 patients (22.6%) presenting with hypotension, 4413 (65.8%) received timely antibiotics (within 3 hours). Among 22 942 (77.4%) patients presenting without hypotension, 17 544 (76.5%) received timely antibiotics (within 5 hours). In multivariable analyses, presenting symptoms were independently associated with antibiotic timing. Among patients with and without hypotension, prehospital fever was associated with increased odds of timely antibiotics (with hypotension: adjusted odds ratio [aOR], 1.34 [95% CI, 1.19 to 1.52]; without hypotension: aOR, 1.26 [95% CI, 1.17 to 1.35]), while gastrointestinal symptoms were associated with decreased odds of timely antibiotics (with hypotension: aOR, 0.84 [95% CI, 0.75 to 0.94]; without hypotension: aOR, 0.80 [95% CI, 0.75 to 0.86]). Urinary and respiratory symptoms were associated with antibiotic timing for patients without hypotension (urinary symptoms: aOR, 1.13 [95% CI, 1.05 to 1.22]; respiratory symptoms: aOR, 0.91 [95% CI, 0.85 to 0.99]). Timely antibiotics were associated with reduced 30-day mortality (adjusted absolute difference, −2.10 [95% CI, −3.01 to −1.10] percentage points) among the overall cohort, after adjustment for presenting symptoms. Conclusions In this cohort study of 29 647 adults presenting with community-onset presumed bacterial sepsis, presenting symptoms were associated with timing of antibiotic therapy. Even after adjusting for presenting symptoms, timely antibiotics remained associated with lower mortality. These findings suggest that proactive screening for sepsis among patients with less obvious clinical presentations may promote timely recognition and treatment.