Fig 2. Estimates of appropriate empirical antibiotic therapy on 30-day mortality (antibiotic monotherapy).

CI, confidence interval. Confounding variables: age, arrival, triage category, direct intensive care unit admittance, chills, need for vasopressors, body temperature, respiratory rate, systolic blood pressure, supplemental oxygen, consciousness, diabetes mellitus with end-organ damage, mild liver disease, malignancy, chronic kidney disease, chronic obstructive pulmonary disease, cerebrovascular accident or transient ischemic attack, and dementia. For a detailed description of statistical adjustment techniques, see S2 Methods. This figure shows attenuation of estimates after adjustment for confounders.