Objective: Sepsis is a frequent cause of admission, but incidence rates based on administrative data have previously produced large differences in estimates. The aim of the study was to estimate the incidence of community-acquired sepsis based on patients' symptoms and clinical findings at arrival to the hospital.
Design: Population-based survey.
Setting: Medical emergency department from September 1, 2010, to August 31, 2011.
Patients: All patients were manually reviewed using a structured protocol in order to identify the presence of infection. Vital signs and laboratory values were collected to define the presence of systemic inflammatory response syndrome and organ dysfunction.
Measurements and main results: Incidence rate of sepsis of any severity. Among 8,358 admissions to the medical emergency department, 1,713 patients presented with an incident admission of sepsis of any severity, median age 72 years (5-95%; range, 26-91 yr), 793 (46.3%) were men, 728 (42.5%) presented with a Charlson comorbidity index greater than 2,621 (36.3%) were admitted with sepsis, 1,071 (62.5%) with severe sepsis, and 21 (1.2%) with septic shock. Incidence rate was 731/100,000 person-years at risk (95% CI, 697-767) in patients with sepsis of any severity, 265/100,000 person-years at risk (95% CI, 245-287) in patients with sepsis, 457/100,000 person-years at risk (95% CI, 430-485) in patients with severe sepsis, and 9/100,000 person-years at risk (95% CI, 6-14) in patients with septic shock.
Conclusions: Based on symptoms and clinical findings at arrival, incidence rates of patients admitted to a medical emergency department with sepsis and severe sepsis are more frequent than previously reported based on discharge diagnoses.