To describe and compare the epidemiology of Sepsis-1 and Sepsis-3 in Yuetan Subdistrict of Beijing, and to estimate the incidence of Sepsis-3 in China. A population-based cohort study was conducted. Through the database of Beijing Public Health Information System, the medical records of all adult residents hospitalized from July 1st, 2012 to June 30th, 2014 in Yuetan Subdistrict were reviewed. According to the clinical data of these patients, patients with Sepsis-1 and Sepsis-3 were enrolled in this analysis and the demographic characteristics of them were compared. Incidence and in-hospital mortality was calculated. Logistic regression method was used to analyze the risk factors of sepsis. Compared with non-septic patients with infections, patients with Sepsis-1 or Sepsis-3 were more likely to be male, older, had more comorbidities and lower body mass index (BMI), had more lower respiratory tract infections, more intra-abdominal infections and more blood flow infections, but had fewer urogenital tract infections, fewer upper respiratory tract infections, fewer gastrointestinal infections and fewer skin and soft tissue infections, and had pure prognosis with longer length of hospital stay [days, Sepsis-1 compared with non-Sepsis-1: 18 (10, 34) vs. 14 (9, 22), Sepsis-3 compared with non-Sepsis-3: 20 (11, 39) vs. 14 (9, 25)] and higher mortality [Sepsis-1 compared with non-Sepsis-1: 20.6% (353/1 716) vs. 2.5% (44/1 733), Sepsis-3 compared with non-Sepsis-3: 32.0% (299/935) vs. 3.9% (98/2 514), all P < 0.01]. Logistic regression analysis showed that male, elder (age ≥ 65 years old), low BMI, bed-rest state, and combined with cerebrovascular disease and hematological malignancies were risk factors for Sepsis-1 [all odds ratio (OR) > 1, all P < 0.05], while the male, elder (age ≥ 65 years old), low BMI, bed-rest state, and combined with cerebrovascular disease, rheumatic immune disease, malignant diseases of blood system and dementia were risk factors for Sepsis-3 (all OR > 1, all P < 0.05). Sepsis-3 was more common in males [OR = 1.19, 95% confidence interval (95%CI) was 1.01-1.40, P < 0.05] and elderly patients (age 65-84 years old: OR = 1.60, 95%CI was 1.28-1.99, P < 0.01; age ≥ 85 years old: OR = 1.76, 95%CI was 1.39-2.23, P < 0.01) as compared with Sepsis-1. After adjusted for gender and age, the standardized incidence of Sepsis-1 was 461 per 100 000 person-year, and that of Sepsis-3 was 236 per 100 000 person-year, with the standardized mortality of 79 per 100 000 person-year and 67 per 100 000 person-year, respectively, in Yuetan Subdistrict of Beijing. Corresponding to a speculative extrapolation of 4 856 532 new cases for Sepsis-1 and 2 487 949 new cases for Sepsis-3, there were 831 674 deaths and 700 437 deaths per year in China, respectively. Male, elder, more comorbidities and low BMI were risk factors for sepsis. The standardized incidence of Sepsis-3 in Yuetan Subdistrict of Beijing was 236 per 100 000 person-year, and speculated there were 2.5 million new cases of Sepsis-3 per year, resulting in more than 700 000 deaths in China. According to the diagnostic criterion of Sepsis-3, 2.36 million new cases per year were reduced, and the mortality was increased by 11.4%, as compared with the criterion of Sepsis-1.