Introduction: Sepsis is a common reason for hospitalization and admission to intensive care units. This condition involves significant use of human and economic resources. Several factors are associated with reduced survival, among them age. With epidemiological and demographic transitions, life expectancy has increased worldwide, leading to a larger population over the age of 80, with a higher comorbid burden and immune alterations due to aging. Objective: To establish the 30-day survival of patients over 80 years old diagnosed with sepsis and admitted to intensive care units in Colombia during 2019. Methodology: This is a survival study conducted on a national cohort of patients. It included hospital admissions to intensive care units of individuals over 80 years old with a diagnosis of sepsis, identified through ICD-10 codes and the administration of antibiotics indicated for sepsis. Administrative databases from 2019 (RUAF-CUPS-ICD-10) were used. The 30-day survival of this population group was estimated using death certificates obtained from the RUAF database, and adjusted hazard ratios were calculated. A multilevel analysis by region was performed to determine the influence of the geographical location of care on mortality in this population group. Additionally, simple and multiple Cox regressions were conducted within a mixed-effects model where regions were treated as random effects. Results: A total of 2,849 patients were included. The overall 30-day survival was 76% (95% CI 0.748-0.779). Patients aged 85-90 years (HR 1.22, 95% CI 1.03-1.46) and those over 90 years old (HR 1.84, 95% CI 1.51-2.23) had a higher risk of death from sepsis compared to the group of patients aged 80-85 years. It was found that certain comorbidities and the geographical region influence mortality in this population. For instance, patients with chronic kidney disease had an increased risk of death with an HR of 1.81 (95% CI 1.46-2.24), as did those from certain geographic regions of Colombia. Discussion: We found that 30-day survival in patients over 80 years old admitted to intensive care for sepsis is similar to what has been reported in the literature. Variations in mortality could be due to different definitions of sepsis, its modification over time, and the follow-up period found in the literature. Factors such as age, geographic region of care, and a history of chronic kidney disease influence the likelihood of surviving a septic episode in this population. These factors should be considered in decision-making processes focused on the patient, with the hope of improving survival rates and reducing the burden of this potentially devastating disease.
