Socioeconomic disparities and mortality among acute patients with sepsis: a retrospective Danish cohort study.

Abstract

OBJECTIVES: Socioeconomic inequalities exist in infectious diseases and sepsis in high-income countries. We investigated the association between income and mortality among patients with sepsis, overall and among those treated in the intensive care unit (ICU) versus general wards. DESIGN: A retrospective register-based cohort study. SETTING: The Region of Southern Denmark (RSD). PARTICIPANTS: All adult patients with an unplanned contact with a hospital in the RSD from 1 January 2016 to 20 March 2018. Patients with sepsis were identified based on the following criteria: (1) blood culture(s) performed within 48 hours of arrival, (2) antibiotic(s) administered within 48 hours of arrival, (3) a discharge diagnosis of infection and (4) a SOFA (Sequential Organ Failure Assessment) score of ≥2. The cohort was divided into quartiles according to household income. PRIMARY AND SECONDARY OUTCOME MEASURES: Cox proportional hazards models were used to estimate the association between income groups and mortality. The primary outcome was 90-day mortality with 7-day and 365-day mortality as secondary outcomes. All outcomes were calculated overall and stratified by general ward treatment only and ICU admission. RESULTS: We identified 7813 first-time visits with community-acquired sepsis, including 886 ICU admissions (11.3%). Among patients in the lowest income group, sepsis was associated with a HR of 1.16 (95% CI 1.01 to 1.34) for 90-day mortality compared with the highest income group. This association was particularly pronounced at 365-day follow-up: HR=1.24 (95% CI 1.10 to 1.39). No difference was observed in 7-day all-cause mortality, HR=1.13 (95% CI 0.89 to 1.45). The association was not observed among patients admitted to the ICU. CONCLUSION: Low income was associated with increased mortality in patients with sepsis, particularly during long-term follow-up. The impact of income disparities was not observed among patient admitted to the ICU.