Urban residence is a risk factor for sepsis mortality
Mohr NM, Zebrowski AM, Gaieski D, Buckler D, Lang M, Carr BG, and Jayaprakash N. Urban residence is a risk factor for sepsis mortality. Acad Emerg Med 2019; 26:S44.
Acad Emerg Med
Background: Sepsis is common in US EDs, and sepsis mortality remains high. Several prior studies have identified treatment in a low-volume hospital and inter-hospital transfer as independent risk factors for death from sepsis, but the impact of rurality of residence remains unclear. The objective of this study was to test the hypothesis that patients who reside in rural areas have higher mortality than those who live in more urban areas. Methods: Observational cohort study of 2013 administrative claims data from all U.S. Medicare beneficiaries aged 65 years and older. Sepsis was defined by ICD-9-CM criteria for ED admissions resulting in acute inpatient hospitalization, and rurality of residence was defined based on Rural-Urban Continuum Codes of the county of residence based on the residence at the time of hospitalization (9 categories). Logistic regression was used to measure the association between the rurality of residence and mortality, adjusting for comorbidities (Elixhauser methodology). The primary outcome was 30-dayin-hospital mortality. Results: A total of 553,854 sepsis cases were identified among Medicare beneficiaries in 2013, and 47.6% were male. Mortality 30 days after index hospital admission was reported in 32.2% (n=178,092). Most sepsis patients (54.7%) resided in a metropolitan area with more than 1 million residents, and 23.1% lived in an area with a population of less than 250,000. Mortality was highest among residents of metropolitan areas with more than 1 million people (32.8%), and lowest in metropolitan areas with fewer than 250,000 residents (30.9%). Only 2.7% of the cohort was transferred between hospitals, but this proportion climbed to 7.2% of cases among residents of non-metropolitan communities with population between 2,500 and 20,000 residents. Adjusting for comorbidities, age, transfer, county income, and hospital volume, mortality was lower among residents of metropolitan areas with 250,000 - 1 million (aOR 0.96, 95%CI 0.94-0.97) and less than 250,000 (aOR 0.93, 95%CI 0.91-0.95) than residents of metropolitan areas of greater than 1 million residents. Conclusion: Medicare beneficiaries residing in the most urban counties have the highest sepsis mortality. Future studies should better characterize the role of access to health care, treatment in high volume sepsis centers, and health system competition on sepsis survival.