BACKGROUND: More than a million people a year in the United States experience sepsis or sepsis-related complications, and sepsis remains the leading cause of in-hospital deaths. Unlike many other leading causes of in-hospital mortality, sepsis detection and treatment are not dependent on the presence of any technology or services that differ between tertiary and non-tertiary hospitals.

OBJECTIVE: To compare sepsis mortality rates between tertiary and non-tertiary hospitals in Washington State.

METHODS: A retrospective longitudinal, observational cohort study of 73 Washington State hospitals for 2010-2015 using data from a standardized state database of hospital abstracts. Abstract records on adult patients (n = 86,378) admitted through the emergency department (ED) from 2010 through 2015 in all tertiary (n = 7) and non-tertiary (n = 66) hospitals in Washington State.

RESULTS: The overall mortality rate for all hospitals was 6.5%. In the fully adjusted model, the odds ratio for in-hospital death was higher in non-tertiary hospitals compared with tertiary hospitals (odds ratio 1.25; 95% confidence interval 1.17-1.35; p < 0.001).

CONCLUSIONS: We observed higher sepsis mortality rates in non-tertiary hospitals, compared with tertiary hospitals. Because most patients who are treated for sepsis are treated outside of tertiary hospitals, and the number of patients treated for sepsis in non-tertiary hospitals seems to be rising, a better understanding of the cause or causes for this differential is crucial.