Abstract
BACKGROUND AND OBJECTIVES: Racial/ethnic and socioeconomic disparities are reported in sepsis, with increased mortality for minority and low socioeconomic status groups; however, these studies rely on billing codes that are imprecise in identifying sepsis. Using a previously validated algorithm to detect pediatric sepsis using electronic clinical data, we hypothesized that racial/ethnic and socioeconomic status disparities would be evident in this group. METHODS: We performed a retrospective study from a large, quaternary academic center, including sepsis episodes from January 20, 2011, to May 20, 2021, identified by an algorithm indicative of bacterial infection with organ dysfunction (cardiac, respiratory, renal, or hematologic). Multivariable logistic regression was used to measure association of race/ethnicity, insurance status, and social disorganization index, with the primary outcome of mortality, adjusting for age, sex, complex chronic conditions, organ dysfunction on day 1, source of admission, and time to hospital. Secondary outcomes were ICU admission, readmission, organ dysfunction-free days, and sepsis therapies. RESULTS: Among 4532 patient episodes, the mortality rate was 9.7%. There was no difference in adjusted odds of mortality on the basis of race/ethnicity, insurance status, or social disorganization. There was no significant association between our predictors and ICU admission. Hispanic patients and publicly insured patients were more likely to be readmitted within 1 year (Hispanic odds ratio 1.28 [1.06-1.5]; public odds ratio 1.19 [1.05-1.35]). CONCLUSIONS: Previously described disparities were not observed when using electronic clinical data to identify sepsis; however, data were only single center. There were significantly higher readmissions in patients who were publicly insured or identified as Hispanic or Latino, which require further investigation.