Covid-19 Icu Admission Disparities among Hispanics and African Americans in Boston

ABSTRACT

INTRODUCTION: Racism has been identified as a driver of health disparities. The COVID pandemic has widened the gap between Whites and racial minorities, resulting in an even greater burden of disease and poorer health outcomes. The Boston area has a greater wealth disparity between these groups compared to the national average. We hypothesize that African American and Hispanic groups in the Boston Area have carried a greater burden of severe disease compared to Whites. METHOD(S) This cross-sectional study included 1,272 single-event adults admitted to Beth Israel Deaconess Medical Center (Boston, MA) due to COVID from March 2020 to April 2022. Patients were grouped by demographics captured in the medical records. Three groups were determined to have the appropriate sample sizes for analysis: Hispanics of any race, African American Non-Hispanics, and White Non-Hispanics. The primary outcome assessed was ICU admission rates;secondary analyses included length of hospitalized and ICU stay and comorbidity rates. Statistical analyses were performed in R Version 3.0. RESULT(S) Out of our sample, 31% were African American (AA), 20% Hispanic, and 49% White. Compared to Whites, ICU admission rates for AA patients were higher than for Hispanics, with an odds ratio (OR) of 1.45 (95% CI, [1.11,1.91]) and 1.21 [1.27, 1.64], respectively. When adjusting for sex, age, and comorbidities, the same pattern was observed ICU admission rates for AAs were positively associated for both males (1.64, [1.11,2.43] and females (1.19, [1.04,1.36]), but only for Hispanic females (1.44, [1.05,1.97]). AAs and Hispanics had a significantly higher OR of having 3+ comorbidities (1.3, [1.13,1.72] and 1.47 [1.12, 1.96]). CONCLUSION(S) Ethnic minority groups have suffered a disproportionately greater burden of disease related to COVID. Our study shows that ICU admission was positively associated with AA and Hispanic race, opposite to what had been previously shown in some publications. Our findings could help reorient public health measures to improve health outcomes in these populations.