Racial and Ethnic Disparities in Multisystem Inflammatory Syndrome in Children in the United States, March 2020 to February 2021

Abstract

Background: The incidence of multisystem inflammatory syndrome in children (MIS-C) varies by race and ethnicity. This study assessed whether disparities in MIS-C in the United States by race and ethnicity exceed known disparities in coronavirus disease 2019 (COVID-19) incidence.

Methods: We compared the distribution of race and ethnicity among patients with MIS-C (<21 years of age, termed children) with onset March 2020 to February 2021 to that of children with COVID-19 and in the general population. Analysis was restricted to 369 counties with high completeness of race and ethnicity reporting for MIS-C and COVID-19. For each racial and ethnic group, observed numbers of patients with MIS-C were compared with expected numbers (observed/expected ratio) in children with COVID-19 and in the general population within these counties.

Results: Compared with children in the general population, MIS-C was more frequent among Hispanic (139% of expected) and non-Hispanic Black children (183%) and less frequent among non-Hispanic White (64%) and non-Hispanic Asian children (48%). Compared with children with COVID-19, MIS-C was more frequent in non-Hispanic Black children (207% of expected) and less frequent in non-Hispanic White children (68%); however, frequency was not different among Hispanic (102%) and non-Hispanic Asian (74%) children.

Conclusions: Disparities in MIS-C by race and ethnicity exist, even after controlling for COVID-19 disparities and geographic variations. The high proportion of MIS-C among Hispanic children and low proportion among non-Hispanic Asian children align with COVID-19 rates, while the high proportion among non-Hispanic Black children and low proportion among non-Hispanic White children are not explainable by COVID-19 rates.

FIGURE 1.

Observed distribution of race and ethnicity among patients with MIS-C in the study subset (n = 1382) compared with the expected distribution of race and ethnicity derived from those with COVID-19 under 21 years of age and the underlying population under 20 years of age within the same counties. *From CDC COVID-19 case surveillance database. †From the 2019 Vintage Census Population Estimates. ‡Significantly different from observed proportion with MIS-C of the same race/ethnicity at P < 0.05.

FIGURE 2.

Sensitivity analyses showing the O/E of MIS-C compared with COVID-19 for each race and ethnicity by age and by onset date. Dark gray boxes denote zero observed patients with MIS-C are available for comparison at that filtering criteria. Boxes are shaded lightly when the O/E ratio is close to 1 and become darker as the O/E ratio moves further from 1. Rows with consistent intensity of coloration/shading and direction of arrows suggest minimal effect of age or of onset date. A, O/E ratios when limiting analysis to specific age groups. B, O/E ratios when limiting analysis to specific onset dates (early: March 1 to June 30, 2020; mid: July 1 to October 31, 2020; late: November 1, 2020 to February 28, 2021). H indicates Hispanic; NHA, non-Hispanic Asian; NHAIAN, non-Hispanic American Indian or Alaska Native; NHB, non-Hispanic Black; NHMult, non-Hispanic multiple races; NHNHPI, non-Hispanic Native Hawaiian or Pacific Islander; NHW, non-Hispanic White. *Statistical significance at P < 0.05.