UCLA

Background

Hispanic ethnicity is associated with an increased risk for severe disease in children with COVID-19. Identifying underlying contributors to this disparity can lead to improved health care utilization and prevention strategies.

This is a retrospective cohort study of children 2–20 years of age with positive SARS-CoV-2 testing from March–October 2020. Univariable and multivariable logistic regression models were fitted to identify demographic, comorbid health conditions, and social vulnerabilities as predictors of severe COVID-19 (need for hospital admission or respiratory support).

We included 1572 children with COVID-19, of whom 45% identified as Hispanic. Compared to non-Hispanic children, patients who identified as Hispanic were more often obese (28% vs. 14%, p < 0.0001), preferred a non-English language (31% vs. 3%, p < 0.0001), and had Medicaid or no insurance (79% vs. 33%, p < 0.0001). In univariable analyses, children who identified as Hispanic were more likely to require hospital admission (OR 2.4, CI: 1.57–3.80) and respiratory support (OR 2.4, CI: 1.38–4.14). In multivariable analyses, hospital admission was associated with obesity (OR 1.9, CI: 1.15–3.08), non-English language (OR 2.4, CI: 1.35–4.23), and Medicaid insurance (OR 2.0, CI: 1.10–3.71), but ethnicity was not a significant predictor of severe disease.

The high rates of severe COVID-19 observed in Hispanic children early in the pandemic appeared to be secondary to underlying co-morbidities and social vulnerabilities that may have influenced access to care, such as language and insurance status. Pediatric providers and public health officials should tailor resource allocation to better target this underserved patient population.