Analysis of School-Level Vaccination Rates by Race, Ethnicity, and Geography in New York City

This cross-sectional study of New York City school data examines differences in COVID-19 vaccination rates by race, ethnicity, and borough.


Introduction
Vaccinations for COVID-19 have proven safe and effective for preventing serious illness for most children. 1,2 Race, ethnicity, and geography are important correlates of vaccination rates for adults, while data have proven limited in understanding whether this is true for children. 3 In this crosssectional study, we took advantage of newly released public school-level vaccination rates in New York City (NYC), where geographically distributed schools with varying racial and ethnic concentration can shed important light on differences by race, ethnicity, and geography for COVID-19 vaccination rates for children.

NYC recently posted school-level vaccination data gathered from the NYC Department of Health and
Mental Hygiene's centralized Citywide Immunization Registry. 4 These data included school-level percentages of students fully (not including boosters) or partially vaccinated for children ages 5 years and older. Using a consistent school identifier, we matched the March 9, 2022, data release to school-level data from the 2020-2021 School Report card. 5 To correspond with the differential timing of vaccination approval for children older vs younger than age 12 years, we categorized schools into 3 groups: elementary (serving grades pre-kindergarten to 5, younger than age 12 years), middle-high (grades 6 to 12, likely older than 12 years), and other grades (eg, schools with grades kindergarten to 8, kindergarten to 12).
We created a set of mutually exclusive and exhaustive categorical variables identifying the majority race or ethnicity of students in each school (Hispanic, Black, White, Asian, or no majority race). We then used ordinary least squares (weighted by share of total enrollment) to determine the regression-adjusted association between school-level vaccination and race and ethnicity, both overall and in a model with race and ethnicity by borough interactions. We also controlled for schoollevel percentage of students that were female, economically disadvantaged, experiencing homelessness, or in foster care. We used the margins command in Stata version 17 (StataCorp) to show the regression-adjusted vaccination percentage. This research was reviewed by the NYU Langone institutional review board and was determined to meet the criteria for exemption 4 under 45 CFR 46. This study followed the STROBE reporting guideline.
Differences in race and ethnicity by borough were also present in the model with a race and ethnicity by borough interaction (Figure, panel B). While schools with a majority of Asian students consistently had the highest vaccination percentage regardless of borough, other differences by

Discussion
We found key differences in vaccination rates by race and ethnicity of student population among schools in NYC, with schools with a majority of Asian students and schools in Manhattan having the highest vaccination rates. Some of these differences are similar to adult data (ie, Black vs White) and the aggregated unadjusted city-level data 6 on children. Differences also emerged by geography, both Vaccination rates refer to the percentages of fully vaccinated students and are projected from the regression analysis. Error bars indicate 95% CIs.