JAMA Health Forum | 2021
Disparities in SARS-CoV-2 Vaccination-to-Infection Risk During the COVID-19 Pandemic in Massachusetts
Abstract
We aggregated community SARS-CoV-2 testing and vaccination data from the Massachusetts Department of Public Health and the Boston Public Health Commission from January 29, 2020, through June 24, 2021. Data were available for 293 distinct communities (278 cities and towns, and 15 Boston neighborhoods) with a cumulative population of 6 755 622 (98.6%). We adhered to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines. This study involved the use of public anonymized data and was therefore designated as exempt by the Mass General Brigham Institutional Review Board. We considered each community’s cumulative incidence of confirmed SARS-CoV-2 infections to be the best available indicator of future infection risk. We used 2 approaches to assess vaccination equity: VIR ratio and Lorenz curves. The VIR ratio was calculated for each community as the quotient of the number of fully vaccinated individuals divided by the cumulative number of confirmed SARSCoV-2 infections. Communities with VIR ratios below the statewide mean have lower vaccination coverage relative to their infection risk. Lorenz curves, which assess equity in resource distribution,3 were used to describe vaccination relative to COVID-19 burden and calculate summaries of inequity (Gini index) and magnitude of vaccine reallocation required to achieve equity (Hoover index). We used population estimates from the Massachusetts Department of Public Health and the American Community Survey to determine community age, race, and ethnic composition. Socioeconomic vulnerability was estimated using the Socioeconomic Status domain of the Social Vulnerability Index (census tract ranks aggregated by community). We fit a negative binomial model using robust sandwich estimators to assess associations between community VIR ratios and a priori–selected predictors: proportion 65 years or older (an early vaccination eligibility criterion), proportion identified as Black and/or Latinx individuals (<20% or 20%), quartile of socioeconomic vulnerability, and community size (<7500 or 7500 residents). Analyses were conducted in R, version 4.0.5 (R Foundation), and a 2-tailed P < .05 was considered statistically significant.