Inequities in COVID-19 Vaccination Rates in the 9 Largest US Cities

The equitable receipt of COVID-19 vaccinations is a national priority.¹ Most jurisdictions in the United States report limited data on vaccinated people, impeding assessment of vaccination equity.² We used neighborhood-level data to estimate inequities in COVID-19 vaccination rates.

The following 9 largest US cities (with surrounding counties except for Chicago, Illinois), representing 40.8 million people, reported neighborhood-level vaccination rates from the beginning of vaccinations through April 13, 2021: New York, New York; Los Angeles, California; Chicago; Houston, Texas; Phoenix, Arizona; Philadelphia, Pennsylvania; San Antonio, Texas; San Diego, California; and Dallas, Texas. In this cross-sectional study, we obtained data on COVID-19 vaccination and death rates for these cites from health authority websites and sociodemographic information from the American Community Survey.³ Race and ethnicity were self-identified by the American Community Survey respondents. This study used only deidentified, publicly available data and was therefore exempt from institutional review board approval and informed consent in accordance with the Common Rule and Columbia University policy. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.

We defined neighborhoods using zip codes. In Los Angeles we used communities, a similar level of aggregation. Within each city, we divided neighborhoods into quartiles according to vaccination rate (adults with at least 1 dose) and calculated the mean sociodemographic characteristics and COVID-19 death rates in each quartile. To measure vaccination rates relative to historical disease burden (defined as the cumulative death rate from COVID-19), we tested the association between COVID-19 death rates and vaccination rates within cities using linear regression and conducted a concentration analysis assessing the share of vaccinations administered in neighborhoods with the highest death rates.⁴

We considered P < .05 from 2-sided tests to be statistically significant. Analyses were performed using Stata/MP, version 16.0 (StataCorp, LLC).

We analyzed 1127 neighborhoods with a mean (SD) COVID-19 vaccination rate of 42.3% (13.4 percentage points). Neighborhoods in the lowest quartile had less than half the vaccination rate of those in the highest quartile (27.6% vs 59.7%; Table). Neighborhoods with high vaccination rates had a greater share of White and Asian people and a lower share of Black and Hispanic or Latino people. These neighborhoods also had higher mean incomes, lower poverty rates, and higher 4-year college completion rates. Employment in health care differed little across quartiles, but in neighborhoods with high vaccination rates, these workers were more likely to be health care practitioners or technologists and less likely to be in support occupations.

Historical COVID-19 death rates (from the first COVID-19 deaths through April 13, 2021) were lowest in neighborhoods with the highest vaccination rates, even though these neighborhoods had more older adults. A 10-percentage-point increase in the vaccination rate was associated with 25 fewer historical COVID-19 deaths per 100 000 population (P < .001, Figure). Of the 863 neighborhoods with death data, the 209 with the highest death rates accounted for half of all historical COVID-19 deaths but 26% of all vaccinations.

In the 9 largest US cities, COVID-19 vaccination rates were disproportionately high in communities with lower burdens of this disease. This study builds on reports of inequitable COVID-19 vaccination rates by race and ethnicity that were frequently based on incomplete demographic data from states² as well as county-level analyses of early vaccination efforts that found comparatively small differences in vaccination rates by county social vulnerability index.⁵ Using more granular neighborhood-level data, we documented substantial inequities in vaccination rates in the first 5 months of vaccine distribution.

Limitations of this study include potential inaccuracies in vaccination and death reporting, the inability to distinguish the role of supply- and demand-side factors, and limits on the generalizability of the results beyond the included cities. We were also unable to assess the role of vaccine supply and eligibility policies in creating or mitigating disparities. However, given that all US adults have been eligible to receive the vaccine since April 2021 and there is current excess supply, other policies are now more actionable targets to improve vaccination equity.

Inequities in vaccination rates across neighborhoods likely reflect several root causes, including systematic underinvestment in public health in segregated communities, unequal access to health care information and services, and medical racism that drives legitimate mistrust among members of marginalized groups.⁶ The findings of the present study emphasize the opportunity and need for cities to address vaccination inequities in marginalized communities.

Accepted for Publication: July 8, 2021.

Published: September 3, 2021. doi:10.1001/jamahealthforum.2021.2415

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Sacarny A et al. JAMA Health Forum.

Corresponding Author: Adam Sacarny, PhD, Columbia University Mailman School of Public Health, 722 W 168 St, 4th Floor, New York, NY 10032 (ajs2102@columbia.edu).

Author Contributions: Drs Sacarny and Daw had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Sacarny.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: All authors.

Administrative, technical, or material support: All authors.

Supervision: Sacarny.

Conflict of Interest Disclosures: None reported.