Since the rollout of COVID-19 vaccines began, one issue that has gained attention has been racial equity in COVID-19 vaccination rates. Ensuring fairness in COVID-19 vaccinations is important given that COVID-19 has disproportionately affected people of color and can exacerbate underlying health disparities. Data is essential to identify disparities in COVID-19 vaccination rates and guide resources and efforts to address them. However, there are gaps in the COVID-19 vaccination data reported by the federal government by race / ethnicity from the Centers for Disease Control and Prevention (CDC). To help fill these federal data gaps, KFF and others have conducted ongoing analysis of state-reported immunization data by race and ethnicity and regular COVID-19 vaccine surveillance surveys of adults. . This data provided additional information on COVID-19 vaccination patterns by race / ethnicity, but is also subject to limitations. This brief provides an overview of these data sources, discusses their limitations, and explains why their conclusions may vary.
Vaccination rate according to data sources
Federal and state administrative data and Vaccine Monitor surveys all show that blacks and Hispanics have been less likely to receive a COVID-19 vaccine compared to their white counterparts since the vaccination rollout began, but that these disparities are growing. are reduced over time. However, they vary in the conclusions of the magnitude of this shrinkage (Figure 1 and Table 1):
- Federal CDC data shows that between late April and late September 2021, the percentage point gap between white and black rates for the general population decreased by 2 percentage points (from 8 to 6 percentage points). percentage) while the gap between white and Hispanic rates decreased by 9 percentage points (from 8 to -1 percentage points).
- State-reported data reveals that the gap between the white and black rates for the total population narrowed by 6 percentage points during this period (from 14 to 8 percentage points), while the difference between the rates of Whites and Hispanics decreased by 9 percentage points (from 13 to 4 percentage points).
- Data from the Vaccine Monitor survey shows the same trend with the difference between the rates of white and black adults decreasing by 8 percentage points (from 9 to 1 percentage points) and the gap between white and Hispanic adults decreasing by 15 percentage points (from 13 to -2 percentage points).
As of September 2021, federal data from the CDC shows similar vaccination rates between Hispanics and whites, with lower rates persisting for blacks, and the highest rate for American Indians and Native Americans. Alaska and Asians. Analysis of state data reveals that blacks and Hispanics are less likely than whites to be vaccinated, but with a narrower gap for Hispanics. Data from the Vaccine Monitor survey shows that the rate gaps for black and Hispanic adults compared to white adults have narrowed, with no statistically significant difference in vaccination rates between these groups. A Pew Research Center survey in August yielded similar results.
This variation in results reflects differences in what the data sources measure. Vaccine Monitor survey vaccination rates are based on adults, while rates based on federal and state administrative data refer to the total population (including children under 12 who are not currently eligible for vaccination) . Inclusion of children in vaccination rates may lead to greater disparities due to racial differences in vaccination rates among adolescents eligible for vaccines (ages 12 to 17) and due to greater racial diversity children versus adults. In addition, both survey data and administrative data are subject to different sources of measurement error, as we will see below.
Federal COVID-19 Vaccination Data by Race / Ethnicity
The CDC reports the distribution of COVID-19 vaccines and the percentage of the total population that received a COVID-19 vaccine by race / ethnicity nationwide. However, as of September 27, 2021, race / ethnicity information was missing for more than 40% of people who received at least one dose. Additionally, the data does not represent all states and jurisdictions, as not all states and territories report vaccine population demographics to CDC. Given these data gaps, the CDC says the data is not generalizable to the entire population of individuals vaccinated against COVID-19. The CDC does not report state-level data on COVID-19 vaccinations by race / ethnicity. Additionally, although the CDC reports vaccinations by race / ethnicity and age separately, it does not publish data to analyze vaccinations by race / ethnicity. and age. As such, the data cannot be used to examine whether there are greater racial disparities in vaccination rates among certain age groups, such as adolescents or young adults.
Indicate COVID-19 vaccination data by race / ethnicity
With no CDC reporting state-level COVID-19 vaccination data by race / ethnicity, KFF has conducted ongoing analysis of data reported directly by states. As of September 20, 2021, 45 states, including Washington, DC, were reporting data on people who had received at least one COVID-19 vaccine by race / ethnicity and KFF was able to calculate total vaccination rates by race / ethnicity out of 43 of these states. (Two states were excluded from the total due to differences in how they report their data). In general, this data is more comprehensive than data reported by the CDC, with lower proportions of vaccinations of unknown or missing race / ethnicity in most states. However, they also have shortcomings, limitations and inconsistencies. As with federal data, it does not include data from all states and jurisdictions, and some states have relatively high shares of vaccinations with an unknown race / ethnicity. For example, in Alabama, 37% of vaccinations were of an unknown race as of September 20, 2021. Additionally, states vary in their racial / ethnic classifications used to report data, including how they classify who self-identifies. as more than one race. Some state-reported data does not include vaccinations administered through federal programs, including the Indian Health Service or the Long-Term Care Partnership Program.
COVID-19 vaccine monitoring and other investigative data
Since December 2020, KFF has been conducting continuous and nationally representative surveys of American adults through the COVID-19 Vaccine Monitor. While these surveys have a broader purpose of measuring confidence in vaccination, information needs, messengers and messages of confidence, we have also used them to track the share of adults who report having been vaccinated against COVID- 19 over time. These surveys rely on probability sampling methods and researchers are taking additional steps to ensure the inclusion of populations that are often missed in surveys (including interviews in English and Spanish and oversampling of prepaid cell phones that are commonly used by low-income adults). Each survey also includes additional interviews with black and Hispanic adults – using weighting to adjust survey respondents to match the distribution of adults in the United States – so that you can have greater statistical confidence when reporting on these groups (see full methodology for details).
Like all surveys, Vaccine Monitor surveys are subject to a margin of sampling error around each estimate. For the September survey, the margin of sampling error was plus or minus 3 percentage points for the full sample, 4 percentage points for white adults, and plus or minus 7 percentage points for black adults. and Hispanics. In addition, surveys may have other sources of error, including non-response error (certain types of people choose not to participate in the survey or refuse to answer the question of whether they have been completed. vaccinated), measurement error (respondents do not understand the question that was asked) and social desirability bias (respondents give answers they think the interviewer wants to hear). Despite these potential sources of error, the Vaccine Monitor surveys have followed CDC’s estimates of the share of adults vaccinated overall over time very closely, and a recent Pew Research Center analysis of 98 different public surveys. conducted by 19 different survey organizations between December 2020 and June 2021 (including the Vaccine Monitor surveys) found that survey estimates of the adult vaccination rate were around 2.8 percentage points, on average, of the rate calculated by CDC.
Data is essential for identifying and addressing disparities in health and healthcare. During the COVID-19 pandemic, gaps in the data available by race / ethnicity have limited efforts to understand and address disparities. The availability and quality of data have improved during the pandemic, but gaps and limitations remain. To help fill gaps in federally reported data on COVID-19 vaccinations by race / ethnicity, KFF and others have conducted an ongoing analysis of state-reported data on COVID-19 vaccinations and Regular COVID-19 vaccine surveillance surveys in adults, which have provided an increased understanding of COVID-19 vaccination patterns by race / ethnicity. This, in turn, has helped direct resources and efforts to address racial disparities in immunization rates. While federal, state and survey data all show a reduction in racial disparities in COVID-19 vaccination rates over time, it varies in the magnitude of that shrinkage, with some surveys showing the gaps have closed. , while administrative data indicate some remaining differences. This variation in results reflects both the differences and the limitations between the data sets. Going forward, continued efforts to increase the availability of comprehensive, high-quality data will be essential to identify and address disparities for COVID-19 and in health and healthcare more generally. Additionally, it will be important to continue to prioritize equity as immunization efforts continue, people become eligible for booster vaccines, and eligibility expands to children, especially given of the significant racial diversity of children.