Blog & News
COVID-19 Vaccination Rates: State-level and Subpopulation Evidence from the Household Pulse SurveyJuly 19, 2021:
Update 6: June 9 to June 21
The COVID-19 vaccines promise to help protect individual Americans against infection and eventually provide population-level herd immunity. After several months of rolling out the various one and two-shot COVID-19 vaccines, which have included hiccups from the Johnson & Johnson vaccine, vaccination rates continue to increase at a slow but steady pace. Although the country fell short of meeting the current administration’s goal of vaccinating 70% of the adult population by July 4 (as measured by official administrative data reported to the Centers for Disease Control [CDC]), several states have achieved this goal.
Over the past several months, all states have increased COVID-19 vaccine rollout by expanding vaccine access to the general adult population and children over 12. However, there are still concerns on prioritization decisions and the existing mechanisms of the vaccine rollout—in addition to evidence that lower-income individuals, people of color, and individuals without strong connections to the health care system are less likely to get vaccinated—which have created challenges in equitable distribution of the COVID-19 vaccine and could worsen existing pandemic-related health inequities.
The available data have not assuaged these concerns, and show patterns of lower vaccination rates among people with lower levels of education, no health insurance coverage, and marginalized racial and ethnic groups. The U.S. Census Bureau recently released updated data on take-up of COVID-19 vaccines from its Household Pulse Survey (HPS), collected June 9-21, 2021. The HPS is an ongoing, biweekly tracking survey designed to measure impacts of the COVID-19 pandemic. These data provide an updated snapshot of COVID-19 vaccination rates and are the only data source to do so at the state level by subpopulation.
This blog post presents top-level findings from these new data, focusing on rates of vaccination (one or more doses) among U.S. adults (age 18 and older) living in households and comparing to results from the last half of March, the most recent time period of comparison from this ongoing blog series.
These data represent the latest release from Phase 3.1 of the HPS, which has a biweekly data collection and dissemination approach. The Census Bureau has indicated that it plans to continue administering the survey through December 2021.
Nationally, over three-fourths of adults had received a vaccination, but this varied by state
According to the new HPS data, 79.1% of U.S. adults had received one or more COVID-19 vaccinations towards the end of June1, though this varied by state from a low of 63.9% in Mississippi to a high of 91.0% in Massachusetts. At least four in five adults had received a vaccine in 22 states and the District of Columbia.
Vaccination rates increased considerably across all states; states with lower rates catching up
Nationally, adult vaccination rates substantially increased from the last half of March, rising from 47.0% during March 17-29, 2021, to 79.1% during June 9-21, 2021. Many states experienced large increases in their vaccination rates. The size of these increases varied from an 11.2 percentage-point (PP) increase in North Dakota to a 46.1 PP increase in Massachusetts. Five states saw increases of 40.0 PP or larger: Maryland, Massachusetts, Oregon, Rhode Island, and Vermont.
Disparities in vaccination rates improved, but slowly and unevenly
Although there have been ongoing strategies to achieve health equity in COVID-19 vaccine rollouts, vaccination rates continued to vary to a great degree by demographic and socioeconomic factors. Gaps in vaccination compared to the national average narrowed slightly for most groups, though some saw larger improvements. As with previous iterations in this blog series, vaccination rates were lower for certain subpopulations such as Black adults, adults who identified as “Some other race/Multiple races,” adults without a high school education, and adults without health insurance coverage. More resources, attention, or new strategies may be needed to close the gaps for these hardest-to-reach groups.
Asian and White adults continued to have above-average vaccination rates at 94.6% and 79.5%, respectively. Rates among Black adults (71.9%) and adults identifying with “Two or more” (Multiple) or “Some other” race (68.5%) continued to be below the national average. Rates among Hispanic/Latino adults improved, and are now above the national average at 79.6%. Hispanic/Latino adults also saw the largest improvement relative to the national average, going from almost 20 percent below the national average in late March (37.9% vs. 47.0%) to nearly one percent above the national average in June (79.6% vs. 79.1%).
Disparities by education level remained, with adults holding a bachelor’s degree or higher continuing to have the highest vaccination rate at 90.8%, and adults without a high school diploma having the lowest vaccination rate at 68.6%. Despite having the lowest vaccination rate, adults without a high school diploma had the largest relative improvement to the national average, going from 38 percent below the national average in March (29.2% vs. 47.0%) to 13 percent below the national average in June (68.6% vs. 79.1%). Vaccination rates among adults with a high school degree or equivalent and adults with some college or an associate’s degree also improved somewhat relative to the national average.
Adults with health insurance coverage had an above-average vaccination rate at 82.4%, while uninsured adults had a below-average vaccination rate at 61.4%. Regardless of having a rate substantially below the average, the rate among those not covered by any type of health insurance had a notable improvement relative to the national average, going from 50 percent below the national average in March (23.7% vs. 47.0%) to 22 percent below the national average in June (61.4% vs. 79.1%).
Notes about the Household Pulse Survey Data
The estimated rates presented in this post were pulled from the HPS COVID-19 Vaccination Tracker published by the Census Bureau. Though these counts are accompanied by standard errors, standard errors are not able to be accurately calculated for rate estimates. Therefore, we are not able to determine if the differences we found in our analysis are statistically significant or if the estimates themselves are statistically reliable. Estimates and differences for subpopulations at the state level should be assumed to have large confidence intervals around them and caution should be taken when drawing strong conclusions from this analysis. However, the fact that these indications of COVID-19 inequities mirror patterns of other vaccinations inequities demonstrate reason for concern.
Though produced by the U.S. Census Bureau, the HPS is considered an “experimental” survey and does not necessarily meet the Census Bureau’s high standards for data quality and statistical reliability. For example, the survey has relatively low response rates (6.4% for June 9-21), and sampled individuals are contacted via email and text message, asking them to complete an internet-based survey. These issues in particular could be potential sources of bias, but come with the tradeoffs of increased speed and flexibility in data collection as well as lower costs. A future post will investigate differences between COVID-19 vaccination rates estimated from survey data (such as the HPS) and administrative sources. The estimates presented this post are based on responses from 68,067 adults. More information about the data and methods for the Household Pulse Survey can be found in a previous SHADAC blog post.
Previous Blogs in the Series
COVID-19 Vaccination Rates: State-level and Subpopulation Analysis of the Household Pulse Survey (Update 5: March 17 to March 29) (SHADAC Blog)
COVID-19 Vaccination Rates: State-level and Subpopulation Analysis of the Household Pulse Survey (Update 4: March 3 to March 15) (SHADAC Blog)
COVID-19 Vaccination Rates: State-level and Subpopulation Evidence from the Household Pulse Survey (Update 3: Feb 17 to March 1) (SHADAC Blog)
COVID-19 Vaccination Rates: State-level and Subpopulation Evidence from the Household Pulse Survey (Update 2: Feb 3 to Feb 15) (SHADAC Blog)
COVID-19 Vaccination Rates: State-level and Subpopulation Evidence from the Household Pulse Survey (Update: Jan 10 to Feb 1) (SHADAC Blog)
COVID-19 Vaccination Rates: New State-level and Subpopulation Evidence from the Household Pulse Survey (Jan 6 to Jan 18) (SHADAC Blog)
Vaccine Hesitancy Decreased During the First Three Months of the Year: New Evidence from the Household Pulse Survey (SHADAC Blog)
State-level Flu Vaccination Rates among Key Population Subgroups (50-state profiles) (SHADAC Infographics)
50-State Infographics: A State-level Look at Flu Vaccination Rates among Key Population Subgroups (SHADAC Blog)
Anticipating COVID-19 Vaccination Challenges through Flu Vaccination Patterns (SHADAC Brief)
New Brief Examines Flu Vaccine Patterns as a Proxy for COVID – Anticipating and Addressing Coronavirus Vaccination Campaign Challenges at the National and State Level (SHADAC Blog)
Ensuring Equity: State Strategies for Monitoring COVID-19 Vaccination Rates by Race and Other Priority Populations (Expert Perspective for State Health & Value Strategies)
SHADAC Webinar - Anticipating COVID-19 Vaccination Challenges through Flu Vaccination Patterns (February 4th) (SHADAC Webinar)
1 Note that it is not unusual for there to be appreciable differences between survey-based estimates and those derived from administrative data, as there are here between the vaccination rates observed in the HPS and those seen in CDC administrative vaccination data. There could be several reasons for these differences, including differences in the population universe (i.e., household-residing adults vs. total adult population), differences in the measured time period, the inaccuracies between self-reported vs. administratively collected data, and differences in the representativeness of survey vs. administrative data. Although administrative data are often thought to be more accurate than survey-based estimates, survey data such as the HPS have the advantage of providing more granular detail about the demographic and socioeconomic characteristics of populations of interest that are often unavailable or incomplete in administrative data.