COVID-19 Vaccination Intent, Perceptions, and Reasons for Not Vaccinating Among Groups Prioritized for Early Vaccination

United States, September and December 2020

Kimberly H. Nguyen, DrPH; Anup Srivastav, PhD; Hilda Razzaghi, PhD; Walter Williams, MD; Megan C. Lindley, MPH; Cynthia Jorgensen, DrPH; Neetu Abad, PhD; James A. Singleton, PhD


Morbidity and Mortality Weekly Report. 2021;70(6):217-222. 

In This Article

Abstract and Introduction


As of February 8, 2021, 59.3 million doses of vaccines to prevent coronavirus disease 2019 (COVID-19) had been distributed in the United States, and 31.6 million persons had received at least 1 dose of the COVID-19 vaccine.[1] However, national polls conducted before vaccine distribution began suggested that many persons were hesitant to receive COVID-19 vaccination.[2] To examine perceptions toward COVID-19 vaccine and intentions to be vaccinated, in September and December 2020, CDC conducted household panel surveys among a representative sample of U.S. adults. From September to December, vaccination intent (defined as being absolutely certain or very likely to be vaccinated) increased overall (from 39.4% to 49.1%); the largest increase occurred among adults aged ≥65 years. If defined as being absolutely certain, very likely, or somewhat likely to be vaccinated, vaccination intent increased overall from September (61.9%) to December (68.0%). Vaccination nonintent (defined as not intending to receive a COVID-19 vaccination) decreased among all adults (from 38.1% to 32.1%) and among most sociodemographic groups. Younger adults, women, non-Hispanic Black (Black) persons, adults living in nonmetropolitan areas, and adults with lower educational attainment, with lower income, and without health insurance were most likely to report lack of intent to receive COVID-19 vaccine. Intent to receive COVID-19 vaccine increased among adults aged ≥65 years by 17.1 percentage points (from 49.1% to 66.2%), among essential workers by 8.8 points (from 37.1% to 45.9%), and among adults aged 18–64 years with underlying medical conditions by 5.3 points (from 36.5% to 41.8%). Although confidence in COVID-19 vaccines increased during September–December 2020 in the United States, additional efforts to tailor messages and implement strategies to further increase the public's confidence, overall and within specific subpopulations, are needed. Ensuring high and equitable vaccination coverage across all populations is important to prevent the spread of COVID-19 and mitigate the impact of the pandemic.

The Advisory Committee on Immunization Practices (ACIP) has issued interim recommendations for COVID-19 vaccine allocation, with initial limited supplies of vaccines recommended for health care personnel and residents of long-term care facilities (phase 1a); frontline essential workers and persons aged ≥75 years (phase 1b); and persons aged 65–74 years, persons aged 16–64 years at high risk for severe COVID-19 illness because of underlying medical conditions,* and other workers in essential and critical infrastructure sectors not included in phases 1a and 1b (phase 1c).[3,4] Vaccinating a large proportion of persons in the United States against COVID-19 is critical for preventing SARS-CoV-2–associated morbidity and mortality and helping bring an end to the global pandemic.

During September 3–October 1, CDC conducted a probability-based Internet panel survey (IPSOS KnowledgePanel)§ of a nationally representative sample of 3,541 U.S. adult panelists aged ≥18 years to assess intent to receive a COVID-19 vaccine and perceptions about the vaccine.[5] During December 18–20, CDC sponsored questions on two probability-based household panel omnibus surveys (IPSOS KnowledgePanel and NORC Amerispeak**) administered to 2,033 panelists (approximately 1,000 panelists each) to reassess COVID-19 vaccination intent and related perceptions.†† This activity was reviewed by CDC and was conducted consistent with applicable federal law and CDC policy.§§ The same questions about COVID-19 vaccine intentions, perceptions, and reasons for not receiving a COVID-19 vaccine were asked in the September and December surveys. However, most respondents were different for each survey; only 123 panelists (3.5%) completed both the September and December IPSOS survey. Intent was assessed by response to the following question: "If a vaccine against COVID-19 were available today at no cost, how likely would you be to get it?" Response options were "absolutely certain," "very likely," "somewhat likely," and "not likely." Respondents who answered "absolutely certain" or "very likely" to receive a COVID-19 vaccination were defined as intending to be vaccinated, and respondents who answered "not likely" were defined as not intending to be vaccinated. Vaccination intentions and related perceptions were stratified by the following three mutually exclusive groups representing the ACIP priorities for initial doses of COVID-19 vaccine after health care providers and long-term care residents: 1) essential workers,¶¶ 2) adults aged 18–64 years with underlying medical conditions, and 3) adults aged ≥65 years.*** Sample size for the December surveys was not large enough to stratify the analysis by age group (65–74 years versus ≥75 years) or essential worker subgroups (health care personnel, other frontline essential workers, and other non-frontline essential workers). Analyses were also conducted to provide estimates among all adults and among adults not included in the initial ACIP priority groups (aged 18–64 years with no underlying medical conditions and who were not essential workers). Responses to questions on intent, perceptions, and reasons for not getting vaccinated were examined by sociodemographic characteristics and priority groups for the September and December surveys. Because of similar sampling methods and characteristics of respondents, the averages of the estimates from the two December surveys were calculated, and the difference between the September survey and the average of the December surveys was determined using t-tests. All surveys were weighted to ensure representativeness of the U.S. population, and all analyses were conducted using SAS-callable SUDAAN (version 11.0; RTI International).

From September to December, the proportion of adults reporting intent to receive COVID-19 vaccine as absolutely certain or very likely increased significantly by 9.7 percentage points (from 39.4% to 49.1%), and the proportion reporting nonintent decreased by 6.0 percentage points (from 38.1% to 32.1%) (Table 1). Among priority groups, intent increased by 17.1 percentage points among adults aged ≥65 years (from 49.1% to 66.2%), by 8.8 percentage points among essential workers (from 37.1% to 45.9%), and by 5.3 percentage points among adults aged 18–64 years with underlying medical conditions (from 36.5% to 41.8%) (Supplementary Figure,

Vaccination nonintent differed by sociodemographic characteristics and decreased across most socioeconomic groups from September to December (Table 2). For example, nonintent decreased by 10.3 percentage points among adults aged 50–64 years and by 11.1 percentage points among adults aged ≥65 years. Although nonintent was higher among women, nonintent among both women and men decreased by 6.0 percentage points between September and December. Nonintent was highest among Black persons in September (56.1%) and December (46.5%) compared with other racial/ethnic groups, with the difference between months (−9.6) not statistically significant. Nonintent was higher among adults with lower educational attainment and lower income but decreased across most education and income categories: among adults with a high school diploma or less, nonintent decreased 7.9 percentage points, and in households with annual incomes of $35,000–$49,999, nonintent decreased by 10.8 percentage points. Vaccination nonintent also decreased in metropolitan statistical areas††† by 6.7 percentage points and among adults in all regions of the United States, except the Northeast, including decreases of 8.3 percentage points in the South, 6.8 in the Midwest, and 6.8 in the West. In December, nonintent was highest among persons without health insurance (44.5%), compared with those who had private health insurance (30.7%) and public health insurance (29.6%), and was similar in September and December.

Among adults in the December surveys who did not intend to get vaccinated, the main reasons most frequently cited were concerns about side effects and safety of the COVID-19 vaccine (29.8%), planning to wait to see if the vaccine is safe and consider receiving it later (14.5%), lack of trust in the government (12.5%), and concern that COVID-19 vaccines were developed too quickly (10.4%) (Table 3). A larger percentage of the December survey participants than September participants reported safety concerns as a main reason (29.8% versus 23.4%), and a smaller percentage reported concern that vaccines were developed too quickly (10.4% versus 21.6%).

*Persons with underlying medical conditions were defined as those who reported having any of the following conditions: cancer; chronic kidney disease; chronic obstructive pulmonary disease (COPD); heart conditions (e.g., heart failure, coronary artery disease, or cardiomyopathies); immunocompromised state (weakened immune system) from solid organ transplant; obesity; pregnancy; sickle cell disease; smoking; and type 2 diabetes mellitus. Respondents aged 18–64 years reporting diagnosis of one or more of these conditions were classified as high-risk in the analyses. This list of underlying medical conditions does not include Down syndrome, which was added to the list on December 23, 2020. A complete list of underlying medical conditions is available at
Other essential workers include those who conduct a range of operations and services that are essential to continued critical infrastructure viability, including staffing operations centers, maintaining and repairing critical infrastructure, operating call centers, working construction, and performing operational functions, among others. Also included are workers who support crucial supply chains and enable functions for critical infrastructure. The industries they support represent, but are not limited to, medical and health care, telecommunications, information technology systems, defense, food and agriculture, transportation and logistics, energy, water and wastewater, and law enforcement.
††The panels from the September and December surveys use an address-based sampling methodology that covers nearly all households in the United States regardless of their phone or Internet status, with a cooperation rate (proportion of all cases interviewed among all eligible units ever contacted) of 69.7% (September IPSOS survey), 38.0% (December IPSOS survey), and 22.8% (December NORC survey). Surveys were fielded in English and Spanish, and non-Hispanic Black and non-Hispanic other race panel members were oversampled to ensure adequate sample size for subgroup analyses by respondent's race/ethnicity.
§§45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.
¶¶Essential workers were defined as those who responded "yes" to the following question: "In your work or volunteer activities, are you classified as an essential worker?"
***Mutually exclusive groups were categorized in the following order: essential workers, adults aged ≥65 years, and adults aged 18–64 years with an underlying medical condition. Anyone who self-identified as an essential worker was categorized as an essential worker, regardless of age. Next, anyone aged ≥65 years was categorized as adult aged ≥65 years. Finally, anyone aged 18–64 years with an underlying medical condition was categorized as adult aged 18–64 years with an underlying medical condition. All others were categorized as adults aged 18–64 years who were not essential workers and had no underlying medical conditions.
†††Metropolitan statistical area (MSA) status was determined by census block group using the panelist's address. For a small number of panelists for whom the address was not available, ZIP code was used to determine MSA status.