Study population and setting
From September 3 to October 1, 2020 (baseline) and December 1 to 3, 2020 (endline), adults aged 18 years and older in the United States were recruited to complete an online survey assessing COVID-19 vaccination intentions, measured using a 4-point scale (absolutely certain, very likely, somewhat likely, not likely). Vaccination intentions, including reasons for vaccine hesitancy (among those who self-reported being “not likely” to receive COVID-19 vaccine), were compared over time and across socio-demographic strata.
Summary of Main Findings
From baseline (N = 3,541) to endline (N = 2,003), vaccination intention increased significantly (39.4% to 49.1%), while vaccine hesitancy declined (38.1% vs. 32.1%). Reasons for vaccine hesitancy included concerns over vaccine side effects and safety (29.8%), mistrust in the government (12.5%), and concern about the pace at which COVID-19 vaccines were developed (10.4%). While vaccine hesitancy declined across socio-demographic strata over time, by endline, participants who were Black (46.5%), lacked health insurance (44.5%), lived in nonmetropolitan areas (39.6%), had low (<$35,000) annual household incomes (38.3%), and had secondary or less education (39.1%) reported significantly higher rates of COVID-19 vaccine hesitancy. Among priority groups for U.S. vaccine rollout, vaccine hesitancy declined over time but remained moderate among adults ages 65 and older (18.7%), essential workers (35.4%), and adults with underlying medical conditions (38.3%) by endline.
The study measured changes in COVID-19 vaccination intentions over time using nationally representative online panel surveys, which allowed investigators to examine longitudinal trends in COVID-19 vaccine acceptance/hesitancy, as well as reasons for vaccine hesitancy.
Response options for questions measuring vaccination intentions were collapsed in the analysis, which introduces potential for misclassifying participants’ vaccine intentions and makes estimates incomparable to results from other studies using more nuanced measures of vaccine acceptance/hesitancy. Because data are derived from two separate panel surveys, observations were not available from the same participants over time; as such, longitudinal estimates are calculated at an ecological scale and may not reflect true changes in individuals’ vaccine intentions over time. Additionally, only univariate associations between vaccine hesitancy and socio-demographic factors were reported in the study; these associations may have been confounded by other unmeasured factors, including behavioral determinants (i.e., perceived susceptibility to COVID-19, norms around vaccination). Lastly, adults who were eligible to participate in the online surveys may be substantially different from those who were excluded, which limits the inferences that can be made about the target population for these surveys (i.e., the US general population).
This is among the first studies to report changes in COVID-19 vaccination intentions among adults in the United States.
This review was posted on: 26 February 2021