Background
COVID-19 continues to pose a significant threat to public health. Widespread uptake of the multiple vaccines authorized by the U.S. Food and Drug Administration for use against COVID-19 represents the safest and most effective strategy for limiting the impact of the disease [
1]. However, with the Centers for Disease Control and Prevention (CDC) COVID Data Tracker reporting that ~20% of adults in the United States have not received a primary dose of a COVID-19 [
2], public hesitancy and refusal to get vaccinated remains a major challenge to realizing the full preventative health benefits of the authorized COVID-19 vaccines.
In order to effectively promote vaccine uptake, it is important to first understand which people are most and least inclined to be vaccinated and why. Over the course of the pandemic, research identifying important demographic (e.g., age, race, ethnicity, and education) [
3,
4] and psychological factors (e.g., COVID-19 risk perceptions [
3‐
6], belief in conspiracy theories [
7], political affiliation [
3,
5,
6,
8], exposure to misinformation [
9], and trust in scientists [
5,
10], and the government [
10,
11]) associated with public attitudes and intentions towards COVID-19 vaccines has accumulated at a rapid rate. This research has been of great value to policy makers and health communicators aiming to develop strategies and interventions to address concerns about COVID-19 vaccines and promote vaccine uptake.
However, although attitudes and intentions towards vaccination are often useful predictors of actual vaccine uptake [
12,
13] this relationship does not always hold true [
14‐
18]. For example, it is well documented that many people who intend to receive an influenza vaccine ultimately do not go on to receive one [
14,
16]. While evidence regarding the predictors of attitudes and intentions towards COVID-19 vaccination and characteristics associated with uptake has accumulated at a dramatic rate during the pandemic, there are fewer studies that have sought to predict actual uptake of COVID-19 vaccination. Among studies that have examined factors associated with COVID-19 vaccine uptake, most have tended to consider the influence of demographic factors such as age, gender, and socioeconomic status [
19‐
21]. Only a minority have also considered potential psychological influences, with those that do focussing on one or two specific factors such as vaccine attitudes [
22,
23], mistrust [
24], and risk perceptions [
23,
25]. As a result, there is a lack of data on the extent to which demographic and psychological factors, when considered together, predict actual uptake of COVID-19 vaccination. Furthermore, many cross-sectional studies of COVID-19 vaccine uptake may not capture changes in public behaviour across evolving periods of the pandemic.
Thus, data is therefore needed to identify attitudinal and sociodemographic factors that predict future vaccine uptake over time. The aim of the present study is to identify factors that predicted uptake of COVID-19 vaccination when vaccines first became available in January and March, 2021. In addition, we report the reasons given for not getting vaccinated by those who had not and did not intend to do so, following the rollout of the COVID-19 vaccines in the US in December, 2020.
We hypothesized that older age, living in a state with a greater proportion of people vaccinated, Veteran status, having a greater number of pre-existing health conditions, higher health literacy, higher numeracy,
1 and being non-Hispanic White, would be associated with having received at least one dose of a COVID-19 vaccine in both January and March, 2021. Based on existing research on psychological factors associated with COVID-19 vaccine attitudes and intentions, we also expected that greater worry about COVID-19, greater COVID-19 risk perceptions, greater confidence in vaccines, greater intentions to get a COVID-19 vaccine, greater trust in health care, greater belief in science, less belief in conspiracies, more liberal political views, and medical maximizing would be associated with COVID-19 vaccine uptake.
Discussion
The aim of the present study was to identify key predictors of, and objections to, COVID-19 vaccine uptake using a nonprobability online longitudinal survey of US Veterans and non-Veterans between December 2020 and March 2021. Building on previous work, we considered a range of demographic and psychological factors that may be associated with COVID-19 vaccine uptake. Findings from our sample, revealed that older age, higher numeracy, higher COVID-19 risk perceptions, and positive attitudes towards COVID-19 vaccines were important predictors of early vaccine uptake (by January 2021). As the rollout progressed, the influence of numeracy and risk perceptions remitted and we found that only older age, positive attitudes towards COVID-19 vaccines, and intentions to receive a COVID-19 vaccine were significant predictors of later vaccine uptake (by March 2021).
Consistent with prior research [
20,
21], older age was the strongest predictor of vaccine uptake for both timepoints, which reflects its emphasis as key criterion for early vaccine eligibility by the CDC [
34]. In contrast, the proportion of other people within the state who had been vaccinated, Veteran status, and the total number of pre-existing conditions were not associated with COVID-19 vaccine uptake at either time points. The combination of numeracy, risk perceptions, and attitudes towards COVID-19 vaccines as predictors of early vaccine uptake supports prior research demonstrating that assessment of the risks and benefits offered by vaccination as well as the threat of the disease that the vaccine protects against have a substantial influence on whether or not someone is likely to get vaccinated [
45,
46]. While clear communication about the risks and benefits associated with the vaccine and the threat posed by the disease is crucial at all times, these findings suggest that it may be particularly effective at encouraging uptake during the early stages of rollouts and for novel vaccines and diseases, given that numeracy and risk perceptions did not remain significant predictors of vaccine uptake later in the pandemic.
Our findings offer important evidence that attitudes and intentions towards COVID-19 predict uptake and provide validation for the many studies that have used these measures as a proxy for vaccination uptake [
3,
5‐
8,
10,
11]. In fact, of our respondents who were 65 years or older, 90% of those who reported that they intended to vaccinate had done so by the March 2021 survey. In addition, the present findings also build on prior research exploring characteristics associated with COVID-19 vaccination uptake, which has tended to overlook psychological and behavioral factors [
19‐
21]. Taken together, these findings reinforce the need to develop effective strategies for addressing people’s concerns and negative attitudes towards COVID-19 vaccines in order to increase uptake.
The findings from the present study may also contribute to informing health communication efforts aimed at those least likely to get a COVID-19 vaccine. Around 10% of the respondents in our study both had not been vaccinated at the time of the final survey in March and also indicated that they did not intend to do so in the future. These respondents tended to be younger, had negative views about the COVID-19 vaccines for adults, low trust in healthcare, and preferred to watch and wait before taking action in medical situations where there is clinical equipoise on whether action is necessary. In addition, the most important reasons given by these respondents for not getting a COVID-19 vaccine focused on safety concerns (particularly regarding side effects and the development process), beliefs that COVID-19 is not a serious threat, personal beliefs conflicting with vaccination and distrust of institutions involved with the vaccines.
Our findings are aligned with prior studies on the reasons given by people who are hesitant towards or refuse COVID-19 vaccines [
3,
47,
48], and offer an empirical basis for targeting public health messages to those who are least likely to vaccinate and tailoring messages to address their concerns. As these beliefs are often deeply held and traditional models of health communication have been largely ineffective at addressing them [
32,
49], we encourage health researchers and communicators to move beyond such traditional models of information provision and instead generate alternative strategies for addressing the concerns of those who are reluctant to get vaccinated. This is particularly important, given that the CDC COVID Data Tracker currently estimates that ~20% of adults in the United States are without a primary dose of a COVID-19 vaccine and 83% have not received an updated booster vaccine [
2].
One limitation of the study is that the findings rely on the accuracy and consistency of respondents’ self-reported data over the duration of the survey period. Although self-reports have been shown to be highly concordant with healthcare utilization and vaccine records [
50,
51], replication of these findings with a method for confirming respondents’ reported vaccine uptake would increase confidence in these findings.
Furthermore, our sample consisted of Veteran and non-Veteran respondents who were unique in being sufficiently motivated and able to complete three online surveys during the pandemic and therefore are not representative of the general U.S. population. The finding that Veteran status did not predict vaccine uptake at either time point was surprising given the efforts and widespread outreach of the U.S. Department of Veterans Affairs in supporting COVID-19 vaccine distribution [
52]. However, it is likely that the greater proportion of older adults in the Veteran sample compared to the non-Veteran sample may have limited our ability to observe a significant effect of Veteran status in the full model.
The unique makeup of our sample may also explain why the only early eligibility and demographic factors associated with vaccine uptake in this study were older age and numeracy. Our findings might also differ from prior research as our sample was overrepresented by respondents without many pre-existing health conditions (70% reported ≤ 1 pre-existing health condition), with high health literacy (94% of respondents reported high health literacy), and who identified as non-Hispanic White (78%). For instance, due to the high proportion respondents in our sample who identified as non-Hispanic White, we did not have sufficient power to explore differences across other specific racial and ethnic subgroups, which have been shown in prior studies to be associated with vaccine intentions and uptake [
11].
Conclusion
Despite these limitations, the findings from the present study offer important insights regarding the predictors of vaccine uptake during the early stages of the COVID-19 vaccine rollout in the US, which can help guide health communications and public outreach. In this study, we found that early uptake of COVID-19 vaccines (i.e., by January 2021) was associated with older age, greater numeracy skills, higher COVID-19 risk perceptions, and positive attitudes towards COVID-19 vaccines, while later vaccine uptake (i.e., by March 2021) was characterized by older age, positive attitudes towards COVID-19 vaccines, and intentions to receive the vaccine. Younger age, negative attitudes towards COVID-19 vaccines, low trust in healthcare, and medical minimizing, were significant predictors of being unvaccinated and not wanting to receive a COVID-19 vaccine, as of March 2021. These findings reinforce the need for developing effective strategies for promoting positive attitudes and intentions towards vaccines to promote uptake and highlight the importance of tailoring efforts to address the unique concerns of those who are least likely to get vaccinated. A major strength of our study is that we were able to cover the initial stages of the COVID-19 vaccine distribution. However, given the changes observed between January and March and the unique characteristics of our sample, further studies are needed to re-evaluate the key predictors of vaccine uptake as the rollout progresses and with wider representation, particularly as individuals become eligible for booster vaccines and considering the circulation of novel SARS-CoV-2 variants.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.