Abstract
Introduction
COVID-19 vaccination rates remain suboptimal in the United States. Clinicians and policymakers need to better understand how likely vaccine-hesitant individuals are to ultimately accept vaccination and what is associated with such changes. This study’s aims were to 1) describe changes between vaccine intentions and actual uptake from June 2021 through February 2022, and 2) identify modifiable factors associated with vaccine uptake among those with initial hesitancy.
Methods
This cohort study included a stratified random sample of adults aged 65 years and older in an integrated health care system. The survey, conducted June through August 2021, elicited intent and perceptions regarding COVID-19 vaccination. Subsequent vaccine uptake through February 2022 was analyzed using electronic health records.
Results
Of 1195 individuals surveyed, 66% responded; 213 reported not yet having received a COVID-19 vaccine and were further analyzed. At baseline, most individuals said they would definitely not (42%) or probably not (5%) get the COVID-19 vaccine or were not sure (26%). During follow-up, 61 individuals (29%) were vaccinated, including 19% of those who initially said they would definitely not be vaccinated. Among vaccine-hesitant individuals, the rate of vaccination was highest for those who initially considered COVID-19 less dangerous than the vaccine (46%) or named short-term side effects (36%) as their most important concern.
Conclusions
COVID-19 vaccine intent among older adults was malleable during the pandemic’s second year, even among those who initially said they would definitely not be vaccinated. Vaccine uptake could be enhanced by increasing awareness of COVID-19 risks and by addressing vaccine side effects.
Introduction
COVID-19 vaccination is a cornerstone of global efforts against the disease, but vaccination rates in the United States remain suboptimal. 1 Vaccination is particularly important for persons aged 65 years and older, who account for 76% of deaths from COVID-19. 2 Black and Latino adults are also at elevated risk of severe disease. 3
To design effective interventions to enhance vaccine uptake among older adults, clinicians and policymakers need to better understand what may lead those who are vaccine hesitant to change their minds. Many cross-sectional surveys of COVID-19 vaccine intent have been reported. 4,5 However, very few studies have followed vaccine-hesitant individuals over time to look at their subsequent vaccine uptake. Limited information exists to describe how often COVID-19 vaccine-hesitant adults ultimately accept vaccination or what factors are associated with acceptance after initial hesitation. 6,7
To address this gap in knowledge, this study observed a diverse population of older adults who had not received a COVID-19 vaccine by early summer 2021, when it had become widely available in the United States. The study aims were to 1) describe individual-level changes between stated vaccine intentions in the summer of 2021 and actual uptake over several months through the beginning of 2022, and 2) identify modifiable factors associated with vaccine uptake among those with initial vaccine hesitancy.
Methods
Design, setting, and population
This cohort study included older adult members of Kaiser Permanente Northern California, an integrated health care system with 21 hospitals and 262 medical offices serving a racially, ethnically, and socioeconomically diverse population. Baseline information on COVID-19 vaccine intent and perceptions was collected via survey in June through August 2021. Vaccine uptake during follow-up, defined as the time the survey was completed through February 2022, was analyzed based on electronic health records (EHRs) as of February 6, 2022.
The baseline survey was fielded to a stratified random sample of 1195 adults aged 65 years and older whose records indicated they had not received at least one vaccine dose by May 14, 2021. The sample was nested within a randomized controlled trial of outreach to promote COVID-19 vaccination. 8 Participants were randomized to receive culturally tailored primary care physician (PCP) outreach, standard PCP outreach, or usual care. The sample was stratified by 4 race/ethnicity–language groups (Black; Latino–English speaking; Latino–Spanish speaking; and Asian, White, other, or unknown) and 4 Kaiser Permanente Northern California service areas (California Central Valley, Fresno, South Sacramento, and San Jose).
The Research Determination Committee for the Kaiser Permanente Northern California Region determined that this project did not meet the regulatory definition of research involving human subjects per 45 CFR 46.102(d).
Survey design and questions
A 10-question survey was designed using the Health Belief Model and Theory of Planned Behavior and Reasoned Action as frameworks. 9,10 An initial screening question evaluated whether the patient had already received a COVID vaccine or scheduled a vaccine appointment. Those who had not were considered eligible to complete the remaining questions.
Respondents were considered vaccine hesitant if they stated that they would “probably not get the vaccine,” would “definitely not get the vaccine,” or were “not sure.” This categorization aligns with other definitions of vaccine hesitancy, 11 but it does not incorporate respondents who said they would “probably get the vaccine.” The survey measured perceived susceptibility by asking whether the respondent personally knew anyone who had been diagnosed with severe COVID. Perceived control was evaluated by asking: “How confident are you that you have the right information to make a choice about COVID vaccine?” Other questions evaluated accessibility of vaccination appointments; vaccine intentions; modifiable factors, such as concerns, attitudes, and beliefs; and how the health care system could be helpful with the respondent considering COVID vaccination. Respondents whose most important concern was that the vaccine was developed too fast or that they mistrusted the health care system were combined into one category for simplicity of analysis.
Data collection
The survey was administered via online, mail, and telephone modes. Individuals were initially sent invitations via email and US mail in Spanish or English, depending on their preferred written language. The email invitation, sent on June 14, 2021, to all individuals with an available valid email address, contained a link to the survey online. Two days later, all individuals in the sample were mailed a packet that included a cover letter with a unique link to the online survey, as well as a paper copy of the survey and a $10 Target gift card. A second wave of emailed survey invitations (sent June 28, 2021) and mailed survey invitations (sent July 16, 2021) was sent to initial nonresponders. The study team conducted telephone follow-up from June 14, 2021, through July 31, 2021, during which the team made 4 attempts to reach individuals. If the patient agreed to participate in the survey, the study team proceeded to administer the online survey over the phone. Survey data collection closed August 13, 2021.
Patient characteristics were drawn from Kaiser Permanente EHRs and geocoded US Census block-group data. Race, ethnicity, and language preference were self-reported by individuals. Individuals were counted as Latino–Spanish preferred if EHR data listed Spanish as their preferred written language. COVID-19 vaccination data were drawn from the EHRs for Kaiser Permanente Northern California, which include data on COVID-19 vaccines administered at all locations, including mass vaccination clinics, pharmacies, and the California state immunization registry. Individuals were counted as vaccinated if they had received at least one dose by February 6, 2022.
Statistical analysis
The primary analysis compared the characteristics of individuals with and without COVID-19 vaccine uptake during follow-up using chi-square analysis for categorical variables and the Wilcoxon rank-sum test for ordinal or continuous nonparametrically distributed variables. Tests of significance were two-tailed with an alpha of 0.05. A Bonferroni or other adjustment for type I error was not applied.
Results
Of the 1195 individuals surveyed, 793 (66%) responded. Compared with nonrespondents, the survey respondents were more likely to be Latino–Spanish preferred, less likely to be Black, and more likely to be from San Jose than the other 3 geographical areas studied. Respondents and nonrespondents did not differ significantly in age, sex, neighborhood deprivation index, or comorbidity scores. Of the respondents, 540 reported they had been vaccinated or clinical records showed they had been vaccinated and were excluded from further analysis, as were 40 whose remaining responses were not complete enough for analysis. Of the 213 respondents included in this study, 144 (67.6%) were aged 65–74 years; 63 (29.6%) were Black, 51 (23.9%) were Latino–English preferred, 34 (16.0%) were Latino–Spanish preferred, and 65 (30.5%) were Asian, White, other, or unknown race (Table 1). Eighty-one (38.0%) had a neighborhood deprivation index at the 75th percentile or higher.
Table 1:
Characteristics of individuals in the COVID-19 vaccination change study, Kaiser Permanente Northern California, June 2021–February 2022
| Characteristicsa | n (%) |
|---|---|
| Total individuals | 213 (100) |
| Age (years) | |
| 65–74 | 144 (67.6) |
| ≥ 75 | 69 (32.4) |
| Female sex | 123 (57.7) |
| Race/ethnicity–language | |
| Black | 63 (29.6) |
| Latino–English preferred | 51 (23.9) |
| Latino–Spanish preferred b | 34 (16.0) |
| Asian, White, other, or unknown | 65 (30.5) |
| Neighborhood Deprivation Index (percentile) | |
| < 25th (least deprived) | 41 (19.2) |
| 25th–74th | 83 (39.0) |
| 75th–89th | 37 (17.4) |
| < 90th (most deprived) | 44 (20.7) |
| Unknown | 8 (3.8) |
| Comorbidity scorec | |
| Low risk | 114 (53.5) |
| Medium risk | 75 (35.2) |
| High risk | 24 (11.3) |
| Geographic area | |
| Central Valley | 83 (39.0) |
| Fresno | 68 (31.9) |
| San Jose | 13 (6.1) |
| South Sacramento | 49 (23.0) |
Patient characteristics were drawn from Kaiser Permanente EHRs and from geocoded US Census block-group data.
Individuals were counted as Latino–Spanish preferred if EHR data listed Spanish as their preferred written language.
The Comorbidity Point Score, Version 2 (comorbidity score), 12 is the validated standard measure used for operational purposes within Kaiser Permanente Northern California.
EHR, electronic health record.
Baseline intentions and perceptions
Of the 213 individuals studied, 34 (16.0%) said they definitely would be vaccinated and 19 (8.9%) said they probably would. Eighty-nine (41.8%) said they definitely would not be vaccinated, 10 (4.7%) said they probably would not, and 56 (26.3%) said they were not sure (Table 2). Only 20 respondents (9.4%) reported that they would “only get the vaccine if required to do so for work or other activities.” To evaluate whether there was interaction between the randomized trial parent study and this cohort, analyses were conducted to determine whether intention to vaccinate was associated with group assignment (culturally tailored PCP outreach, standard PCP outreach, or usual care) in the randomized trial. There were no significant differences in vaccine intentions between groups.
Table 2:
Baseline intentions and perceptions among individuals in the COVID-19 vaccination change study, Kaiser Permanente Northern California, June–August 2021
| Intentions/perceptions | n (%) |
|---|---|
| Total individuals | 213 (100) |
| Intention regarding COVID-19 vaccinea | |
| Definitely get | 34 (16.0) |
| Probably get | 19 (8.9) |
| Not sure | 56 (26.3) |
| Probably not get | 10 (4.7) |
| Definitely not get | 89 (41.8) |
| Confident has the right information to make a choiceb | |
| Not at all | 46 (21.6) |
| A little/somewhat | 69 (32.4) |
| Very | 85 (39.9) |
| Most important concern about vaccine | |
| Developed too fast, mistrust of health care system | 52 (24.4) |
| Long-term side effects | 33 (15.5) |
| Religious/personal beliefs | 25 (11.7) |
| Short-term side effects | 24 (11.3) |
| Personal health issues | 16 (7.5) |
| Getting COVID is less dangerous than the vaccine | 15 (7.0) |
| Mistrust of government | 15 (7.0) |
| Needs more information, choice of vaccines, or access | 12 (5.6) |
| No response or unclassifiable | 21 (9.9) |
| Knows anyone who has had severe COVID-19 | |
| No one | 123 (57.7) |
| Household member | 11 (5.2) |
| Other family member or friend | 52 (24.4) |
| Other person | 27 (12.7) |
| How their health care system can be most helpfulc | |
| Nothing more, due to patient wanting to wait for more experience with vaccine | 79 (37.1) |
| Nothing more, due to patient’s beliefs | 52 (24.4) |
| More information, choice of vaccines, or access | 43 (20.2) |
| More information on personal health issue | 11 (5.2) |
Five responses were missing.
Thirteen responses were missing.
Twenty-eight responses were missing.
Confidence in having the right information to make a decision about vaccination varied widely and was associated with intention to be vaccinated. Eighty-five individuals (39.9%) were very confident, 69 (32.4%) were somewhat or a little confident, and 46 (21.6%) were not at all confident. Individuals who said they would definitely or definitely not be vaccinated expressed higher confidence than those without definite vaccine intentions (who said they were not sure, probably would, or probably would not be vaccinated, p < 0.001). Of the 85 individuals without definite intentions, only 23 (27.1%) were very confident in having the right information, while 37 (53.6%) were somewhat or a little confident, and 24 (52.2%) were not at all confident.
Regarding their most important concerns about vaccination, individuals most commonly said it was developed too fast or that they mistrusted the health care system (n = 52, 24.4%). Another 33 individuals (15.5%) named concerns about possible long-term side effects, while 25 (11.7%) named religious or personal beliefs, 24 (11.3%) named short-term side effects, and 16 (7.5%) named personal health issues. Other responses are shown in Table 2.
Changes between baseline vaccine intent and subsequent uptake
Among the 213 adults in this study, baseline intent was correlated with subsequent vaccination (p = 0.006), but baseline confidence in having the right information was not (p = 0.18; Table 3). Among the 89 individuals who initially said they would definitely not be vaccinated, 17 (19.1%) were vaccinated during the follow-up period. None of the 10 individuals who said they would probably not be vaccinated and 19 (33.9%) of the 56 who were not sure were subsequently vaccinated.
Table 3:
Changes between baseline intent and subsequent uptake of COVID-19 vaccine among older individuals, Kaiser Permanente Northern California, June 2021–February 2022
| Baseline intenta | Baseline n | n (% of category) vaccinated during follow-upb | p |
|---|---|---|---|
| Total individuals | 213 | ||
| Intention regarding COVID-19 vaccine | 0.006 | ||
| Definitely get | 34 | 16 (47.1) | |
| Probably get | 19 | 9 (47.4) | |
| Not sure | 56 | 19 (33.9) | |
| Probably not get | 10 | 0 (0) | |
| Definitely not get | 89 | 17 (19.1) | |
| No response | 5 | 0 (0) | |
| Confident has the right information to make a choice | 0.18 | ||
| Not at all | 46 | 10 (21.7) | |
| A little/somewhat | 69 | 24 (34.8) | |
| Very | 85 | 21 (24.7) | |
| No response | 13 | 6 (46.2) |
Age, sex, race or ethnicity, neighborhood deprivation index, comorbidity score, and geographic area were not associated with change in vaccine intent.
The baseline survey was conducted in June–August 2021. Vaccine uptake was analyzed using electronic health records during the follow-up period, which ended February 6, 2022.
In contrast, 16 (47.1%) of the 34 individuals who said they would definitely be vaccinated and 9 (47.4%) of the 19 individuals who said they would probably be vaccinated were subsequently vaccinated. Of the 53 individuals who said they would definitely or probably get the vaccine, those aged 65–74 years were more likely than those aged 75 years or older to actually be vaccinated (58% vs 23%, p = 0.02). No other factors were significantly associated with the probability of being vaccinated in this group.
Changes in vaccine-hesitant older adults
To identify associations between baseline intent and subsequent vaccine uptake, a subgroup analysis was conducted for the 155 individuals who were initially vaccine hesitant (survey responses of not sure, would probably not, or would definitely not get vaccinated; Table 4). Overall, 36 (23.2%) of these individuals were vaccinated during follow-up, including 17 (19.1%) of those who initially said they would definitely not get vaccinated. Individuals were more likely to become vaccinated if they had initially said they needed more information or choice of vaccines; if they said their most important concern was short-term side effects, religious, or personal beliefs; or if they believed that COVID-19 was less dangerous than portrayed or less dangerous than vaccine side effects (p < 0.001 for the chi-square test of association between all concerns and receipt of vaccine). In addition, individuals who were Latino or who initially said they were not sure were more likely than others to be vaccinated, but these differences were not statistically significant. The individuals least likely to be vaccinated were those who named long-term side effects or personal health issues as their primary concern. Baseline confidence in having the right information about the vaccine was not associated with subsequent vaccination (p = 0.38).
Table 4:
Factors associated with COVID-19 vaccine uptake among initially vaccine-hesitant older individuals, Kaiser Permanente Northern California, June 2021–February 2022
| Categorya | n | n (% of category) vaccinated during follow-up | pb |
|---|---|---|---|
| Total individuals in this subgroup analysis | 155 | 36 (23.2) | |
| Race/ethnicity–language | |||
| Black | 41 | 9 (22.0) | 0.08 |
| Latino–English preferred | 39 | 14 (35.9) | |
| Latino–Spanish preferred | 22 | 6 (27.3) | |
| Asian, White, other, or unknown | 53 | 7 (13.2) | |
| Intention regarding COVID-19 vaccine | 0.16 | ||
| Not sure | 56 | 19 (33.9) | |
| Probably not get | 10 | 0 (0) | |
| Definitely not get | 89 | 17 (19.1) | |
| Most important concern about vaccine | < 0.001 | ||
| Needs more information, choice of vaccines, or access | 4 | 4 (100) | |
| Getting COVID is less dangerous than the vaccine | 11 | 5 (45.5) | |
| Short-term side effects | 14 | 5 (35.7) | |
| Religious/personal beliefs | 20 | 7 (35.0) | |
| Mistrust of government | 13 | 2 (15.4) | |
| Developed too fast, mistrust of health care system | 42 | 6 (14.3) | |
| Long-term side effects | 27 | 3 (11.1) | |
| Personal health issues | 9 | 0 (0) | |
| No response or unclassifiable | 15 | 4 (26.7) |
Age, sex, neighborhood deprivation index, comorbidity score, and geographic area were not associated with change in vaccine intent.
p Value is from a chi-square analysis of a 2 × 2 table created by merging the “not sure” and “probably not get” categories compared to the “definitely not get” category.
Discussion
Major findings
In this study, more than 1 in 5 older adults who voiced hesitation to receive COVID-19 vaccine in summer 2021 received at least 1 vaccine dose approximately 7 months later, including those who initially said they would definitely not be vaccinated. Vaccine-hesitant individuals who identified needs for more information or short-term side effects or believed that COVID-19 was less dangerous than the vaccine were more likely than others to subsequently accept vaccination. These patterns suggest that COVID-19 vaccine intent is modifiable in some individuals and can potentially be influenced by providing emerging information about the risks of both COVID-19 disease and vaccine side effects.
Comparisons and implications
These findings are novel in that the study observed continuing change between COVID-19 vaccine intent and subsequent uptake in older adults who were still vaccine hesitant in summer 2021. Very few studies have attempted to evaluate changes between COVID-19 vaccine intent and actual uptake over time at an individual level. A study by the Kaiser Family Foundation found that of US adults who were vaccine hesitant in January 2021, 24% had become vaccinated by July 2021. 7 A later study found that 7% of adults who expressed vaccine hesitancy in April/May 2021 had become vaccinated by June/July 2021. 6 Other studies of changes in vaccine intent compared with uptake have been conducted in refugees 13 and in Poland. 14
The present study evaluated a later time period than the Kaiser Family Foundation or Szilagyi studies, with the baseline survey conducted several months after the vaccine had become widely available to older US adults. The relatively high proportion (23%) of vaccine-hesitant older adults who changed their minds in this study, including those who were initially not sure, may reflect 2 factors. First, the study’s time period between June 2021 and February 2022 included the COVID-19 surge due to the Delta variant and part of the surge due to the Omicron variant. These surges may have led to a perception of higher risk from COVID-19 disease than at the time of the baseline survey in summer 2021. This could have contributed to vaccine uptake, consistent with the Health Belief Model. 9 Second, more experience with the vaccine had accrued by late 2021, which likely offered reassurance or at least perceived predictability regarding its possible short-term side effects.
The study results suggest that continuing efforts to encourage COVID-19 vaccination are warranted, even for adults who state strong intentions against it. They also indicate that some subgroups are more amenable to change than others, depending on their most important concerns. Clinicians and policymakers could consider directing specific efforts toward adults who identify short-term side effects as their key concern; these adults seem more likely to be movable than those focused on long-term side effects. In addition, a higher likelihood of vaccine uptake in adults who initially believed that vaccine side effects could be worse than COVID-19 disease or that the disease was less dangerous than portrayed was observed. The Delta and Omicron surges during the follow-up period may have increased the perception of the risk of disease, particularly among this subgroup. In addition, in a study of more than 118,000 adults in this health care system from July 2021 through January 2022, 94% of those vaccinated completed their vaccination series, 15 suggesting that respondents receiving at least 1 dose likely became fully vaccinated.
This study also found that fewer than half of the individuals who said they were probably or definitely planning to get the vaccine were actually vaccinated during follow-up. Adults in this group were more likely to follow through on being vaccinated if they were aged 65–75 years, relative to those 75 years old or older. Various factors may account for the lack of follow-through, including accessibility (reliance on others for transportation) 16 or simply a change of mind. Future research could elucidate why some individuals who indicate initial intent to be vaccinated subsequently remain unvaccinated.
Limitations
This study population was highly diverse, although low-income individuals were somewhat underrepresented. The study group included older adults in Northern California and oversampled those who were Black and Latino; the findings may not be representative of younger adults or those in other states. Although the study’s survey response rate was 66%, the survey respondents who were unvaccinated and included in this study may not be representative of all unvaccinated adults. The capture of COVID-19 vaccination administration in the computerized data sources used here is believed to be relatively complete, but no data system is perfect. If some of the adults classified as unvaccinated were truly vaccinated, this would strengthen the study’s main finding that many adults with initial hesitancy subsequently accepted vaccination.
The present study’s findings are based on inference from the baseline survey responses; individuals were not directly asked why they changed from their original intent. Multiple tests of association were performed, but a formal statistical adjustment for type I error was not applied; thus, the findings on factors associated with change in intent should be interpreted as exploratory.
Conclusions
COVID-19 vaccine intent among older adults was malleable during the pandemic’s second year, even among those who initially said they would definitely not be vaccinated. Vaccine uptake could be enhanced by increasing awareness of COVID-19 risks and by addressing vaccine side effects.
Acknowledgments
We appreciate the excellent survey work of our research team, including Chelsy Alfaro, BS/BA, Savinnie Ho, Ananya Krishnapura, Evan Layefsky, BA, and Ria Sood. We are grateful to Matthew Daley, MD, for advising us on survey design. We thank Doug Corley, MD, for support from the TPMG Delivery Science and Applied Research program and Jacek Skarbinski, MD, for generously sharing staff to support this work. We are deeply grateful to the many physicians and members of Kaiser Permanente who supported this study, including the physicians-in-chief of the participating medical centers: Shahzad Jahromi, MD, Sanjay Marwaha, MD, Peter Miles, MD, and Efren Rosas, MD. Access to Data: Eric P. Elkin had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. We are thankful for the Kaiser Permanente Northern California members who participated in this study.
Footnotes
Author Contributions: Cimone Durojaiye, MPH, participated in the concept and design, acquisition and interpretation of data, and drafting and critical revision of the manuscript. Stephanie Prausnitz, MS, participated in the concept and design, acquisition of data, and critical revision of the manuscript. Eric P Elkin, MPH, participated in the analysis of data, including statistical analysis, and the drafting and critical revision of the manuscript. Patricia Escobar, MPH, participated in the concept and design, as well as the critical revision of the manuscript. Lucy Finn, MPH, participated in the concept and design, as well as the critical revision of the manuscript. Yi-Fen Irene Chen, MD, participated in the concept and design, obtaining funding, supervision, and critical revision of the manuscript. Tracy A Lieu MD, MPH, participated in the concept and design, obtaining funding, interpretation of data, and drafting and critical revision of the manuscript.
Conflicts of Interest: All authors are employees of The Permanente Medical Group. The authors have no other relationships or conflicts of interest to declare.
Funding: This study was funded by The Permanente Medical Group via its Delivery Science and Applied Research Program.
References
- 1. CDC (Centers for Disease Control and Prevention) . COVID-19 vaccinations in the United States. Updated May 20, 2022. Accessed 16 December 2021. https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total
- 2. CDC . Demographic trends of COVID-19 cases and deaths in the US reported to CDC. Updated May 20, 2022. Accessed 16 December 2021. https://covid.cdc.gov/covid-data-tracker/#demographics
- 3. Escobar GJ, Adams AS, Liu VX, et al. Racial disparities in COVID-19 testing and outcomes: Retrospective cohort study in an integrated health system. Ann Intern Med. 2021;174(6):786–793. 10.7326/M20-6979 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Nguyen KH, Nguyen K, Corlin L, Allen JD, Chung M. Changes in COVID-19 vaccination receipt and intention to vaccinate by socioeconomic characteristics and geographic area, United States, January 6 – March 29, 2021. Ann Med. 2021;53(1):1419–1428. 10.1080/07853890.2021.1957998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Ruiz JB, Bell RA. Predictors of intention to vaccinate against COVID-19: Results of a nationwide survey. Vaccine. 2021;39(7):1080–1086. 10.1016/j.vaccine.2021.01.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Szilagyi PG, Thomas K, Shah MD, et al. Changes in COVID-19 vaccine intent from April/May to June/July 2021. JAMA. 2021;326(19):1971–1974. 10.1001/jama.2021.18761 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kirzinger ASG, Brodie M. KFF COVID-19 vaccine monitor: In their own words, six months later. Updated July 13, 2021. Accessed 3 May 2021. https://www.kff.org/coronavirus-covid-19/poll-finding/kff-covid-19-vaccine-monitor-in-their-own-words-six-months-later/
- 8. Lieu TA, Elkin EP, Escobar PR, et al. Effect of electronic and mail outreach from primary care physicians for COVID-19 vaccination of Black and Latino older adults: A randomized clinical trial. JAMA Netw Open. 2022;5(6):e2217004. 10.1001/jamanetworkopen.2022.17004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Becker MH. The health belief model and sick role behavior. Health Education Monographs. 1974;2(4):409–419. 10.1177/109019817400200407 [DOI] [Google Scholar]
- 10. Montano DE, Kasprzyk D. Chapter 6: Theory of reasoned action, theory of planned behavior, and the integrated behavioral model. In: Glanz K, Rimer BK, Viswanath K, eds. Health Behavior: Theory, Research, and Practice. 5th ed. Jossey-Bass/Wiley; 2015:95–124. [Google Scholar]
- 11. Larson HJ, Gakidou E, Murray CJL. The vaccine-hesitant moment. N Engl J Med. 2022;387(1):58–65. 10.1056/NEJMra2106441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Escobar GJ, Gardner MN, Greene JD, Draper D, Kipnis P. Risk-adjusting hospital mortality using a comprehensive electronic record in an integrated health care delivery system. Med Care. 2013;51(5):446–453. 10.1097/MLR.0b013e3182881c8e [DOI] [PubMed] [Google Scholar]
- 13. Shaw J, Anderson KB, Fabi RE, et al. COVID-19 vaccination intention and behavior in a large, diverse, U.S. refugee population. Vaccine. 2022;40(9):1231–1237. 10.1016/j.vaccine.2022.01.057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Maciuszek J, Polak M, Stasiuk K. Declared intention (not) to be vaccinated against COVID-19, and actual behavior—the longitudinal study in the Polish sample. Vaccines (Basel). 2022;10(2):147. 10.3390/vaccines10020147 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Skarbinski J, Wood MS, Chervo TC, et al. Risk of severe clinical outcomes among persons with SARS-CoV-2 infection with differing levels of vaccination during widespread Omicron (B.1.1.529) and Delta (B.1.617.2) variant circulation in Northern California: A retrospective cohort study. Lancet Reg Health Am. 2022;12:100297. 10.1016/j.lana.2022.100297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bedford H, Attwell K, Danchin M, Marshall H, Corben P, Leask J. Vaccine hesitancy, refusal and access barriers: The need for clarity in terminology. Vaccine. 2018;36(44):6556–6558. 10.1016/j.vaccine.2017.08.004 [DOI] [PubMed] [Google Scholar]