Association of Race/Ethnicity With Likeliness of COVID-19 Vaccine Uptake Among Health Workers and the General Population in the San Francisco Bay Area

Surveys have demonstrated racial differences in the public’s willingness to receive a COVID-19 vaccine^1,2 but have not directly compared vaccine intentions among health workers and the general public.³ We investigated COVID-19 vaccine intentions among racially and ethnically diverse samples of health workers and the general population.

We conducted a cross-sectional survey from November 27, 2020, to January 15, 2021, nested within 2 longitudinal cohort studies of prevalence and incidence of SARS-CoV-2 infection in 6 San Francisco Bay Area counties. The general population cohort comprised 3935 community-residing adults sampled from randomly selected households, and the medical center employee cohort comprised 2501 employees of 3 large medical centers, who volunteered for biweekly to monthly COVID-19 testing. The main outcome measure was likeliness of vaccine uptake, derived from 2 survey items: (1) “How likely are you to get an approved COVID-19 vaccine when it becomes available?” (using a 1-7 Likert scale anchored at “not at all likely” and “very likely”), and (2) “How early would you ideally like to receive the COVID-19 vaccine?” (asked of respondents scoring ≥3 on the first item). The survey also included items asking about reasons to get, and to not get, vaccinated. Respondents self-identified race/ethnicity (see eMethods in the Supplement for details on sampling and the survey instrument). Crude results were compared using 2-tailed χ² tests, with P < .05 considered significant. Logistic regression models stratified by cohort tested association of race/ethnicity with vaccine willingness, adjusting for age, gender, and level of education. All statistical analyses were performed using SAS, version 9.4 (SAS Institute). American Association for Public Opinion Research Response Rate 1 definition was used.

The University of California, San Francisco, and Stanford Institutional Review Boards designated the general population cohort study a public health surveillance study and approved the medical center employee cohort study protocol. Written electronic informed consent was obtained at enrollment.

A total of 3161 of 3935 (80.3%) participants in the general population cohort and 1803 of 2501 (72.1%) participants in the medical center employee cohort responded to the vaccine survey (Table). Although a higher proportion of medical center employees than members of the general population reported likeliness of vaccine uptake, racial/ethnic differences in likeliness were comparable in both cohorts (Figure). In the medical center cohort, the adjusted odds ratio (aOR) (95% CI) of likeliness of vaccine uptake relative to White respondents was 0.24 (0.10-0.60) for Black respondents, 0.50 (0.31-0.79) for Latinx respondents, 0.37 (0.27-0.51) for Asian respondents, 0.28 (0.15-0.53) for respondents of other races, and 0.49 (0.29-0.82) for respondents of multiple races. In the general population cohort, the aOR (95% CI) relative to White respondents was 0.29 (0.20-0.43) for Black respondents, 0.55 (0.43-0.71) for Latinx respondents, 0.57 (0.47-0.70) for Asian respondents, 0.62 (0.38-1.02) for respondents of other races, and 0.65 (0.46-0.92) for respondents of multiple races. Ratings of reasons to get vaccinated were similar across racial/ethnic groups, but Black, Latinx, and Asian respondents were significantly more likely than White respondents to endorse reasons to not get vaccinated, especially less confidence in the vaccine preventing COVID-19 (aOR [95% CI] for Black, Latinx, and Asian respondents having low confidence relative to White respondents, 2.39 [1.58-3.61], 2.04 [1.58-2.64], and 1.85 [1.51-2.27], respectively); less trust in companies making the vaccine (aOR [95% CI] for Black, Latinx, and Asian respondents having low trust relative to White respondents, 3.08 [2.00-4.73], 1.85 [1.38-2.48], and 1.34 [1.04-1.72], respectively); and more worry that government rushed the approval process (aOR [95% CI] for Black, Latinx, and Asian respondents relative to White respondents, 2.10 [1.44-3.05], 1.68 [1.34-2.10], and 1.81 [1.53-2.15], respectively).

In this survey study including a diversity of racial/ethnic groups, occupational immersion in a health care setting did not offset disparities in COVID-19 vaccination intentions. We found that Asian individuals, multiracial individuals, and those of other races were more similar to Black and Latinx individuals than White individuals in their likeliness of vaccine uptake. Limitations of this study include that the sample was drawn from people sufficiently concerned about their risk of COVID-19 and trusting of research to volunteer for a study involving repeated COVID-19 testing and the survey not including additional domains, such as perceived access, that might influence reported likeliness of vaccine uptake. However, it is striking that even among individuals motivated to participate in a longitudinal COVID-19 testing study, there were racial/ethnic differences in COVID-19 vaccination intentions and concerns about the vaccine.

Black, Latinx, Asian, and Native American communities have borne a disproportionate toll of the COVID-19 pandemic in the US⁴; inequities in vaccination would compound these disparities. Our survey was fielded at the time of the first emergency use authorization of COVID-19 vaccines in the US. Vaccination rollout since then has revealed barriers to accessing vaccination among historically marginalized populations who are highly motivated to be vaccinated.⁵ Vaccination intentions must be understood as a deliberative and dynamic process; a focus on intentions must not distract from the importance of ensuring equitable access to vaccination.⁵ Special effort is required to reach historically marginalized populations, including those in health occupations, to support informed vaccination decision-making and facilitate access. Efforts must acknowledge a history of racism that has degraded the trustworthiness of health and medical science institutions among historically marginalized populations,⁶ undermined confidence in COVID-19 vaccines, and perpetuated inequitable access to care.

Accepted for Publication: March 6, 2021.

Published Online: March 30, 2021. doi:10.1001/jamainternmed.2021.1445

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Grumbach K et al. JAMA Internal Medicine.

Corresponding Author: Kevin Grumbach, MD, San Francisco General Hospital, Department of Family and Community Medicine, University of California, San Francisco, 1001 Potrero Ave, Ward 83, Room 310, San Francisco, CA 94110 (kevin.grumbach@ucsf.edu).

Author Contributions: Dr Grumbach had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Grumbach, Judson, Jain, Lindan, Doernberg, Holubar.

Acquisition, analysis, or interpretation of data: Grumbach, Desai, Jain, Lindan, Doernberg, Holubar.

Drafting of the manuscript: Grumbach, Judson, Holubar.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Desai, Holubar.

Administrative, technical, or material support: Grumbach, Judson, Jain, Lindan.

Supervision: Grumbach, Judson, Jain, Lindan, Doernberg.

Conflict of Interest Disclosures: Dr Jain reported receiving grants from the Centers for Disease Control and Prevention/President’s Emergency Plan For AIDS Relief not related to this work during the conduct of the study. Dr Lindan reported receiving grants from the Chan Zuckerberg Initiative during the conduct of the study. Dr Doernberg reported receiving grants from the Chan Zuckerberg Initiative during the conduct of the study and receiving personal fees from Genentech and Basilea Pharmaceutica for consulting outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by the Chan Zuckerberg Initiative. Dr Grumbach’s effort was partly supported by a grant from the National Institutes of Health Community Engagement Alliance Against COVID-19 Disparities program (21-312-0217571-66106L).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The contents of the article are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.

Additional Contributions: Yvonne Maldonado, MD (Division of Pediatric Infectious Diseases, Stanford University School of Medicine), and George W. Rutherford, MD (Department of Epidemiology and Biostatistics, University of California, San Francisco), serve as principal investigators of the California Pandemic Consortium and obtained study funding, designed the cohort studies, and provided input into survey study design. Yingjie Weng, MS, Di Lu, MS, and Derek Boothroyd, PhD, of the Quantitative Sciences Unit, Department of Medicine, Stanford University, provided consultation on analytic methods and performed data analysis. Jenna Bollyky, MD (Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine), and Hannah Sample, BS (Department of Epidemiology and Biostatistics, University of California, San Francisco), managed the study cohorts. Beatrice Huang, BA (Department of Family and Community Medicine, University of California, San Francisco), contributed to survey instrument development and administration. The authors thank all additional project staff for their dedicated work on this study and community partners who collaborated on recruitment of participants for the general population cohort. These individuals did not receive compensation for their contributions beyond their employment salaries.