Who is getting boosted? Disparities in COVID-19 vaccine booster uptake among health care workers

Abstract

COVID-19 vaccination remains one of the most effective tools to reduce the risk of SARS-CoV-2 infection. Unfortunately, vaccine hesitancy has limited primary vaccination and booster uptake among the general population and HCWs. To gain a better understanding of factors associated with booster vaccine uptake, we analyzed COVID-19 vaccine booster rates among HCWs and identified risk factors associated with nonacceptance.

Of the 62,387 HCWs included in our analysis, the overall booster uptake rate was 64.8%. Older age, Non-Hispanic White racial group, early initial vaccine uptake and longer duration of employment were associated with higher booster uptake. Significant differences were observed between different job categories.

This persistence of vaccine hesitancy and disparities in COVID-19 booster uptake among HCWs, almost 2 years after the rollout of the COVID-19 vaccination, call for further efforts to increase vaccine confidence among HCWs and the general population in light of the continued need for further COVID-19 protection.

1. Background

COVID-19 vaccination remains one of our most effective tools to reduce the risk of infection, reinfection, severity of illness and death from SARS-CoV-2 infection [1], [2], [3], [4], [5].

Despite the demonstrated effectiveness and safety of the vaccine, [6] vaccine hesitancy has limited uptake, threatening the likelihood of attaining herd immunity [7]. Unfortunately, health care workers (HCWs) have not been immune to vaccine hesitancy, with several studies reporting significant disparities in primary series vaccine uptake among HCWs from different age groups, ethnic origins, and job roles [8], [9], [10], [11]. While vaccine mandates have led to higher and more equitable primary COVID-19 vaccination among HCWs, [12] surveys of HCW’s intent to receive booster doses suggest that vaccine hesitancy persists and may cause ongoing disparities in booster uptake [13], [14], [15].

While hesitancy to accept primary COVID-19 vaccination has been extensively examined among United States (US) HCWs, [8], [9], [10], [11] to our knowledge the evidence regarding likelihood to accept a booster vaccine among HCW in US is limited to one survey of 1374 HCW demonstrating lower odds of intent to accept a booster among HCW who were initially hesitant to accept a primary COVID-19 vaccination [14]. The persistent hesitancy among HCWs raises concern given their ongoing potential risk to contract and transmit infection within healthcare settings and the significant role they play in building vaccine confidence among the general population, especially as boosters are likely to be crucial to address waning immunity and newly emerging variants [4], [5].

To gain a better understanding of factors associated with booster vaccine uptake, we analyzed COVID-19 vaccine booster rates among HCWs and identified risk factors associated with nonacceptance.

2. Methods

This is a retrospective analysis of COVID-19 vaccine booster data through 04/01/2022 for HCWs at a large multi-site US academic medical center with facilities in four states (Arizona, Florida, Minnesota, and Wisconsin). We included all actively employed HCWs with a hire date before 07/01/2021, who were eligible for the COVID-19 vaccine booster during the study period. Booster eligibility was defined as completion of a primary COVID-19 mRNA vaccination series or a single dose of Janssen vaccine more than 5 or 2 months, respectively, before the study end date.

2.1. Vaccination program

The COVID-19 vaccination program was initiated in December 2020 on voluntary basis and in a staged rollout based on occupational risk [16]. All HCWs were eligible to receive the vaccine by 03/31/2021. HCWs hired after 06/30/2021 were required to receive COVID-19 primary vaccination before starting work, and all existing employees were required to complete a primary vaccination series by 1/5/2022, with only approved medical or religious exemptions allowed. Boosters have been available on a voluntary basis starting in late September 2021 and are offered to all HCWs at the appropriate interval after their primary vaccination. Employees receiving booster vaccination outside Occupational Health Services (OHS) by other providers in the practice complete an electronic release of information allowing the vaccination record to be shared with OHS.

2.2. Data Collection

Vaccine and booster data (vaccination status, booster status and dates as applicable), demographic, and occupational data (age, gender, location, employment duration, job) and most recent positive SARS-CoV-2 molecular assay results and dates were obtained from occupational health records. All records were deidentified before analysis. The definition of the included variables is fully described in a supplementary table 1. (S-table-1) For the purpose of this study unvaccinated HCWs were excluded from analysis. Institutional vaccination data management was described previously [9].

2.3. Statistical analysis

Data were summarized as median and interquartile range for continuous variables and absolute and relative frequencies (%) for categorical variables. Comparisons between groups were performed using nonparametric Wilcoxon rank-sum test for continuous variables and the chi-square test for categorical variables. A logistic regression analysis with robust standard errors was used to assess the associations between COVID-19 vaccine booster uptake and risk factors, including job category, demographics (age categories, gender and practice location), early acceptance of initial vaccine dose and history of previous COVID-19 infection. We adjusted for job category (nonclinical support staff (support staff in patient care environments with no direct patient care duties) vs other job categories), race (White vs other race groups), age (<25 vs other age categories), gender, and history of previous COVID-19 infection. To account for variation in vaccine hesitancy among US population, we also adjusted for practice location (Midwest, AZ, FL) and early acceptance of initial vaccine dose. An early initial dose was defined as a first dose received before 8/01/2021. Variance inflation factor testing was performed to assess for multicollinearity. All statistical analyses were performed using a standard software package (Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC.). The study was approved by the Mayo Clinic Institutional Review Board (IRB application #20–007051).

3. Results

During the study period, 62,387 HCWs met the inclusion criteria. Seventy-one percent of HCWs were female, with an average age of 42.9 years (IQR: 18 to 100) and an average duration since hire of 85.6 months (IQR: 0 to 778.5), the majority were Non-Hispanic White (82.4 %) and from a midwest practice site (76.5 %). Nursing staff and support staff accounted for 49.4 % of the HCWs evaluated, followed by physicians (10.5 %) and administrative staff (20.2 %). The majority of the HCWs (89.2 %) received their primary COVID vaccination series before 08/01/2021 with average of 22 days (IQR: 0 to 226) from the initial vaccine availability day compared to 301 days (IQR: 227 to 462) among those receiving an initial dose after 08/01/22. (Table 1).

Table 1.

Demographic Characteristics of the participants and SARS-CoV-2 booster vaccination.

Variables	Booster Dose N:40,430 (64.8 %)	No Booster Dose N:21,957 (35.2 %)	P value**
Age Groups, years			<0.0001
<25	1,299 (45.1 %)	1,580 (54.9 %)
25–34	9,109 (58.3 %)	6,513(41.7 %)
35–44	10,667 (64.4 %)	5,899 (35.6 %)
45–54	9,105 (67.9 %)	4,307 (32.1 %)
55–65	8,636 (73.1 %)	3,178 (26.9 %)
>=65	1,446 (78.3 %)	400 (21.7 %)
Gender			<0.0001
Male	12,336 (67.96 %)	5,817 (32.04 %)
Female	27,926 (63.5 %)	16,060 (36.5 %)
Race/Ethnicity			<0.0001
Non-Hispanic White	33,891 (65.9 %)	17,538 (34.1 %)
Black	1,047 (44.6 %)	1,299 (55.4 %)
Hispanic	1,515 (56.8 %)	1,151 (43.2 %)
Asian	3,135 (69.4 %)	1,382 (30.6 %)
Other	674 (57.1 %)	507 (42.9 %)
Job location			<0.0001
Midwest region	32,677 (68.5 %)	15,056 (31.5 %)
Arizona	3,516 (49.6 %)	3,572 (50.4 %)
Florida	3,567 (54.6 %)	2,963 (45.4 %)
Job category			<0.0001
Nonclinical Support Staff	5,531 (55.1 %)	4,509 (44.9 %)
Physician	5,566 (84.7 %)	1,005 (15.3 %)
Advanced Practice	2,068 (80.7 %)	495 (19.3 %)
Nurse	9,106 (65.2 %)	4,853 (34.8 %)
Healthcare Professional	3,602(71.1 %)	1,465 (28.9 %)
Clinical Support Staff	3,409 (51.3 %)	3,238 (48.7 %)
Research	3,285 (69.9 %)	1,411 (30.1 %)
Administrative Staff	7,695 (61.1 %)	4,901 (38.9 %)
Patient Care Role			<0.0001
Direct Care	22,768 (68.9 %)	10,285 (31.1 %)
No Direct Care	17,494 (60.1 %)	11,592 (39.9 %)
Vaccine			<0.0001
Early Vaccination	40,218 (72.6 %)	15,211 (27.4 %)
Late Vaccination	44 (0.7 %)	6,666 (99.3 %)
Duration of hire, months Median (IQR)	108.9 (0–778.5)	64.7 (0–605.5)	<0.0001
Prior positive SARS-CoV-2 molecular assay***	3,985 (32.2 %)	8,385 (67.8 %)	<0.0001

SARS-CoV-2: severe acute respiratory syndrome coronavirus 2.

*Missing variables job location were excluded from the analysis.

** Nonparametric Wilcoxon rank-sum test for continuous variables and the chi-square test for categorical variables.

*** Positive tests after booster dose were excluded for booster dose group.

The overall COVID vaccine booster uptake rate was 64.8 % (40,262 HCWs) during the study period. Older age, Non-Hispanic White racial group, early initial vaccine uptake and longer duration of hire factors were associated with higher booster uptake (P <.0001). (Table 1).

While direct patient care status was associated with increased odds of booster vaccine uptake, differences were observed between different job categories. Physicians, advanced practice staff, nurses and healthcare professionals were more likely to receive a booster dose than administrative staff and support staff (P <.0001). (Table 1).

The multivariable analysis (Fig. 1) illustrated differences between job categories, gender, age groups, location, early uptake of primary COVID vaccination and previous positive SARS-CoV-2 molecular assay. The odds of being boosted increased with increasing age, with the highest odds of receiving booster dose among HCWs older than 65 years (OR, 3.09; 95 % CI, 2.6–3.6). The differences in booster rate by race and ethnicity were demonstrated in the multivariable analysis, with Black or African American HCWs having the lowest odds of being boosted (OR, 0.6; 95 % CI, 0.55–0.68) followed by other and Asian ethnic groups (OR, 0.87; 95 % CI, 0.75–1.01 and OR, 0.94; 95 % CI, 0.87–1.02 respectively). A history of previous positive SARS-CoV-2 molecular assay was associated with a lower booster rate (OR, 0.24; 95 % CI, 0.23–0.25P <.0001).

Fig. 1.

Multivariable Logistic Regression Analysis of Factors Associated with COVID-19 Vaccine Booster Uptake.

4. Discussion

In our study, 64.8 % of HCWs received a COVID-19 booster vaccination during the study period. While this is higher than the 48–51 % uptake reported among the general population [17], [18], it is lower than the 83.6 % rate predicted by Pal et al [14] or the reported 73.8 % booster uptake among HCWs in Singapore [13].

Given the need for booster vaccinations to improve the immunogenicity of the vaccine and prolong protection [4], the lower rate of booster uptake observed in our cohort demonstrates the need to understand the patterns and disparities associated with COVID-19 booster vaccination acceptance in HCWs.

Our cohort demonstrated that early uptake of primary COVID-19 vaccination was associated with significantly higher odds of receiving a booster dose, which aligns with the data reported by others [13], [14]. The lower uptake of booster dose among initially hesitant HCWs highlights the need for different strategies to address the factors driving vaccine hesitancy.

Previously reported data on primary COVID-19 vaccination has highlighted significant disparities in vaccine uptake by age, gender, ethnic origins, and job roles and previous COVID-19 infection status [8], [9], [10], [11]. Our cohort redemonstrated these disparities in COVID-19 booster vaccination uptake among subgroups even after adjusting for early acceptance of primary vaccination.

Consistent with findings in primary vaccination uptake, we found increased booster uptake in older age groups, which may reflect higher perceived vulnerability to COVID-19 [14], [19]. We also found disparities in COVID-19 booster vaccination uptake among racial and ethnic groups, similar to the disparities observed in primary vaccination and booster uptake among the general population and other HCWs [11], [17], [20]. After adjusting for other factors including geographic location rates of booster vaccination, uptake continued to be lower in Black or African American HCWs, compared to Non-Hispanic White HCWs which highlights the complexity of addressing vaccine hesitancy within racial and ethnic groups. Additionally, the observed difference in booster uptake among different geographic location aligns with variation in booster vaccine uptake among the general population, highlighting the influence of environmental, social and cultural factors in vaccine hesitancy [17], [21].

Persistent disparities in booster vaccination uptake among the job categories in the patient care environment, which have also been observed by others [13], [19], are concerning given that some undervaccinated groups represent those with lower educational and socioeconomic status who may be at greater risk for adverse health outcomes. The lower booster vaccination acceptance among HCWs who were previously infected with COVID-19, can likely be attributed to perceived natural immunity.

Overall our results showed a slightly lower booster uptake than actual and predicted rates already published [13], [14], while demonstrating persistent vaccine hesitancy among HCWs who were hesitant or late to receive their primary COVID-19 vaccination. In comparison to previously published studies [13], [14], [19] the strengths of our study include the large sample size (n = 62,387) and use of an integrated occupational health database across all clinic and hospital sites which allowed us to obtain actual vaccination and booster uptake data and to assess variation in booster uptake by geographic region. This decreased the risk of ascertainment bias in vaccination status and prior COVID-19 infection.

However, our study is subject to several limitations. First is the utilization of Human Resources demographic data which can include omission of racial and ethnic data for some HCWs. However, in our cohort <2 % of demographic information was missing with no significant difference between our primary variables and missing and non-missing values. Second, the use of documented data of positive COVID-PCR results to identify SARS-CoV-2 infections can contribute to under-reporting of cases; however, test documentation was necessary to issue work restrictions for the isolation period, and supervisors needed this official work restriction from OHS to correctly code the absence. Therefore, the risk of underreporting is low in this occupational setting. Third, we were unable to assess other factors that could contribute to vaccine hesitancy such as experienced side effects following primary covid-19 vaccination, medical comorbidities, health beliefs, socioeconomic factors, or political views. Fourth, while our study includes a large cohort of HCWs, it may not be fully representative of all HCWs across the US. Although the inclusion of HCWs from three different regions partially mitigates this limitation, over 80 % of our cohort are Non-Hispanic White compared to just 60 % of the national population. Finally, since booster vaccination is not a requirement for employment, and employees may obtain booster doses externally without reporting this to OHS, the overall booster coverage rates may be underestimated.

Our study adds to the limited available data regarding COVID-19 booster vaccination uptake among HCWs, which can help guide future vaccine campaigns and public health strategies to build vaccine confidence among HCWs and the general population.

5. Conclusion

We found a persistence of vaccine hesitancy and disparities in COVID-19 booster vaccine uptake in a large cohort of US HCWs. Lower booster uptake was observed in administrative and support staff, younger workers, Black or African American HCWs, those who were slower to accept initial vaccination, and HCWs with prior infection. These findings, almost 2 years after the rollout of the COVID-19 vaccination, call for further efforts to increase vaccine confidence among HCWs and the general population in light of the continued need for further COVID-19 boosters.

Funding

This work was supported by the Division of Public Health, Infectious Diseases and Occupational Medicine at Mayo Clinic, Rochester.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.