key: cord-0695840-q1049zfy
authors: Bell, S.; Clarke, R. M.; Ismail, S. A.; Ojo-Aromokudu, O.; Naqvi, H.; Coghill, Y.; Donovan, H.; Letley, L.; Paterson, P.; Mounter-Jack, S.
title: COVID-19 vaccination beliefs, attitudes, and behaviours among health and social care workers in the UK: a mixed-methods study
date: 2021-04-25
journal: nan
DOI: 10.1101/2021.04.23.21255971
sha: d717913a8da05dcaeb14a0b2f8ea7667fc344dc1
doc_id: 695840
cord_uid: q1049zfy

Background: The UK began delivering its COVID-19 vaccination programme on 8 December 2020, with health and social care workers (H&SCWs) given high priority for vaccination. Despite well-documented occupational exposure risks, however, there is evidence of lower uptake among some H&SCW groups. Methods: We used a mixed-methods approach involving an online cross-sectional survey and semi-structured interviews to gain insight into COVID-19 vaccination beliefs, attitudes, and behaviours amongst H&SCWs in the UK by socio-demographic and employment variables. 1917 people were surveyed 1658 healthcare workers (HCWs) and 261 social care workers (SCWs). Twenty participants were interviewed. Findings: Workplace factors contributed to vaccination access and uptake. SCWs were more likely to not be offered COVID-19 vaccination than HCWs (OR:1.453, 95%CI: 1.244-1.696). SCWs specifically reported uncertainties around how to access COVID-19 vaccination. Participants who indicated stronger agreement with the statement "I would recommend my organisation as a place to work" were more likely to have been offered COVID-19 vaccination (OR:1.28, 95%CI: 1.06-1.56). Those who agreed more strongly with the statement "I feel/felt under pressure from my employer to get a COVID-19 vaccine" were more likely to have declined vaccination (OR:1.75, 95%CI: 1.27-2.41). Interviewees that experienced employer pressure to get vaccinated felt this exacerbated their vaccine concerns and increased distrust. In comparison to White British and White Irish participants, Black African and Mixed Black African participants were more likely to not be offered (OR:2.011, 95%CI: 1.026-3.943) and more likely to have declined COVID-19 vaccination (OR:5.55, 95%CI: 2.29-13.43). Reasons for declining vaccination among Black African participants included distrust in COVID-19 vaccination, healthcare providers, and policymakers. Conclusion: H&SCW employers are in a pivotal position to facilitate COVID-19 vaccination access, by ensuring staff are aware of how to get vaccinated and promoting a workplace environment in which vaccination decisions are informed and voluntary.

Ambulance (operational) 9 (90.0%) 0 (0%) 1 (10.0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)

Public Health 19 (63.3%) 1 (3.3%) 3 (10.0%) 2 (6.7%) 1 (3.3%) 2 (6.7%) 2 (6.7%) 0 (0%)

Commissioning 4 (50.0%) 0 (0%) 1 (12.5%) 2 (25.0%) 0 (0%) 1 (12.5%) 0 (0%) 0 (0%) Registered Nurses and Midwives 385 (67.3%) 33 (5.8%) 55 (9.6%) 20 (3.5%) 25 (4.4%) 25 (4.4%) 23 (4.0%) 6 (1.0%)

Nursing or Healthcare Assistants 135 (68.9%) 12 (6.1%) 10 (5.1%) 9 (4.6%) 9 (4.6%) 11 (5.6%) 9 (4.6%) 1 (0.5%)

Wider Healthcare Team 44 (62.0%) 2 (2.8%) 5 (7.0%) 7 (9.9%) 1 (1.4%) 6 (8.5%) 5 (7.0%) 1 (1.4%)

General Management 49 (52.1%) 3 (3.2%) 11 (11.7%) 8 (8.5%) 11 (11.7%) 5 (5.3%) 6 (6.4%) 1 (1.1%)

Do not know/Did not answer 4 (80.0%) 0 (0%) 0 (0%) 1 (20.0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 

Residential 36 (75.0%) 1 (2.1%) 2 (4.2%) 2 (4.2%) 2 (4.2%) 2 (4.2%) 1 (2.1%) 2 (4.2%) Domiciliary 79 (88.8%) 3 (3.4%) 4 (4.5%) 0 (0%) 0 (0%) 1 (1.1%) 0 (0%) 2 (2.2%) Day 6 (35.3%) 3 (17.6%) 2 (11.8%) 2 (11.8%) 0 (0%) 0 (0%) 3 (17.6%) 1 (5.9%) Community 63 (82.9%) (%) 4 (5.3%) 3 (3.9%) 1 (1.3%) 2 (2.6%) 2 (2.6%) 1 (1.3%) 0 (0%) Regulated Professionals 6 (85.7%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 1 (14.3%) 0 (0%)

Other 16 (69.6%) 2 (8.7%) 0 (0%) 0 (0%) 1 (4.3%) 1 (4.3%) 0 (0%) 3 (13.0%)

Do not know/Did not answer 1 (100%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (%) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 A third of survey participants (n=640; 33.5%) provided their details to be contacted for interview. 200

Most of these indicated in the survey that they had been vaccinated (n=534; 83.4%) and were HCWs 201 (n=563; 88.0%). All participants that reported declining COVID-19 vaccination and left their contact 202 details were contacted (n=28). We also contacted 34 H&SCWs that reported they had been or 203 planned to get vaccinated, 16 H&SCWs who had not yet been offered the vaccine, and 1 H&SCW 204 who did not provide their vaccination status. In total, 81 participants were contacted, of these 16 205

HCWs and 4 SCWS were interviewed. 206

Participants were recruited from across ethnic groups, age categories, job roles, and geographical 207 areas (see Table 4 ). In the period between the survey and the interview, three participants had 208 changed their decision from declining to accepting COVID-19 vaccination. At interview all 209 participants had been offered vaccination, 13 had been vaccinated, 1 participant had booked in to 210 receive their vaccine, and 7 had declined vaccination. 211 212 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) Sector . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. No ꝉ ------

No flu vaccine uptake ꝉ ------ 

White British and White Irish ꝉ . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. Quantitative findings: differences in beliefs by ethnicity 249 Table 7 indicates the differences in beliefs and trust in sources of information across the ethnic 250 minority groups as compared to White British or White Irish participants. 251 252 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101/2021.04.23.21255971 doi: medRxiv preprint Fifty-six percent of participants wanted to be vaccinated to protect themselves/avoid catching 273 COVID-19 (see Figure 1 ). The next most given reasons to accept vaccination were to protect family 274 and/or friends (40% of participants), to protect service users (28%), to protect others/society/the 275 community (24%), to get life back to 'normal', to control the pandemic (20%), and to protect 276 colleagues (10%). 277 We highlight where specific themes were reported more dominantly by HCWs or SCWs. We use the 309 term participants when themes emerged in interviews and survey free-text responses, and highlight 310 where themes emerged in interviews or free-text responses only. 311 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101/2021.04.23.21255971 doi: medRxiv preprint

Most NHS HCWs reported that it was easy to organise vaccination through formal invitations via 313 their employer. For others, including SCWs and those with non-mainstream NHS employers (private, 314 bank staff, locums, pharmacists) COVID-19 vaccination was described as 'quite a battle' as they were 315 not given advice about how to organise vaccination by their employer(s), or received mixed 316 messages around whether they should arrange an appointment through their general practitioner or 317 employer. In trying to access COVID-19 vaccination, some of these HCWs had reached out to their 318 general practice but were not able to access the vaccine through this route. One interviewee was 319 told to wait until their age category joined the priority list to be vaccinated, and in one case was 320 advised to book vaccination through an NHS web link. 321

One interviewed HCW felt that although some were able to access vaccination appointments during 322 working hours, in practice this was not happening for all staff, noting that staff in job grades 5 and 323 below, often belonging to ethnic minority groups, were less likely to have their time freed up during 324 working hours to access vaccination. 325

How participants perceived their personal risk of COVID-19 was a major influence on vaccination 327 decision-making. Most vaccinated participants considered themselves at risk of COVID-19 exposure, 328

i.e. because of working in frontline roles, and/or at risk of severe COVID-19 due to one or a 329 combination of factors including older age, clinical vulnerability, and being from an ethnic minority 330 background. In interviews, participants who perceived themselves to be at higher risk of COVID-19 331 elaborated that this sense of personal risk outweighed vaccine safety concerns. Several survey 332 participants felt that family members of H&SCWs should also have been offered vaccination as a 333 priority. 334

Most unvaccinated participants considered themselves to be at low risk of COVID-19. These 335 participants included younger adults and those without underlying health conditions, H&SCWs with 336 limited/no service user contact at the time of vaccine offer, and those not living with household 337 members at higher risk of COVID-19. One interviewee considered that gaining natural immunity 338 through infection was better than vaccinating for those not in higher risk groups. 339 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101/2021.04.23.21255971 doi: medRxiv preprint Vaccinated interviewees discussed COVID-19 as a severe disease and had often witnessed serious 345 and life-threatening COVID-19 symptoms and complications first-hand (e.g. through being badly 346 affected by COVID-19 themselves or seeing others with severe COVID-19). 347

Several participants that had declined vaccination considered COVID-19 severity as overplayed by 348 the government or the media (with an over-representation of cases in younger people), or 349 considered the disease severe for others (e.g. people in clinically vulnerable groups) but not for 350 themselves. These views were also based on some having had COVID-19 but only experiencing mild 351 symptoms, or not knowing people that had been affected. 352

Two interviewees (one HCW and one SCW) expressed scepticism around how COVID-19 deaths are 353 reported, questioning how many of these might be linked to non-COVID-19 causes exacerbated by 354 disruptions to NHS care. 355

Vaccination decisions were influenced by trust at multiple levels: in the vaccine, the provider, and 357 policymakers. In particular, lower levels of trust at each level were reported amongst H&SCWs from 358

Black and Asian ethnic minority groups where trust was undermined by beliefs in systemic racism 359 (e.g. racism in healthcare, in medical research, in government). SCWs also reported lower levels of 360 trust, which appeared more linked to relationships with their employer and feeling pressurised to 361 vaccinate. 362

Vaccinated and unvaccinated participants expressed concern over potential COVID-19 vaccine side 364 effects, particularly long-term effects. Many participants were concerned about the speed of vaccine 365 development and delivery, stating that COVID-19 vaccines remain in the trial phase and there is a 366 lack of long-term outcome data. Some participants reported that they or their colleagues had 367 wanted to 'wait and see' how the vaccination worked in others before getting vaccinated 368 themselves. In free-text responses, several HCWs noted that the indemnification of the Pfizer 369 vaccine (i.e. protecting Pfizer from civil lawsuits in the event of unforeseen vaccine complications) 370 also undermined trust in the vaccine. 371

Questions were raised about whether vaccines might interact with medications, how vaccination 372 might affect people who already had COVID-19, and the safety of COVID-19 vaccines and of mRNA 373 vaccines in particular. These views were expressed in interviews and free-text responses. 374 "Because I have had Long Covid and I cannot get enough information to support 375 me getting this information that taking this vaccine will not exacerbate my 376 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 symptoms and ill health, to feel safe about this choice. I feel totally lost in this 377

to make an uninformed decision and it is totally overwhelming" (Survey ID387-379

Concerns were raised that COVID-19 vaccines had not been sufficiently trialled in ethnic minority 381 groups, and therefore there was insufficient safety and effectiveness data for use in people from 382 ethnic minority backgrounds. One interviewee, of Indian ethnicity, spoke specifically about wanting 383 a particular vaccine because it had been trialled in India, and had wanted to wait for this vaccine to 384 become available before vaccinating. 385

Participants raised concerns about past medical racism (e.g. the Tuskegee syphilis study) as being 387 commonly reported amongst colleagues from Black ethnic groups. Some pointed to reports of 388 unethical medical experimentation by Pfizer in Africa in the past, and one interviewee reported that 389 staff had been given the option of accessing the AstraZeneca vaccine, which was available at a 390 different site. 391

Several participants reported that the medical racism they or others within their ethnic group had 392 experienced in the past fuelled their vaccine hesitancy, such as Black women being treated adversely 393 by health care providers. Two survey participants voiced scepticism in being asked to have data 394 recorded around ethnicity prior to vaccination, as they were concerned this might influence the 395 vaccine given to them. 396

Participants voiced their distrust in the pharmaceutical industry and scientists, drawing on examples 397 of harms caused by medicines and vaccines introduced in the past, such as swine flu vaccination 398 (linked to increased risk of narcolepsy (16, 17) , which also undermined trust in COVID-19 vaccines. 399

Several interviewees and survey participants reported that distrust in the government and its 401 handling of the pandemic had affected their own or colleagues' trust in vaccination. Amongst 402 unvaccinated interview participants, notably SCWs, questions were raised around the length of time 403 NHS services had been restricted and reduced, the changing guidance e.g. on mask wearing, and the 404 motivation for continual lockdowns. 405

Participants shared their frustrations about changes to the interval between vaccine doses, against 406 the manufacturers recommended protocol, and for some this negatively affected COVID-19 407 vaccination confidence and left them feeling vulnerable. Several participants felt the dosing interval 408 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. A small number of participants voiced some critique around the use of the collective term BAME 423 (Black, Asian, and Minority Ethnic), which does not highlight heterogeneity between or within 424 groups and may mask particular inequalities. One participant reported: 'I'm tired of people talking 425 about those BAME people without understanding that they're different groups, and without trying to 426 understand why.' (Interviewee#14 -HCW, female, Black or Black British: -Caribbean). Several survey participants were concerned that the effectiveness of COVID-19 vaccination may be 438 undermined by new variants of . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101/2021.04.23.21255971 doi: medRxiv preprint

Vaccinated participants reported that they or their colleagues viewed vaccination as important to 441 protect themselves and others, including clinically vulnerable family members, and also as a route 442 out of the pandemic and a return to 'normality' (e.g. being able to travel and visit loved ones). 443

Several vaccinated participants considered vaccination necessary for them to keep being able to 444 work. One participant also considered vaccination important to protect against newer strains of 445 Unvaccinated participants tended to consider vaccination as less important. One interview 447 participant and several survey participants felt being vaccinated would not impact on their need to 448 wear personal protective equipment or make a difference in terms of their ability to see loved ones. Participants were alarmed that their vaccination decisions were relayed to managers when health 466 related issues would usually be supported by occupational health or general practitioners only. 467

Several participants reported that managers had access to lists of unvaccinated staff and were 468 contacting them directly to discuss their vaccination status. 469

Feeling pressurised had damaging effects, eroding trust and negatively affecting relationships at 470 work. It cemented several participant's stances on declining vaccination, making them more vaccine 471 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 hesitant, and pushed one person into having the vaccine when they would have preferred not to. 472

Participants strongly felt that vaccination should be an individual's choice, informed and voluntary. 473

Several participants voiced concerns about allergies and a lack of early information on their 475 suitability for vaccination. One unvaccinated interviewee had concerns about having had a severe 476 adverse reaction to a past vaccination while another had delayed vaccination until a specific vaccine 477 became available, which they were advised was more suitable for people with allergies. 478 3.1.8 Vaccination concern in women of childbearing age 479 Two interviewees and several survey participants who wanted to become pregnant in the future 480 were concerned about unfounded rumours that COVID-19 vaccines cause fertility problems. Another 481 interviewee was concerned about the safety of COVID-19 vaccination and breastfeeding and had 482 decided to delay vaccination. Updated guidance on vaccinating during pregnancy enhanced 483 uncertainty amongst some pregnant women about vaccinating. 484

Access to information varied amongst staff, with differences reported across job roles and also by 486 ethnicity, and this affected beliefs and behaviours. Participants recommended that information 487 should be communicated regularly by employers across various platforms to reach different 488 audiences e.g. email, social media, webinars. One recommendation, voiced by several interview and 489 survey participants, was that engagement with faith groups was particularly important in 490 communicating with staff from ethnic minority backgrounds. Participants also wanted more 491 openness in communication about uncertainty, so they could make an informed decision around 492 vaccination. 493

All interviewees reported actively searching for COVID-19 vaccine information and most discussed 494 vaccination with colleagues. Participants tended to avoid social media sources, in favour of accessing 495 government or NHS sources. Many participants preferred to access research articles directly, with 496 concerns that second-hand reporting of the information (i.e. in the media) may be incomplete or 497 biased. 498

Unvaccinated participants reported that although they might receive information about vaccination, 499 they were not given the opportunity to have any discussion around vaccine questions and concerns. 500 Instead, they felt they were dismissed as being anti-vaccination and treated as stupid, something 501 they felt was exacerbated by some of the terminology around vaccination, including anti-vax and 502 myth busting. One participant voiced that Black people have historically been silenced and told not 503 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ;  to ask questions around health care, and that this was happening with COVID-19 vaccination and 504 undermining trust. 505 'I wanted to be fully informed rather than just told to take it… I was left 506 frightened and with little information which was totally avoidable.'" (Survey 507 ID350 -HCW, female, Mixed: -White and Black Caribbean) 508

Participants said that it was difficult to keep up-to-date and understand reasons for changes to the 509 COVID-19 vaccination programme e.g. updated guidance on vaccination in pregnancy. There were 510 also reports of participants being given mixed messages about vaccination. For example, two 511 interviewees that had declined vaccination had been given different information from either 512 healthcare professionals or their employer on vaccines e.g. advised that younger people should 513 access a certain vaccine. Finally, many survey participants also suggested that more transparency 514 was needed about side effects, which they felt had been downplayed. 515 were offered and accepted vaccination); however, we identified variations in vaccination offer and 531 uptake by job sector and ethnicity. 532

We found that SCWs were offered COVID-19 vaccination at a lower rate than HCWs (87.0% vs 92.7%) 534 and that SCWs were less clear about who was responsible for their COVID-19 vaccination offer; 535 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101/2021.04.23.21255971 doi: medRxiv preprint whether GP or employer. This is likely to be linked to the organisational structure and nature of roles 536 within social care. Social care can be provided by local authorities, private sector companies or 537 voluntary organisations. Funding is either paid for by the individual or by local authorities where the 538 individual cannot self -fund. Care is often provided in small care home settings or in an individual's 539 own home so the workforce is more disperse. In addition, workforce vaccination as part of the 540 seasonal influenza vaccination programme is more embedded within the NHS than social care and 541 therefore NHS organisations may be better prepared for COVID-19 vaccination delivery. The 542 effectiveness of the SCW influenza vaccination programme is arguably also undermined by 543 conflicting guidance on who should provide the vaccination. Since 2017/18 the SCW workforce has 544 been able to access seasonal influenza vaccination freely through primary care but this service is 545 'intended to complement, not replace, any established occupational health schemes that employers 546 have in place' (18). Influenza vaccination is not consistently monitored amongst SCWs and uptake 547 remains low (19). In our study, in several instances, the onus appeared to be on SCWs to organise 548 vaccination, rather than on their employers. This was also reported by interviewed non-mainstream 549 NHS HCWs, suggesting that outside of the NHS greater barriers to vaccine offer and access may be 550 experienced. 551 552 Our analysis indicated that Black African or Mixed Black African participants were being offered 553 COVID-19 vaccination at a lower rate than White British and White Irish participants (87.5% vs 554 92.1%). This appeared as a significant finding in our multivariate regression analysis (OR:2.59, 95%CI: 555 1.292 -5.177) and is difficult to explain and requires further investigation. Compared to White 556

British and White Irish participants, we also found that rates of COVID-19 vaccination decline were 557 higher amongst Black African or Mixed Black African participants and Black Caribbean or Mixed Black 558 Caribbean participants. These findings are similar to data from an NHS Trust in England, which 559 reported differential uptake by ethnicity (70.9% in White staff v 58.5% in South Asian and 36.8% in 560 Black staff; p<.001) (7), and are also supported by international data showing lower intention to take 561 up COVID-19 vaccination among healthcare workers from certain ethnic groups (20). Comparisons 562 by ethnicity is further demonstrated by the interaction between ethnicity and COVID-19 vaccine 563 beliefs, showing for example, less confidence in vaccine safety and importance amongst Black 564

H&SCWs. 565 566 Through free-text responses and interviews we found that factors influencing vaccination-decisions 567 were multi-layered and often involved the weighing up of perceived risks and benefits of 568 vaccination. Factors found to influence COVID-19 vaccine uptake amongst H&SCWs included 569 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 perceptions around COVID-19 risk and COVID-19 severity, beliefs around vaccine effectiveness, 570 safety and importance, concerns about allergies, concerns about safety in women of childbearing 571 age, and religious beliefs. Interviewees discussed that perceptions around personal COVID-19 risk 572 and the severity of COVID-19 were central to decision-making, and younger participants and/or 573 those without underlying health conditions did not necessarily see the rationale for vaccinating, 574 particularly without evidence on the link between vaccination and reducing disease transmission. 575

Concerns around safety -especially in light of the speed of development of the new vaccines -also 576 accord with findings from attitudinal work conducted internationally (21). 577 578 Distrust, reported in the vaccine, the provider and the policymaker, was also central to vaccination 579 decisions. Distrust was particularly expressed by ethnic minority groups, who emphasised how they 580 have borne the brunt of health inequalities before and during the pandemic. Black participants in 581 particular felt stigmatised and patronised, and unable to voice questions and concerns, and obtain 582 responses. This is consistent with existing evidence (22) and a recent PHE report (23). Participants 583 criticised the media characterisation of ethnic minority groups as anti-vax, which fails to recognise 584 underlying reasons for lower vaccine confidence. 585 586 Organisational factors and workplace culture played an important role in the likelihood of both being 587 offered and getting vaccinated, as described elsewhere (24). This is consistent with previous 588 research that indicates that an organisational culture framing seasonal influenza vaccination within a 589 broader staff wellbeing programme was conducive to higher uptake (24, 25) . The coherence 590 between staff well-being and COVID-19 vaccination might be particularly important in a context 591 where staff have been subjected to high levels of stress over a long period (26). Importantly, the 592 survey revealed that participants that reported greater agreement with the statement 'I feel/felt 593 under pressure from my employer to get a COVID-19 vaccine' were significantly more likely to decline 594 the vaccine even after demographic factors were controlled for. Interviews suggest placing staff 595 under pressure to vaccinate may increase intention to refuse the vaccine. This was particularly 596 evident in SCWs, for whom pressure was exacerbated by hearing of care sector employers making 597 COVID-19 vaccination mandatory for staff, and the vulnerability of SCW positions (e.g. employment 598 on zero-hours contracts). This is consistent with previous research that shows promoting a positive 599 choice around the seasonal influenza vaccine rather than resorting to a more coercive approach, 600 such as mandating, was supportive of fostering continuous improvement of vaccination uptake (24, 601 27). 602 603 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 25, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 

19-vaccination-advice-from-the-jcvi-2-december-2020. 688 2. Public Health England

COVID-19 weekly announced vaccinations 15 699

Seasonal flu vaccine uptake in healthcare workers 2020 to 701 2021: provisional monthly data for 1

Risk of hospital 705 admission with coronavirus disease 2019 in healthcare workers and their households: nationwide 706 linkage cohort study

COVID-19: review of disparities in risks and outcomes

Demographic and 711 occupational determinants of anti-SARS-CoV-2 IgG seropositivity in hospital staff

Differential 714 occupational risks to healthcare workers from SARS-CoV-2 observed during a prospective 715 observational study. Elife. 9

Exclusive: deaths of NHS staff from covid-19 analysed

Association of demographic and occupational factors with SARS-CoV-719 2 vaccine uptake in a multi-ethnic UK healthcare workforce: a rapid real world analysis

Trends and clinical characteristics of COVID-19 722 vaccine recipients: a federated analysis of 57.9 million patients' primary care records in situ using 723 OpenSAFELY

Conceptualization and measurement of ethnic identity: 725 Current status and future directions

Using thematic analysis in psychology

Incidence of narcolepsy 729 after H1N1 influenza and vaccinations: Systematic review and meta-analysis

Risk of Narcolepsy 732 after AS03 Adjuvanted Pandemic A/H1N1 2009 Influenza Vaccine in Adults: A Case-Coverage Study 733 in England

Direct Enhanced Service Specification: Seasonal 735 influenza and pneumococcal polysaccharide vaccination programme 2020/21. 2020

Concerns over low uptake of flu vaccination in social care

COVID-19 Vaccine Acceptance among Health Care Workers in the 739 United States

Attitudes of 741 healthcare workers towards COVID-19 vaccination: a survey in France and French-speaking parts of 742 Belgium and Canada

Beyond the data: Understanding the impact of COVID-19 on 746 BAME groups

Organisational factors affecting performance in delivering influenza vaccination to staff in NHS Acute 752 Hospital Trusts in England: A qualitative study

Independent Expert Advisory Group for Health 754

Factors and considerations for establishing and improving seasonal 755 influenza vaccination of health workers: Report from a WHO meeting

What should be 758 done to support the mental health of healthcare staff treating COVID-19 patients?

Is flu vaccination opt-out feasible? 761 Evidence from vaccination programme implementers and managers in the English National Health 762 Service

Association of demographic and occupational factors with SARS-CoV-2 767 vaccine uptake in a multi-ethnic UK healthcare workforce: a rapid real-world analysis