Research and analysis

Second quarterly report on progress to address COVID-19 health inequalities

Updated 1 March 2021

This was published under the 2019 to 2022 Johnson Conservative government

1. Introduction

In June 2020, the Prime Minister and the Secretary of State for Health and Social Care asked the Minister for Equalities, Kemi Badenoch MP, to lead cross-government work following publication of the Public Health England (PHE) report COVID-19: review of disparities in risks and outcomes.^{[footnote 1]}

The Minister for Equalities published her first progress report on 22 October. This is the second report in this year-long project, which has been submitted to the Prime Minister and the Secretary of State for Health and Social Care in line with the terms of reference.^{[footnote 2]}

The first report concluded that a range of socioeconomic and geographical factors coupled with pre-existing health conditions were contributing to the higher infection and mortality rates for ethnic minority groups, with a part of the excess risk remaining unexplained for some groups.

This second report looks at those causes in more detail and sets out some of the work undertaken to fill the gaps in our understanding and to mitigate the risks of COVID-19 infection. It sets out the progress made under the terms of reference and in implementing the recommendations from the first report (set out in Annex A).

The focus of this work has been the disproportionate impact COVID-19 has had on ethnic minorities. There is wider work underway across government to consider the impact the virus has had on other groups, such as disabled people.

Race Disparity Unit, Cabinet Office
February 2021

2. Executive summary and next steps

2.1 Summary

In June 2020, the Prime Minister and Secretary of State for Health and Social Care commissioned the Minister for Equalities, with support from the Cabinet Office Race Disparity Unit (RDU), to take forward work on the disproportionate impact of COVID-19 on ethnic minority groups.

This commission included developing the understanding of the drivers of disparities in infection and death rates of COVID-19, reviewing the effectiveness of current actions being undertaken by the government to lessen disparities, and modifying or developing policy where needed. This report details the work undertaken across government since the minister’s first quarterly report was published on 22 October.

It is clear that government departments have made significant efforts to address disparities in outcomes from COVID-19. In order to improve public health communications with those communities most at risk from COVID-19, the government released £23.75 million in funding to local authorities last month under the Community Champions scheme, following an expressions of interest exercise.

This funding is enabling local authorities to work with grassroots advocates to tailor public health communications and to use trusted local voices to promote healthy living, encourage vaccine uptake and counter misinformation. The government will monitor the impact of the scheme and share the findings with other local authorities.

To improve our understanding of the health, social, cultural and economic impacts of COVID-19, the government has just invested a further £4.5 million of funding in new research projects looking at ethnic minority groups. In order to prevent the spread of the virus and to protect frontline workers, many of whom are from an ethnic minority background, the Department for Transport and the Department for Health and Social Care included transportation workers in mass testing pilots covering the Christmas travel period. These are now being rolled out more widely.

The government also successfully piloted community-led, localised, asymptomatic testing at places of worship in ethnically diverse areas, building trust within the community and enabling a higher number of positive cases to be detected.

Efforts need to continue into the next quarter, driven by the latest data and evidence and focused on those most affected by the second wave of the virus.

A light-touch review of local authority actions has shown that there are also considerable efforts underway at a local level to address COVID-19 disparities for ethnic minority groups, led by local authorities and Directors of Public Health, and using trusted voices in the community. For example, Birmingham City Council has established 645 Champions across the 69 wards in the City, and has established a system of 19 community partners to support wider dissemination of accessible information and engagement with specific minority communities. Other examples include utilising local data and insights on the communities that have been disproportionately impacted by COVID-19 to tailor and target messaging.

This exercise has shown that some areas with strong existing links to community groups, and with access to relevant and up-to-date data, have fared well in supporting those communities who have been disproportionately impacted by COVID-19. However, some areas have cited difficulties around adapting to the virtual delivery of services, and engaging communities in a COVID-19 secure manner. The government will share with local authorities examples of good practice and solutions to overcome some of the barriers identified by the review.

This report sets out our increased understanding of the drivers of these disparities. In particular, the impact of COVID-19 on certain ethnic minority groups has changed between the first wave and the early second wave; changes within such a short time period strongly suggest that ethnic inequalities in COVID-19 outcomes are driven by risk of infection, as opposed to ethnicity itself being a risk factor for severe illness or death from COVID-19.

The direct impacts of COVID-19 improved for ethnic minorities as a whole during the early second wave. For example, in the first wave, Black African men were 4.5 times more likely to die from COVID-19 than White British men of the same age but in the early part of the second wave the risk of death was the same for Black African and White British men. At the same time, however, the second wave has had a much greater impact on some South Asian groups. Work is underway to consider why the second wave to date has had such a disproportionate impact on Pakistani and Bangladeshi groups. Relevant considerations include regional patterns in first and second waves of the virus, household occupancy and multigenerational households, deprivation, and occupational exposure.

These findings strengthen the argument that ethnic minorities should not be considered a single group that faces similar risk factors in relation to COVID-19. Different ethnic groups have experienced different outcomes during both waves of the virus.

This report also summarises the findings of the research commissioned by the RDU into a small group of ethnic minority people’s personal experiences of COVID-19. Some important themes emerged from this work. For example, participants felt that communications tended to frame ethnic minorities as a homogeneous group that is vulnerable to COVID-19, which they found stigmatising. The research also showed the challenges some participants had in navigating public health advice and applying it to everyday situations, as well as adapting to the pace of change with the guidance. These insights will be shared with other government departments to improve policy-making.

The data also shows that deprivation continues to be a major driver of the disparities in COVID-19 infection rates for all ethnic groups and this will be a particular focus of government work in the third quarter.

A significant development since the first quarterly report is the approval and roll out of COVID-19 vaccines. This report summarises how the vaccination programme is being prioritised and the implications of this for ethnic minority groups, as well as the analysis of likely take up rates for these groups. In addition, on 13 February the government published its UK COVID-19 vaccine uptake plan.^{[footnote 3]} This highlights a range of barriers to uptake and some of the work taking place across government and at a local level to minimise the impact of these.

Data from the UK Household Longitudinal study, the Office for National Statistics (ONS), and REACT-2 suggest lower levels of vaccine uptake among some ethnic minority groups. The government has put in place a programme of work to understand and address this. This includes establishing NHS vaccination centres in suitable sites in the community, such as places of worship.

The report also sets out the measures the government has taken through communications campaigns to encourage uptake of vaccines among ethnic minority groups and to counter misinformation, both nationally and locally. This has included use of digital advertising on Facebook and Twitter, traditional media, social media posts featuring well-known medical professionals and disseminating important messaging to hundreds of local contacts, such as faith leaders. These efforts will continue over the coming weeks, linking in with the new Community Champions to target those areas most at risk.

In addition, over the last quarter the Minister for Equalities and the RDU have engaged within government and externally to share the findings from the first report and any new data, to encourage development of new government interventions to tackle COVID-19 disparities for ethnic minority groups and to identify any barriers and solutions to vaccine uptake. This includes holding roundtables with those representing the adult social care sector, independent pharmacists and South Asian groups, and further work to address maternal health disparities for ethic minority women. This engagement work will continue in the next quarter, with a particular focus on promoting vaccine uptake.

2.2 Next steps

This report sets out the following next steps:

Central and local government interventions

MHCLG to share with local authorities examples of good practice from the review of local authority activity.

MHCLG to share with local authorities the findings from the initial, one-month review of returns from Community Champions.

Vaccination programme

Minister for Equalities to write to the Joint Committee on Vaccination and Immunisation (JCVI) summarising the latest data and evidence set out in this report, to inform future advice on vaccine prioritisation.

The government will continue to monitor data on vaccine uptake among ethnic minority groups and, if necessary, take further steps to address any barriers among these groups.

Data and evidence

The RDU will share the findings from the qualitative research into people’s personal experiences of COVID-19 across government, particularly in relation to the stigmatisation felt by a number of participants in relation to being singled out as ‘BAME’.

Departments and other agencies should publish a statement on GOV.UK outlining their plans to move their data collections to the Government Statistical Service’s (GSS) harmonised ethnicity data standard. Harmonisation is hugely important as it allows analysts to gain deeper insight and value from data.

NHSE/I, working with DHSC and others, should publish a quarterly report on progress in improving the recording of ethnicity in health care records.

Departments should provide updated datasets on COVID-19 risk factors and secondary impacts for publication on the Ethnicity facts and figures website^{[footnote 4]} in line with the schedule in Annex C. This provides transparency of process to users - promoting trust and authority - as well as informing them when the most up-to-date data will be made available.

Engagement

The Minister for Equalities, the government advisers on COVID-19 and ethnicity, and the RDU will continue a programme of engagement over the next quarter. This will include work to promote vaccine uptake, alongside the engagement led by the Minister for COVID-19 Vaccine Deployment.

Communications

The government will continue to tailor its communications strategy on vaccine roll out to reflect the latest evidence on vaccine uptake among ethnic minority groups.

The government will work closely with the new Community Champions to disseminate important public health messages, promote uptake of vaccine and tackle misinformation

Government communications will reflect the findings of the qualitative research into people’s personal experiences of COVID-19 and will ensure that ethnic minorities are not treated as a single group and that public health messaging is not stigmatising.

3. Measures to address COVID-19 disparities

3.1 Summary

The first quarterly report recorded the actions government departments and their agencies had implemented in the early stages of the pandemic to mitigate impacts of COVID-19. This section updates on progress with some of these initiatives, highlights new measures and summarises the outcome of the light-touch review of local authority actions.

This section also provides an update on progress departments have made in creating appropriate systems for monitoring the impacts their policies are having.

3.2 Approach and results: central government

The RDU has continued to work closely with other government departments and agencies, and in particular the Department of Health and Social Care (DHSC), Public Health England (PHE), the wider Cabinet Office and the Ministry of Housing, Communities and Local Government (MHCLG), to assess current initiatives to reduce COVID-19 disparities amongst ethnic minority communities. The RDU has also worked closely with the Disability Unit, also located within the Cabinet Office Equality Hub, which is leading similar work in relation to the impact of COVID-19 on disabled people.

The RDU established a cross-Whitehall working group in November in order to share the latest evidence from the review to inform policy development. The RDU commissioned updates through this group and a letter from the Minister for Equalities to relevant ministerial colleagues on 18 December 2020. A summary of these returns is at Annex B.

Overall, departmental returns have highlighted the ongoing commitment to tackle the disparities through a number of means, from guidance to relevant industries to establishing funds to support disproportionately impacted groups, including ethnic minority groups.

Some important initiatives include:

• MHCLG has released £23.75 million in funding to local authorities under the Community Champions scheme (see next section).
• DHSC and MHCLG collaborated to run pilots of community-led, localised, asymptomatic testing at places of worship in ethnically diverse areas such as Brent and Wolverhampton with the aim of removing some of the main identified barriers to engaging with Test and Trace, including trust and access. These have reported significant success. For example, a pilot at a Gurdwara in Wolverhampton, which ran during a Sikh religious festival, used local volunteers to open up testing to a cohort of people who might not otherwise have engaged. This drove up testing rates, with almost 3,000 kits registered which showed a much higher prevalence rate (5.3%) than other pilot sites. This enabled a higher number of COVID-19 cases to be detected and the pilot was extended at the request of religious leaders and the local council. Further pilots are planned.
• NHS England is using the Al-Abbas Islamic Centre in Balsall Heath, Birmingham, as a vaccination centre, which is helping to build trust within the local community and encourage vaccine uptake among groups more at risk from COVID-19 and whom the data show are more reluctant about being vaccinated.
• DfT and DHSC worked together to secure the inclusion of transportation workers, a significant proportion of whom are from an ethic minority background, in mass testing pilots covering the Christmas travel period. These are now being rolled out more widely.
• Providing additional funding to the Public Health England Better Health Campaign in order to target Black African, Black Caribbean, Indian, Bangladeshi and Pakistani ethnic groups. This campaign, which runs until March, aims to reduce obesity and other comorbidities within these groups, which are associated with worse COVID-19 outcomes.
• Following engagement with trade associations, DfT issued new guidance to private hire vehicle (PHV) and taxi drivers in November (and updated in January) about how to protect themselves from COVID-19. 53% of such drivers are from an ethnic minority group and 98% are men.
• UK Research and Innovation has invested a further £4.5 million in funding for 4 new research projects looking at the health, social, cultural and economic impacts of COVID-19 on ethnic minority groups.^{[footnote 5]}
• The RDU continues to assist the DHSC and NHSE in improving outcomes for pregnant women from ethnic minority backgrounds during the pandemic. This remains a priority due to the pre-existing health inequalities in this area, and a commitment to ensure that COVID-19 does not exacerbate these (see section 3).

While good progress has been made to address COVID-19 disparities, government departments must redouble their efforts, taking account of the latest available data and evidence. In particular, departments must consider measures that will benefit those most affected by the second wave of the virus, and in particular those from the Bangladeshi and Pakistani ethnic groups.

Community Champions

The Community Champions scheme was announced on 22 October to enhance existing communication strategies in a target group of councils and to fund work with grassroots advocates from those communities most at risk from COVID-19. The scheme is backed by up to £25 million in government funding.

MHCLG, which is sponsoring the scheme, developed a longlist of 65 local authority areas with larger proportions of at-risk communities and entrenched community transmission of COVID-19, using a variety of data sources. Expressions of interest have been received from 60 of these local authorities, summarising the measures they have already implemented to reduce COVID-19 disparities and how additional funds would be used to enhance and expand their interventions. MHCLG evaluated these and released funding in January.

All funded local authorities are expected to implement the following:

• A community connection, outreach and engagement strategy, with a steering group, involving local community leaders, public health providers, voluntary groups, and organisations in each area.
• Community Champions, including from within the disproportionately impacted ethnic minority and disabled groups or local hard to reach communities, to undertake one-to-one support, build trust and counter misinformation.
• Engagement activities with residents from disproportionately impacted ethnic minority and disabled groups in greatest need of local support.
• Creation and delivery of practical sustainable tools to increase outreach, communication and engagement which are tailored, appealing, visual and multi-language messages to reach diverse populations, who may be digitally excluded, and mobilise local communities.

By implementing the scheme in this way, community leaders and voluntary and community groups who specialise in working with the communities most at risk from COVID-19 will be able to deliver appropriate communications to their communities. This will increase personalised and localised communications in relation to COVID-19.

To reach disproportionately impacted communities beyond these 60 areas, MHCLG has funded 2 national voluntary, community and social enterprise (VCSE) partners who have a proven track record of engaging harder to reach communities to carry out communications and engagement activities, including health promotion and encouraging vaccine uptake.

MHCLG will monitor the success of the Community Champions scheme by requiring returns at one, 3 and 6 month intervals and holding regular meetings with funded partners.

3.3 Approach and results: local government

In the first quarterly progress report, the Minister for Equalities committed to a rapid, light-touch review of actions taken by local authorities and Directors of Public Health to support people from ethnic minority backgrounds, in order to understand what works at a local level. To minimise the burden on local authorities at this particularly challenging time, the RDU worked with MHCLG and DHSC to develop an approach that provided rich data on actions undertaken to date, but without imposing reporting burdens.

RDU conducted this rapid review by focusing on local authority areas, identified by MHCLG under the Community Champions applications process, with larger proportions of at-risk communities and entrenched community transmission of COVID-19, using a variety of data sources. The longlist of local authority areas was developed using DHSC/PHE data on COVID-19 incidence (the data used for assessing tiers) alongside social integration data (higher residential segregation and lower English language skills) and higher levels of disability to identify areas with larger proportions of at-risk communities and entrenched community transmission of COVID-19. Each local authority was asked to provide examples of the actions it had taken to minimise the disparities, such as promoting public health messaging. This enabled a rapid review of local authorities where COVID-19 disparities remain particularly pertinent, whilst capturing the broad range of ideas and practices across a variety of geographical locations.

It is clear that there is a huge effort underway to break down barriers at a local level. Some common themes emerging from this work include:

• utilising existing partnerships and networks, including faith organisations and the voluntary community sector
• providing targeted communications messaging – this is being undertaken through a number of means including translating important messages, distributing information in alternative locations and formats, and providing messaging that is appropriate for specific community events and activities that have a higher proportion of ethnic minority participation
• collecting, mapping and utilising data and insights on the communities within their area that have been disproportionately impacted by COVID-19

This exercise has shown that those areas with strong existing links to community groups, and those with access to significant quantities of high quality, relevant and up-to-date data, have fared better in supporting those communities who have been disproportionately impacted by COVID-19. Directors of Public Health have provided similar views, indicating for example that health messages on vaccines have been best received when delivered by individuals respected by the communities they are seeking to influence. This includes trusted health professionals and leaders in the faith communities.

At the same time, some areas cited difficulties around adapting to the challenges of COVID-19 such as through the virtual delivery of services. This was reported as particularly challenging when delivering in partnership with third sector organisations that had previously relied on a physical presence within the community, and whose members did not necessarily have the technological skills and capability to adapt to the changing circumstances. Some have also reported difficulty in engaging directly with communities in a COVID-19 secure manner, such as through online services, where these groups have traditionally relied upon face-to-face council and third sector engagement. The reasons for this prior reliance on face-to-face engagement are multifaceted, but some reasons cited were: low levels of English language within the community, lack of access to suitable technology, and reluctance to engage with the council directly.

The RDU will work with MHCLG to draw up a list of examples of good local authority practice and suggestions on how to overcome some of the barriers identified in this review and will share these widely. MHCLG will also share feedback from the regular reviews of progress with the Community Champions scheme.

Examples of local government actions

Birmingham City Council has established 645 Champions across the 69 wards in the City, and has established a system of 19 community partners to support wider dissemination of accessible information and engagement with specific minority communities. This has ensured that information is accessible to those with language barriers. The Council has also developed social media campaigns on TikTok and Instagram for young people.

Bassetlaw has utilised local partnerships with the District Council, community and voluntary service and health to maintain community engagement through distanced engagement methods and new Facebook Live events, as well as undertaking several thousand ‘safe and well’ calls to vulnerable residents.

Hackney Council is currently running 2 one-year programmes to proactively engage and communicate with residents who are disproportionately affected by COVID-19. This includes a Public Health Community Champions programme, which currently works with 75 Champions who have been trained in ‘Making Every Contact Count’ and given messages about COVID-19, which has provided valuable insight about how to adapt and adjust public health messaging to maximise reach.

Newcastle City Council has a programme of work to address COVID-19 disparities including creating an ethnic minorities Community Leaders WhatsApp group to allow delivery of timely messages and enable a 2-way dialogue, and producing COVID-19 message videos in Bengali, Urdu, Hindi, Arabic, Romanian, Czech and Romani, which reached 106,000 people.

Oldham council has a community engagement network that works closely with community leaders to gather insight and engage with communities. It uses targeted door-to-door engagement led by intelligence about hotspots of COVID-19 cases, and targeted communications using insight from the community engagement network (such as broadcasting messages through Mosque radio and creating accessible graphics distributed via WhatsApp).

Stoke on Trent has facilitated community groups to provide practical support and ensure community venues are COVID-secure. To date over £215,000 has been invested in such groups.

Wirral has utilised ‘connectors’ within its local communities, allowing the council to react and respond to local needs. The connectors undertake a range of activities including wellbeing calls, delivering food, supporting, local contact tracing and delivering leaflets. More importantly they act as the voice of the community, allowing the Council to understand people’s fears and behaviours and to tailor its communications accordingly.

3.4 Individual risk and the clinically extremely vulnerable

The last report summarised work to develop a new COVID-19 predictive risk model (“QCOVID”). This has been developed by an expert group commissioned by the Chief Medical Officer and takes into account a wide range of factors now known to increase risk of infection and serious outcomes from COVID-19. The associated research was peer-reviewed and published in the British Medical Journal in October 2020, approved by the Medicines and Healthcare products Regulatory Agency (MHRA) in December and independently validated by the ONS in January 2021.

DHSC is working at pace to apply the model in the NHS, incorporating the feedback from private testing of a clinical decision support tool. The RDU (and wider Cabinet Office Equality Hub) has fed into this work. The clinical tool was made available across primary and secondary care as a secure public beta webtool from 16 February.

Through providing detailed data about clinical risk, the QCOVID model has enabled DHSC to incorporate the findings from the research into national policy and has used it to identify a new cohort of patients at equivalent risk to the Clinically Extremely Vulnerable. This group is being added to the Shielded Patient List as a precautionary measure, and is entitled to advice and support, including priority access to the COVID-19 vaccine if they have not already been offered it. The Joint Committee for Vaccination and Immunisation (JCVI) has also reviewed the underlying data from the QCOVID model in shaping its advice on COVID-19 vaccine prioritisation.

Future plans include exploring embedding the clinical tool in GP systems, and the potential development of a citizen-facing tool. As more is learnt about why the threat posed by COVID-19 varies across the population, the QCOVID risk model will be updated with the latest evidence and individuals will be given more nuanced advice on risk.

3.5 COVID-19 vaccinations

The JCVI is an independent expert committee which provides expert scientific advice on prioritisation of COVID-19 vaccines. In its interim advice published on 2 December^{[footnote 6]}, updated on 30 December, the JCVI advised that for Phase 1, the vaccine deployment programme should prioritise care home residents and their carers, followed by people over 80 and health and social care workers, then to the rest of the population over 50 years of age in order of age and clinical risk factors. This is because the current evidence strongly indicates that the single greatest risk of mortality from COVID-19 is increasing age and that the risk increases exponentially with age. Prioritising in this way will protect those most at risk of morbidity and mortality, and to protect the NHS. Phase 2 will follow, covering adults under 50 who are not health or social care workers, clinically extremely vulnerable or who have underlying health conditions.

The JCVI, which will keep its advice under review, found that there was evidence that certain ethnic minority groups have higher rates of infection, and higher rates of serious disease, morbidity and mortality. It concluded that: “There is no strong evidence that ethnicity by itself (or genetics) is the sole explanation for observed differences in rates of severe illness and deaths. What is clear is that certain health conditions are associated with increased risk of serious disease, and these health conditions are often overrepresented in certain [ethnic minority] groups. It is also clear that societal factors, such as occupation, household size, deprivation, and access to healthcare can increase susceptibility to COVID-19 and worsen outcomes following infection. These factors are playing a large role in the inequalities being seen with COVID-19”.

The JCVI concluded that good vaccine coverage in ethnic minority groups will be the most important factor within a vaccine programme in reducing disparities in outcomes for these groups. It added that prioritisation of persons with underlying health conditions will also provide for greater vaccination of ethnic minority groups, who are disproportionately affected by such conditions.

The JCVI recognises that tailored local implementation to promote good vaccine coverage in ethnic minority groups will be the most important factor within a vaccine programme in reducing health inequalities in these groups. The NHS, PHE and DHSC will provide advice and information at every possible opportunity, including working closely with ethnic minority communities, to support those receiving a vaccine and to anyone who has questions about the vaccination process.

The JCVI’s advice also highlights the need for good vaccine uptake and coverage in ethnic minority groups as the most important factor in reducing the inequalities in infection rates and outcomes. The Minister for Equalities is working closely with the Minister for COVID-19 Vaccine Deployment to encourage take up (see section 4 for further detail). The Minister for Equalities is also personally participating in the Novavax clinical vaccine trial at Guys’ and St Thomas’ and has encouraged other ethnic minority MPs to do the same in order to break down barriers and to help increase the very low participation levels of ethnic minorities in clinical research.

The Minister for Equalities wrote to the JCVI in December to share the main findings from the first quarterly report. She will write again to the Committee shortly, summarising the evidence gathered in this report.

In addition, NHS England has established a Vaccine Equalities Committee, bringing together government departments with national representatives from the Association of Directors of Public Health, Local Authorities, Fire and Police services and third sector organisations to advise and guide the vaccine deployment programme on addressing inequalities. This Committee is considering data on vaccine uptake by different ethnic minority groups, and supporting targeted work to address concerns about uptake within particular communities.

3.6 Next steps

MHCLG to share with local authorities examples of good practice from the review of local authority activity.

MHCLG to share with local authorities findings from the initial, one-month review of returns from Community Champions.

Minister for Equalities to write to the JCVI summarising the latest data and evidence set out in this report, to inform future advice on vaccine prioritisation.

The government will continue to monitor data on vaccine uptake among ethnic minority groups and, if necessary, take further steps to address any barriers among these groups.

4. Data and evidence

4.1 Summary

The disproportionate impact on ethnic minorities - apparent during the first wave and continuing for some ethnic groups during the second wave to date - is largely a result of higher infection rates for some ethnic groups.

Based on critical care data, deaths in hospital data, mortality rates and hazard ratios analysis from ONS and OpenSAFELY, the direct outcomes of COVID-19 have improved for all ethnic minority groups between the first and early second waves of the virus, except people from South Asian backgrounds, in particular Bangladeshi and Pakistani groups. For these particular groups, the risk of dying from COVID-19 compared with White British increased between the first and second waves, possibly driven by high infection rates in the early second wave.

General health inequalities are complex; while all are entitled to access primary care services, people from ethnic minority groups are more likely to report poorer experiences using health services. Prior to the pandemic, overall mortality (adjusted for age) was lower in most ethnic minority groups but COVID-19 has reversed the mortality advantage in some ethnic minority groups.

It is vital to understand the differences between ethnic groups, rather than using “BAME” as an aggregate, and to consider the circumstances that may have contributed to health inequalities from COVID-19, as different ethnic groups experience different outcomes.

Uptake of previous national vaccination programmes has been lower in Black African and Black Caribbean groups and REACT-2 data suggests that while likelihood to accept has increased, Black adults are still least likely to get the COVID-19 vaccine. OpenSAFELY analysis has found that vaccine uptake is lower in Black and South Asian over 80s.

Some initial analysis of the impact of COVID-19 on disabled people is set out in Annex C. Further analysis and research is planned over the coming months.

A Statutory Instrument and accompanying Directions were laid in December which make changes to the regulations governing GP contracts. When ethnicity data is provided by the patient, the GP is now mandated to record that information in general practice.

There is further quality review work ongoing: the harmonisation of ethnicity data, including the UISPC review of NHS classifications; and other reviews suggesting how the quality of ethnicity data in health datasets such as Hospital Episode Statistics might be improved.

A publication schedule of updates to datasets on the RDU’s ‘Ethnicity facts and figures’ website has been created, to facilitate the timely provision of data related to COVID-19 risk factors and secondary impacts.

4.2 Approach

The first quarterly report concluded that a range of socioeconomic and geographical factors coupled with pre-existing health conditions were contributing to the higher infection and mortality rates for ethnic minority groups, with a part of the excess risk remaining unexplained for some groups. During the second quarter, the RDU has worked with ONS, other government departments, academics and the SAGE ethnicity subgroup to get a better understanding of these drivers and to address, where possible, the gaps in our understanding.

Work undertaken following the first report includes:

• Commissioning analysis by PHE of the relationship between ethnicity, pre-existing health conditions and COVID-19 infection and mortality.
• Working closely with ONS analysts, who have published more data and analysis on infections^{[footnote 7]}, the social and economic impacts of COVID-19^{[footnote 8]}, the relationship between occupation, ethnicity and COVID-19, and a plain English explanation of the disparities^{[footnote 9]}. ONS has also introduced a regularly updated COVID-19 Dashboard^{[footnote 10]} (including a section about ethnicity), which provides an easily-accessible and up-to-date summary of the main statistics about COVID-19.
• Supporting research by helping academics to access data and information vital for their research and meeting those leading the NIHR-funded projects.
• Commissioning Policy Lab to undertake a study of ethnic minority people’s lived experience, building on the wider stakeholder engagement in the PHE ‘Beyond the Data’ report.
• Advising ONS colleagues in their development of a larger scale qualitative investigation into the compliance behaviours of different population groups, including ethnic minority communities.

4.3 Results

Drivers of disparities for ethnic minority groups

Previous reports revealed^{[footnote 11]} that people from ethnic minority groups faced a disproportionate impact from COVID-19 during the first wave of the pandemic in the UK. However, recent analysis of the first wave from PHE^{[footnote 12]} and evidence from the second wave of COVID-19 intensive care admissions in the UK^{[footnote 13]} has shown that some ethnic minority groups did not face increased risk of death from COVID-19 once infected.

According to Public Health England^{[footnote 14]}, comparing survival rates (once tested positive) with those of people from the White group:

• people belonging to the Mixed and Other ethnic (aggregate) groups, and the Black African, Black Caribbean and Asian Other (detailed) groups did not have poorer survival rates than White people. The high death rates reported previously by ONS for these groups are therefore likely to be largely determined by a high risk of getting COVID-19 rather than a higher risk of dying once infected.
• Bangladeshi, Chinese, Pakistani, Black Other and Indian ethnic groups had an increased risk of death. This difference is reduced with longer follow up after death and therefore PHE suggests that the analysis be repeated with a longer follow-up period in order to draw more reliable conclusions about poorer survival in some ethnic groups.

Figure 1: Adjusted odds^{[footnote 15]} of dying after getting infected with COVID-19 by detailed ethnic group compared with White ethnic group, positive tests occurring between 20 March and 13 July 2020

source: Public Health England, Note: adjusted for age, sex, socioeconomic characteristics and pre-existing conditions.

This is consistent with research from the International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC), which found that, compared with the White group and accounting for age, sex and geography, South Asian people had a higher risk of mortality (hazard ratio (HR) 1.19) but this was not the case for East Asian, Black or Other ethnic groups (HRs 1.00, 1.05, 0.99 respectively).^{[footnote 16]}

Risk factors for being critically ill or dying from COVID-19 include factors such as age, sex, pre-existing health conditions and disabilities. Further information about these risk factors is shown in Annex C.

The disproportionate impact on ethnic minorities - apparent during the first wave and continuing for some ethnic groups during the second wave to date - is largely a result of higher infection rates for some ethnic groups.

Ethnicity itself is not a risk factor for infection but people from ethnic minority groups are more likely to experience various risk factors for infection. SAGE reported that “Modifiable social factors such as poverty and occupation make a large contribution to the greater burden of COVID-19 in ethnic minorities”.^{[footnote 17]} Infection risk factors include geography, age, deprivation, overcrowding, multigenerational households, certain occupations (in particular those that are public-facing) and lifestyle factors.

In addition, the interaction between risk factors needs to be considered, where the impact to risk could be additive or multiplicative, as many of the risk factors are interrelated. If a household at increased risk of infection includes people with disabilities, pre-existing health conditions including obesity, or older people, there is also an increased risk of mortality once infected.

The likelihood of experiencing different risk factors varies by ethnic group, just as infection rates and mortality rates vary by ethnic group.

Finally, as different ethnic groups experience different outcomes, it is not analytically useful to aggregate all ethnic minority groups under a heading of “BAME” or to draw conclusions about COVID-19 outcomes by looking only at aggregated data for all ethnic minorities. It is vital to understand the differences between ethnic groups and to consider the circumstances that may have contributed to health inequalities from COVID-19.

According to a study from academics affiliated to St George’s University of London^{[footnote 18]}, migrants in high-income countries are at high risk of exposure to, and infection with, COVID-19. In general, migrants have higher levels of many risk factors and vulnerabilities relevant to COVID-19, including increased exposure due to high-risk occupations and overcrowded accommodation, and barriers to health care including inadequate information, language barriers, and reduced entitlement to healthcare coverage related to their immigration status.

This study had limited evidence for the UK. However, academics from University College London (UCL) are preparing an analysis using the Million Migrants study^{[footnote 19]} to understand the relationship between migration and COVID-19 health outcomes. Further details of this analysis will be reported when the analysis is complete.

Differences between the first wave and the early second wave

Overall the direct outcomes of COVID-19 have improved for all ethnic minority groups except people from South Asian background, in particular the Bangladeshi and Pakistani ethnic groups, between the first and early second waves of the virus to date, according to evidence from the Intensive Care National Audit and Research Centre (ICNARC), NHS England, ONS and OpenSAFELY.

For the purpose of this analysis, the first wave is deemed to be the period up to 31 August, and the early second wave is the period from 1 September to 28 December. Recent PHE data on mortality for January 2021, the approximate peak of the second wave, reveals higher mortality rates for the Black ethnic group than research on the early second wave suggests.^{[footnote 20]} However, these data do not show a detailed ethnicity breakdown and because there has not been any analysis of hazard ratios, it is difficult to come to conclusions about specific ethnic groups such as Black African and Black Caribbean.

Research by ONS analysts and academics uses both age standardised mortality rates (ASMRs) and Hazard ratios (HRs) to explore risk of death.^{[footnote 21]} ASMRs allow for comparison of mortality between ethnic groups, or over time. It is worth bearing in mind that the differences in ASMRs between the first and second wave may be impacted by the different lengths of the 2 waves (see Annex C); in a longer time period more deaths may have occurred. Hazard ratios quantify the risk of death from COVID-19 for ethnic minorities, relative to the risk experienced by White British people. This supports exploration of the role played by different risk factors in the excess risk in wave 1 and wave 2 faced by different groups. ASMRs measure absolute mortality, while HRs measure relative risk of mortality.

Age standardised mortality rates

The most notable improvements are the outcomes for the Black African ethnic group.^{[footnote 22]} In the first wave, age-standardised mortality rates (ASMRs) of COVID-19 mortality were greatest among individuals identifying as Black African (402.5 and 174.1 deaths per 100,000 population in men and women respectively); however, during the second wave ASMRs of COVID-19 mortality fell sharply for Black African people (79.7 and 32.0 deaths per 100,000 population in men and women respectively). Comparing annualised mortality rates from the first and second waves - see Annex C - both Black African men and Black African women show a decrease of over 60% compared with the first wave. However, the second wave mortality rates have risen by 124% and 97% for men and women from Pakistani backgrounds, respectively.

During the first wave ASMRs were lowest among those identifying as White British (119.1 and 65.1 deaths per 100,000 population in men and women, respectively) and risk of death was greater among all ethnic minority groups compared with the White British population. In the second wave, the risk was similar among ethnic minority groups compared with the White British population, with the exception of Pakistani, Indian and Bangladeshi people, where it remained higher.

In the second wave, the ASMRs of COVID-19 mortality were highest among men and women identifying as Pakistani (339.9 and 166.8 deaths per 100,000 population in men and women) and Bangladeshi (318.7 and 127.1 deaths per 100,000 population in men and women). Comparatively, the ASMRs of COVID-19 mortality were low for White British populations (77.8 and 42.6 deaths per 100,000 population in men and women) and lowest of all for people identifying as Chinese (43.7 and 44.0 deaths per 100,000 population in men and women). High Pakistani and Bangladeshi mortality may reflect the high infections during the second wave for these groups. Pakistani infection rates have historically been higher than other ethnic minorities (with the exception of the Other ethnic groups population, which is likely inflated by misclassification) but increased sharply, compared with most other ethnic groups, from early October to early December.^{[footnote 23]}

Researchers affiliated to the Office for National Statistics, University of Oxford, London School of Hygiene and Tropical Medicine, and University of Leicester behind the ASMRs analysis state that “An appreciable reduction in the difference in COVID-19 mortality in the second wave of the pandemic between people from Black ethnic background and people from the White British group is reassuring, but the continued higher rate of mortality in people from Bangladeshi and Pakistani backgrounds is alarming and requires focused public health campaign and policy response”.

Figure 2: COVID-19 age-standardised mortality rates (ASMR) by ethnic group and sex separately for the first (24 January 2020 to 31 August 2020) and the second (1 September 2020 to 28 December 2020) waves of the pandemic

Source: Office for National Statistics

Hazard ratios

In the first wave, Black African, Pakistani and Bangladeshi men were 4.5, 2.7 and 3.5 times more likely to die from COVID-19 respectively than White British men of the same age. In the second wave (to date) the mortality risk remained high for Pakistani and Bangladeshi men (4.8 and 4.1) but not for men from Black ethnic groups who had the same rate as White British men. A large part of this continued disparity for South Asian populations can be explained from geographic, socioeconomic and health factors.

In both waves, adjustment for geographical factors, socio-demographic characteristics and pre-pandemic health substantially reduced the estimated disparities between most ethnic groups and the White British population.^{[footnote 24]} This suggests that the differences in mortality between ethnic groups are partly mediated by these factors. However, these factors reduced the hazard ratios (HR) more strongly in wave 1 than in wave 2 (to date). In addition, the factors that most strongly affected the HRs differed in the 2 waves.

In wave 1, adjusting for geographical factors more than halved the estimated hazard ratios for all ethnic minority groups. For most groups, the HRs were further reduced by adjusting for socio-demographic factors and pre-pandemic health status, especially amongst women. After adjusting for all these factors, women from Bangladeshi and Mixed backgrounds were no longer at greater risk of COVID-19 related death than White British people. For women from all other groups except Black African, the fully adjusted hazard ratios were below 1.4. However, despite the reduction of the HRs after adjusting for all these factors, men from all ethnic minority groups except Other White remained at greater risk, but with hazard ratios greatly reduced.

In wave 2 (to date), adjusting for geographical factors did not substantially reduce the HRs in men and women from Bangladeshi background, but reduced the HRs for people from Pakistani background. Adjusting for socio-demographic factors reduced the elevated risks of people from Bangladeshi and Pakistani backgrounds similarly in the 2 waves. Further adjustment for pre-pandemic health status also attenuated the relationship. However in wave 2, even after adjusting for all these factors, people from Pakistani and Bangladeshi backgrounds remained at substantially greater risk of COVID-19 death than White British people, with HRs of 2.7 and 2.0 in men and women from a Pakistani background, and 2.5 and 2.2 in men and women from a Bangladeshi background, respectively. The adjustments had little impact on the HRs for people from an Indian background.

Analysis from OpenSAFELY^{[footnote 25]} found similar results: “Ethnic inequalities in the risk of dying from COVID-19-related death appear to have changed between the first and early second wave of the epidemic in England. In the period between September and early November 2020, South Asian ethnic groups (in particular those of Bangladeshi and Pakistani ethnicity) were at markedly increased risk compared with people of White ethnicity, while people of Black and other ethnicities had similar risks to White [people]”.

Based on this evidence, the changes observed in COVID-19 outcomes among ethnic groups within such a short time period strongly suggest that ethnic inequalities in COVID-19 outcomes are driven by risk of infection as opposed to ethnicity itself being a risk factor for severe illness or death from COVID-19.

According to ONS researchers “The widespread dissemination of research findings and government reports published during the first wave of infection, that highlighted that people from ethnic minority groups were disproportionately affected by COVID-19, may have helped raise the awareness of these disparities amongst the general public”.^{[footnote 26]} The information provided by academia and government, in combination with the government response to address disparities, may have led to reduced disparities during wave 2, especially for the Black African and Black Caribbean ethnic groups. Different geographic ‘hotspots’ throughout the pandemic may also have contributed to the changing risk experienced by different populations, for example, Black people in London during the first wave or Pakistani and Bangladeshi people in the North West during the early second wave.

There is work underway to consider why the second wave to date had such a disproportionate impact on South Asian groups, led by the SAGE ethnicity subgroup. Relevant considerations include:

• The early part of the second wave was particularly evident in the North West, North East, Yorkshire and Midlands (regions that they have high proportions of people from the Pakistani group).^{[footnote 27]} From December onwards, leaving the early second wave and entering the peak, COVID-19 infection rates in London, East of England and the South East increased.^{[footnote 28]} At the same time, COVID-19 infection rates for Indian, Other Asian and Black populations notably increased, similar to the infection rate seen for Pakistani people. Different regional patterns in how the waves have emerged may contribute to the differing patterns of infections by ethnic group.
• Bangladeshi and Pakistani households in England have the highest rates of overcrowding at 24% and 18% respectively.^{[footnote 29]} They also have the highest rates of large households; 33.6% of Pakistani adults and 32.0% of Bangladeshi adults live in households of 6 or more people.^{[footnote 30]}
• There is higher excess mortality risk for South Asian women when compared with other minority groups, which is partially explained through higher rates of living in multigenerational households - less than 10% of White women aged 65 and above are estimated to live in a multigenerational household, compared with 60.0% of Bangladeshi women and 58.4% of Pakistani women.^{[footnote 31]} These were similar proportions to Pakistani and Bangladeshi men aged 65 and above. However this does not explain the excess mortality experienced by South Asian men or other ethnic groups. According to SAGE, “This may be associated with the different circumstances in which women live in intergenerational households, for example, they are more likely to reside with other family members once widowed, and potentially more vulnerable. This may in part be because for minority ethnic groups household composition is more strongly correlated with other factors which are already adjusted in the analysis”.
• Additionally, COVID-19 infection^{[footnote 32]} and risk of mortality^{[footnote 33]} are both associated with deprivation. Asian populations are over-represented in deprived areas in England, with 15.7% of Asian people living in the most deprived 10% of neighbourhoods.^{[footnote 34]} This is more evident in certain Asian populations; 31.1% of Pakistani people and 19.3% of Bangladeshi people live in the most deprived 10% of neighbourhoods.
• The COVID-19 rate of mortality is also higher in urban areas and highest in major urban conurbations - this category includes areas such as London, Birmingham, Oldham and Bradford^{[footnote 35]}, all of which have large South Asian populations.
• Differences in occupational exposure could also account for some of the differences in mortality between groups, as a higher proportion of Pakistani and Bangladeshi men work as taxi drivers, shopkeepers and proprietors than men in any other ethnic backgrounds.^{[footnote 36]} For example, 33.0% of male taxi and private hire vehicle (PHV) drivers and chauffeurs and 10.9% of male security guards and related occupations are Bangladeshi or Pakistani men, compared with 3.1% of men in all occupations.^{[footnote 37]} Both occupations have higher mortality rates, 101.4 and 100.7 per 100,000 respectively, compared with 31.4 per 100,000 for working age men in England and Wales.^{[footnote 38]}

Health inequalities

This section looks beyond inequalities in the impact of COVID-19, and at health inequalities by ethnicity more generally. The material here is a relatively high-level summary.

A recent briefing from The King’s Fund^{[footnote 39]} considers how selected outcomes in health differ between different ethnic groups in England. Some summary points from the briefing are as follows:

Access to primary care health services is generally equitable for ethnic minority groups, but less consistently so across other health services.^{[footnote 40][footnote 41]} However, people from ethnic minority groups are more likely to report being in poorer health than their White counterparts^{[footnote 42][footnote 43][footnote 44]} and to report poorer experiences of using health services.^{[footnote 45][footnote 46][footnote 47]}

Despite this, prior to the COVID-19 pandemic, overall mortality (adjusted for age) was lower among most ethnic minority groups than the White population.^{[footnote 48][footnote 49][footnote 50][footnote 51][footnote 52]} This underlines the complexity of the picture and the importance of distinguishing between the inequalities experienced by different ethnic groups:

• People from the White Gypsy or Irish Traveller, Bangladeshi and Pakistani communities have the poorest health outcomes across a range of indicators.^{[footnote 53]}
• Compared with the White population, disability-free life expectancy is estimated to be lower among several ethnic minority groups.^{[footnote 54]}
• While the incidence of cancer is highest in the White population^{[footnote 55]}, rates of infant mortality^{[footnote 56]}, cardiovascular disease (CVD) ^{[footnote 57][footnote 58]} and diabetes^{[footnote 59]} are higher among Black and South Asian groups. CVD and diabetes cause significant morbidity among these groups, much of which can be prevented by public health measures aimed at tackling risk factors such as obesity, poor diet, inadequate physical activity and smoking.

The COVID-19 pandemic has had a disproportionate impact on ethnic minority groups, reversing the mortality advantage in some ethnic minority groups.^{[footnote 60]}

The King’s Fund briefing provides contextual information for some of the factors driving these ethnic differences in health.

Non-COVID-19 deaths

An OpenSAFELY study^{[footnote 61]} found that, between 1 February and 9 November 2020, people from ethnic minority groups were more likely to die from COVID-19 than their White counterparts. After adjusting for age and sex, the odds ratios for the South Asian, Black, Mixed and Other ethnic groups for dying from COVID-19 were 2.33, 2.20, 1.75 and 1.54 respectively. However, during the same period, ethnic minority groups had lower odds than the White ethnic group of non-COVID-19 death (odds ratio for South Asian: 0.78, Black: 0.88, Mixed: 0.86, Other: 0.65).

The study reports that “the evidence from both the internal comparison in the present study, and related data from other studies, suggests that the higher risk of poor COVID-19 outcomes reflects unique features of the pandemic rather than a generalised higher risk of death in non-White groups.” Other evidence suggests that prior to the COVID-19 pandemic, deaths from infectious diseases - as well as non-infectious disease deaths - were lower among ethnic minority groups compared with the White population.^{[footnote 62]} The disproportionate deaths from COVID-19 experienced by ethnic minorities may be due to something more specific to COVID-19, rather than infectious diseases more generally. The OpenSAFELY study continues “reasons might include a high likelihood of working in at-risk occupations with high exposure risk, such as health and social care, hospitality and public transportation; and a high likelihood of living in large, high-density or multigenerational households, which might act individually or in combination to increase the risk of infection, and thus the overall risk of COVID-19 death, particularly if a high exposure risk in younger people leads to increased infection in older people via households and community settings”.

Looking at the specific causes of death, the OpenSAFELY study found that people from Mixed and Asian backgrounds were more likely to die from cardiovascular disease than their White counterparts, whilst the Black ethnic group were more likely to die from Dementia or Alzheimer’s than White people. For all other types of non-COVID-19 death (for example, cancer, respiratory), people from the White ethnic group were more likely to die than those from the Black, Asian, Mixed or Other groups. In all age groups the risk of death from cancer and cardiovascular disease was higher than the risk of COVID-19 death. In individuals over 80, dementia/Alzheimer’s was also a leading cause of death.

Vaccinations

According to analysis published by SAGE^{[footnote 63]}, in previous national vaccination programmes in the UK, reported uptake has been lower in areas with a higher proportion of minority ethnic group populations. There is a risk that vaccine uptake for COVID-19 may also be lower among minority ethnic groups.

Primary care data analysed by QResearch indicates that, in the case of several viruses, Black African and Black Caribbean groups are less likely to be vaccinated compared with White groups; over-65s from the Black Caribbean population are half as likely to have had the influenza vaccine, compared with over-65s from the White group.^{[footnote 64]} Furthermore, for vaccines for viruses that have emerged since 2013, all adults in minority ethnic groups were less likely to be vaccinated compared with those in White groups (by 10-20%).

Recent survey data from the UK Household Longitudinal study shows that Black groups are the most likely not to be vaccinated, the next group being the combined Pakistani and Bangladeshi group. White groups are the most likely to be vaccinated.

According to ONS^{[footnote 65]}, from early December 2020 to early January 2021, less than half (49%) of Black or Black British adults reported that they were likely to have the vaccine. Higher percentages were reported amongst people from White (85%) and Mixed ethnicity (80%) backgrounds.

The ONS study mentions that of those who said they were unlikely to have the vaccine:

• over 6 in 10 adults of ethnic minority background (64%) reported they were worried about the side effects of the vaccine, compared with 45% of adults of White ethnic background
• around 2 in 10 adults of White ethnic background (21%) reported they did not feel COVID-19 was a personal risk, compared with 11% of ethnic minority background

More recent research from REACT-2, undertaken from 26 January to 8 February 2021, suggests that 72.5% of Black adults would accept a vaccination^{[footnote 66]}; an increase on the ONS estimate but still lower than the figures for adults from the Mixed, Other, Asian and White groups (83.1%, 84.4%, 87.6% and 92.6% respectively).

Lower COVID-19 vaccine uptake is evident for some ethnic minority groups in vaccines administered to over 80s as at 4 February.^{[footnote 67]} The vaccine rates for eligible Black African, White and Black African, Other Black, Bangladeshi and Pakistani people are notably low (45.1%, 51.5%, 53.4%, 54.6% and 55.2% respectively). This compares to 82.8% of eligible White British people having received a vaccine.

A further study of vaccine uptake among all staff at University Hospitals of Leicester NHS Trust found that ethnic minority healthcare workers are far less likely to take up COVID-19 vaccination than those of White ethnicity.^{[footnote 68]} This has implications for delivery of the COVID-19 vaccination programme.

Understanding people’s personal experiences of COVID-19

As noted in the first report, RDU commissioned the Policy Lab^{[footnote 69]} to undertake a deep dive into the experiences of 12 people from different ethnic minority backgrounds. Using in-depth interviews and observing daily activities over 8 weeks, the research provides deep insight into the impact of COVID-19 on participants’ experience of everyday life in the autumn of 2020, changes in views as circumstances change and brings quantitative studies to life. The study was designed to include people from a variety of ethnic groups and circumstances with which to compare experiences and perspectives (see Annex C); however, participants are spokespeople only for their own lives and should not be considered as representative of their ethnic minority community or any other communities that they belong to.

6 themes were identified (see Annex C) which cut across all ethnic groups, economic circumstances and geographic locations. The most noteworthy points for the RDU are:

• participants felt that their ability to adapt and adjust, along with their degree of hope and pragmatism, was largely dictated by the social and economic resources available
• participants’ beliefs, exposure to media and personal experiences of COVID-19, among other factors, influenced their perceptions of risk. As such, management strategies were based on each individual’s assessment and interpretation of these complex and cumulative risks

These findings align with supporting evidence which shows that some ethnic minorities are likely to experience occupation related risks, such as working in health and social care.^{[footnote 70][footnote 71][footnote 72]} SAGE reported that “Modifiable social factors such as poverty and occupation make a large contribution to the greater burden of COVID-19 in ethnic minorities (high confidence)”.^{[footnote 73]}

These findings point to the need to understand individual risk and support the development of, and wider access to, objective risk assessment tools as discussed in the first report.^{[footnote 74]}

Further noteworthy findings from this research include that:

• Participants from various ethnic minority groups felt that the use of the term ‘BAME’ in communications framed ethnic minority groups as a homogeneous group that is vulnerable to COVID-19. This association of higher risk with ethnic minorities exacerbated participant’s feelings of being stigmatised and distrusted; particularly when generalised statements were made about this group’s risk without providing the appropriate rationale behind the statement.
• The RDU did not observe a consistent narrative around perceptions of compliance. Policy Lab found that most participants had experienced instances where they felt avoided, blamed or berated about their lifestyles and alleged non-adherence to the rules. One participant felt that although he and his family were compliant, his ethnic minority group (in the abstract) were not, while other participants felt that they or their communities were more stringently following social distancing to avoid further blame for spreading COVID-19.
• This also influenced how some participants felt about the vaccines. While most participants were positive about the vaccines and a return to ‘normal’, some participants reported distrust of the vaccines. These feelings were influenced by their life experiences, encounters with misinformation, confusion over media narratives surrounding risk to ethnic minorities, and pre-existing feelings of trust towards institutions.
• Most participants reported receiving their news from official sources such as government announcements or GOV.UK and all were very aware of the guidelines and tiers as a result; however, nearly all encountered misinformation. Most participants did not believe the content and some perceived it as dangerous.
• Some reflected on the difficulty in navigating the advice and applying it to everyday situations, as well as adapting to the pace of change of the guidance. This, coupled with influences outside of the reliable news sources (such as information from the influence of close relations), led to one participant expressing views that COVID-19 could be viewed as a belief rather than fact, although they still followed restrictions carefully as a precaution.
• Furthermore, many of the conversations with participants centred around discussions on what they did or did not believe in relation to COVID-19. This highlights the variety of views and different perceptions surrounding COVID-19 across participants

Research from SAGE and SPI-B highlights the impact of messaging on behaviour. For example SPI-B’s report explains, “For health messages to be effective, they need to reflect evidence on the relationship between cognitions such as beliefs and attitudes and specific behaviour. When messages target how people think about the target behaviour, they are more effective in changing that behaviour”.^{[footnote 75]}

RDU will share Policy Lab’s findings, as appropriate, with officials across government to allow policy makers to see the real time and longitudinal impacts of policy interventions.

Improving data quality

The RDU and others have been working to improve the quality of data used to measure the impacts of COVID-19 on people in ethnic minority groups.

Improving the recording of ethnicity in GP practices

The DHSC laid a Statutory Instrument and accompanying Directions in December which made changes to the regulations governing GP contracts. When ethnicity data is provided by the patient (or someone lawfully acting on their behalf where the patient is a child or someone who lacks capacity), the GP is now mandated to record that information in general practice. This Amending Regulation was made to the GP Contract in response to perceived problems in capturing and recording patient ethnicity data that was highlighted by a PHE report on COVID-19. It is a permanent change to GP Regulations.

The amendment seeks to improve recording of ethnicity data to enable NHS services and programmes to be more effective for patients. The information may be used by NHS England and other healthcare commissioners in the future to gain a better understanding of population demographics and assist in public health management and the identification and monitoring of public health inequalities.

Data harmonisation

The preferred standards for the collection and publication of data for the protected characteristics (including ethnicity) are the Government Statistical Service (GSS) harmonised standards.

Harmonisation is hugely important as it allows analysts to gain deeper insight and value from data and the RDU has already set out its commitment to the harmonisation of ethnicity data across government in its Quality Improvement Plan.^{[footnote 76]}

For ethnicity, the GSS harmonised standard is currently based on the 2011 Census and has 18 groups. The GSS Harmonisation Team will update the harmonised standard soon, reflecting the approaches adopted for the 2021 Censuses in England, Wales and Northern Ireland, and 2022 in Scotland.

Data owners in departments and agencies should commit to aligning their ethnicity data collections to the harmonised ethnicity standard as defined by the GSS, and publish their commitment to doing so, including timescales. As the harmonised standard may change over time, data owners are advised to consult with ONS Best Practice and Impact Division and the RDU before making any change.

A significant harmonisation priority is the publication of recommendations from the Unified Information Standard for Protected Characteristics (UISPC) project. This is a commission from DHSC to NHS England designed to enable the DHSC and the wider NHS to identify viable options for improving the consistency, detail and overall quality of equality data, including ethnicity. The UISPC recommendations could help to move NHS classifications from the 2001 harmonised Census ethnicity categories to an updated set of classifications.

The full set of UISPC recommendations is being considered within NHS England and NHS Improvement, and advice will be given to DHSC in due course. This activity should be prioritised.

While respondents to the Census have always had the option of writing in their ethnicity, moving data collection to the new Census classifications (or something similar) could facilitate the capture of better quality data for some groups. This could be through the new ‘Roma’ tick box and new write-in option for ‘Black African’ in England, for example.

The quality of ethnicity data in health datasets

Ethnicity information on COVID-19 cases and deaths following a positive test for COVID-19 is obtained through linkage to Hospital Episode Statistics (HES) and determined by the most recent ethnicity given by a person - people might give their ethnicity a number of times during different visits to hospital. There are several limitations with this approach:

• Ethnicity is presumed to be self-reported by the patient in hospital records, but this may not always be the case.
• Patients may also report different ethnicities in different episodes of care (and the recording of different ethnicities could also happen if staff ascribe ethnicity for some episodes)
• People from certain ethnic backgrounds may be less likely to have complete records with which to complete the linkage to hospital data.

Previous analysis has shown higher population-based diagnoses and death rates in the Other ethnic group due to a mismatch between ethnicity assigned in the population data (based on the 2011 Census) and hospital records. Further work is ongoing within PHE to improve the allocation of ethnicity to COVID-19 cases and deaths, and improvements will be implemented over the coming months.

In October 2020, The King’s Fund and UCL Institute of Health Equity submitted a paper to NHS England and NHS Improvement highlighting some of the limitations in both the coverage and quality of ethnicity recording of people using NHS health and care services in England and suggesting how this can be improved.^{[footnote 77]}

Improvements to data might come in a number of ways. One way is through harmonisation. Others include improvements through data linkage or applying analytical techniques on data that have already been collected, an example being reclassifications by PHE. Finally, improvements can, and should, come through data collection.

The The King’s Fund and UCL Institute of Health Equity paper suggested ways that data might be improved when ethnicity is collected from patients.

It also recommended that “to ensure that ethnicity recording in health records is fit-for purpose to support the many key functions it is designed to do … DHSC, NHSEI and NHS Digital (should) take steps to ensure that NHS organisations and staff, and GPs, are aware of how this information should be collected from patients and recorded”. The RDU endorses this recommendation. The publication of a project plan for this work by NHSE/I (in collaboration with DHSC and others), along with regular progress updates, would be welcome.

A further project being undertaken by the Nuffield Trust will examine the completeness, validity, and consistency of ethnicity coding within NHS health datasets in England (excluding GP records). The project aims to establish the extent and nature of data quality issues, and understand the implications for analysis and decision making about ethnicity and health. The project is expected to report in April 2021, and is being undertaken in collaboration with the NHS Race and Health Observatory.

This report will provide additional information to inform action to improve the quality of ethnicity data in health records. The recommendation from the first quarter report to collect ethnicity information during death certification is dependent on ethnicity being taken from health records. It is therefore critically important that ethnicity data in health records are fit-for-purpose.

This report also considers ongoing improvements to, and limitations of, 2 other ONS surveys, the COVID-19 Infection Survey (CIS), and the Opinions and Lifestyle Survey (OPN). Both surveys are important in measuring different aspects of the impact of COVID-19.

There have been a number of quality improvements to both of these surveys. For example:

• the sample size of the CIS has increased. In the 12 months from the start of the survey, it is expected that ONS will recruit approximately 500,000 individuals from approximately 240,000 households in total across the 4 countries of the UK.
• The time between the OPN survey fieldwork closing and results being published has also reduced to allow more timely analysis of how attitudes and experiences are changing through the pandemic.
• Starting with data collected between 21 and 25 October, the sample size for the weekly OPN has approximately tripled in England, to around 3,000 people.

However, while the sample size of the CIS has increased, small sample sizes in each of the 4 aggregate ethnic minority groups (and the 18 detailed groups) remain too small to produce robust estimates for these groups.

Also, there is an underrepresentation of ethnic minority groups in the OPN. Ethnic minorities constitute around 6% of the sample in the OPN, compared with 14% in the 2011 Census.

RDU will continue to work with ONS analysts to improve the quality of data for ethnic minority groups from these (and other) surveys.

ONS has also recently initiated a wider project to improve how they engage with under-represented groups. The project will develop evidence-based recommendations to ensure that future mixed-mode social survey designs are more representative.

As part of this project ONS are also going to consider their approach to sample design to investigate whether the samples drawn could be more inclusive and representative of minority groups than at present.

Quality of vaccinations data

The Office for Statistics Regulation has recently called for producers of COVID-19 vaccine statistics across the UK to address some data quality issues with vaccinations data.^{[footnote 78]} These include providing more detailed characteristics of recipients, for example their ethnicity, age, and sex.

RDU will provide an assessment of the quality of data about vaccinations among ethnic groups in the next quarterly report.^{[footnote 79]}

Data publication and access

It is important that the datasets are as timely as possible to facilitate analysis within and outside of government. To this end, a publication schedule of updates to datasets on the RDU’s Ethnicity facts and figures website that relate to risk factors and secondary impacts will be published (on the website) and is presented in Annex C.^{[footnote 80]} Departments should provide the datasets to the timescales set out in that schedule.

The datasets listed in the schedule are highlighted in the Ethnicity and COVID-19 part of the website.^{[footnote 81]} RDU is maintaining this page as a way to collect data and analysis from across government. This page provides links to data about:

• deaths and disparities in outcomes among people, disaggregated by ethnicity
• risk factors associated with getting and dying from COVID-19
• economic, educational and emotional impacts

To enable users to have access to up to date data, ONS has also made available a number of the most important micro-level datasets for the analysis of COVID-19 disparities between different ethnic groups. These are available from the ONS Secure Research Service on application and are listed in Annex C.

4.4 Next steps

The RDU will continue to share the findings from the qualitative research into people’s personal experiences of COVID-19 across government, particularly in relation to the stigmatisation felt by a number of participants in relation to being singled out as ‘BAME’.

Departments and other agencies should publish a statement on GOV.UK outlining their plans to move their data collections to the Government Statistical Service’s (GSS) harmonised standard.

NHSE/I, working with DHSC and others, should publish a quarterly report on progress in improving the recording of ethnicity in health care records.

Departments should provide updated datasets on COVID-19 risk factors and secondary impacts for publication on the Ethnicity facts and figures website in line with the schedule in Annex C. This provides transparency of process to users - promoting trust and authority - as well as informing them when the most up-to-date data will be made available.

5. Stakeholder engagement

5.1 Summary

Over the last quarter, the Minister for Equalities, the government advisers on COVID-19 and ethnicity and RDU officials have led a programme of engagement both within government and with external stakeholders. This has comprised working groups, bilateral meetings, conference speeches and targeted roundtables. This work will continue in the next quarter.

5.2 Approach

Within government, the Minister for Equalities shared the findings from her first quarterly report with other ministers through meetings and correspondence. This included updating ministerial colleagues on the likely drivers behind the higher COVID-19 infection and mortality rates for ethnic minority groups. The minister stressed the importance of not mistaking risk factors that are prevalent among ethnic minorities with portraying ethnicity itself as a risk factor, and potentially stigmatising ethnic minority groups. The Minister for Equalities also set out the importance of measuring the impact of policy interventions and shared with other ministers the technical annex prepared by the RDU.

The RDU also established a cross-government policy working group in November as a means of sharing the main findings from the research into why COVID-19 has had such a disproportionate impact on ethnic minority groups, and to encourage development of new policy interventions to address these findings. The RDU has also worked closely with DHSC, MHCLG, PHE and other parts of the Cabinet Office on a range of issues including new guidance and accompanying, targeted communications on preventing household transmission of COVID-19, particularly within multigenerational households.

The Minister for Equalities also met the Minister for COVID-19 Vaccine Deployment in January to discuss how data relating to ethnicity would be collected during the vaccination programme and how to promote vaccine uptake among ethnic minority groups.

In terms of external engagement, the Minister for Equalities has met a number of important stakeholders including the British Medical Association and the Health Foundation. She met the BAME Communities Advisory Group (representing the adult social care sector) to consider their report on the impact of COVID-19 on ethnic minority people who are in receipt of social care services or work in the social care sector.^{[footnote 82]}

The Minister for Equalities, alongside the Minister for COVID-19 Vaccine Deployment and the Minister for Prevention, Public Health and Primary Care also attended a round table with the National Pharmacy Association (NPA), a network of 14,000 pharmacies across the UK. This was an opportunity to share the findings of her work, to hear directly from pharmacists about tackling health inequalities and to consider how to encourage understanding and uptake of the COVID-19 vaccines. The NPA is a particularly important advocate as ethnic minority workers represent 43% of the pharmacy profession.

The Minister for Equalities gave the keynote speech at the Westminster Health Forum conference on tackling ethnic minority inequality in health and social care, held on 9 December.

The RDU and the Minister for Equalities have continued engagement with maternal health stakeholders over the last quarter. Following a joint Ministerial roundtable event in September 2020, the RDU has supported DHSC in driving positive actions in maternity services to improve outcomes for ethnic minority women. This includes the recently launched NHS campaign ‘Help us Help You’ informing pregnant women about the importance of attending check-ups, and providing reassurance that the NHS is there to see them safely.

The RDU and DHSC have also engaged with midwives from Birmingham and East London NHS Trusts who are on the frontline of the local COVID-19 response, in order to identify best practice as well as areas for improvement.

The Minister for Equalities also gave a keynote speech at Action on Pre-eclampsia’s Annual Expert Meeting in November and met the Chief Midwife for England on 15 December.

In February, No.10 and the RDU hosted 2 roundtables on promoting vaccine uptake amongst South Asian groups. These groups were selected because of the disproportionate impact COVID-19 has had on them in the second wave and because of concerns about vaccine uptake amongst these and other ethnic minority groups..

The events generated a number of ideas and insights on how to improve vaccine uptake in the South Asian cohort such as:

• expanding the use of places of worship as vaccination centres
• the significant role women can play in countering misinformation and encouraging uptake
• the importance of high profile appearances and interactions with the community in order to improve vaccine uptake, such as the Prime Minister’s visit to the Al Hikmah centre in Batley on 2 February which was cited as particularly impactful

The work to promote vaccine uptake among ethnic minority groups, and particularly those that have been disproportionately impacted by the second wave, will continue in the next quarter.

Reducing maternal health disparities during the pandemic

An example of best practice is the ‘Birmingham Pathway’. Midwives in the University Hospitals Birmingham NHS Trust identified that local ethnic minority women in the top ten percentile of deprivation were at higher risk for COVID-19. In response, the Trust convened a multidisciplinary team and launched a maternity surveillance programme to share data and monitor pregnant women testing positive.

The programme aimed to detect at an early stage COVID-19-related problems in pregnant women. COVID-19 phone lines made daily calls to patients, guaranteeing rapid access to midwives. 45% of women on the surveillance programme were from ethnic minority groups. Women’s feedback to midwives was that they felt supported and reassured by calls. This initiative provided a swift, data-informed response to prevent poor outcomes for women at higher risk of COVID-19 complications.

Additional actions undertaken by maternity services in areas with higher ethnic minority populations include conducting community outreach to address misinformation and disinformation, hosting weekly online Q&A forums for pregnant women, and engaging with local faith leaders who provide trusted avenues for myth-busting. It is vital that best practices like these continue to be shared across maternity services networks to ensure that women and their babies receive appropriate care and present early to hospital during COVID-19.

5.3 Next steps

The Minister for Equalities, the government advisers on COVID-19 and ethnicity and the RDU will continue a programme of engagement over the next quarter. This will include work to promote vaccine uptake, alongside the engagement led by the Minister for COVID-19 Vaccine Deployment.

6. Communications

6.1 Summary

The government has continued work to build on the way that public health messages are effectively delivered to ethnic minority people. There have been ongoing improvements within this multichannel communications strategy to address language, cultural and accessibility barriers, with a particular focus on targeted community engagement to better inform communications through bespoke creative content and messaging.

In the second quarter, as the new vaccines were announced, a concerted effort is being made to understand and overcome concerns about the vaccines among ethnic minority people. This has had a particular focus on encouraging uptake of vaccines and countering misinformation. Ministers, Dr Raghib Ali (government adviser on COVID-19 and ethnicity) and other healthcare experts have played important roles in supporting this effort.

6.2 Approach

To improve vaccine confidence and uptake among ethnic minority people, a multi-channel approach is being taken which includes but is not limited to the following activity:

The government has worked with specialist agencies to hold a series of roundtables for ethnic minority healthcare professionals and religious and community leaders to act as ambassadors within their communities. These sessions provided an opportunity to feed back concerns, recommend approaches and have specific questions answered by health experts such as Dr Ali and Dr Mary Ramsey, the Head of Immunisation from Public Health England.

To improve its understanding of vaccine concerns, the government is now working with over 90 faith, healthcare provider networks, influencers and experts from a range of communities. As a result of these new relationships, further Q&A sessions have been confirmed with the Sikh Council, NHS Muslim Network, Sarwat Tasneem who are networked into various Muslim women’s groups, the Polish Vaccine Foundation, Oxford Polish Association and the National Pharmacy Association.

Editorial content packages are being developed with trusted voices among healthcare professionals, who will feature in media opportunities and digital content.

Utilising the ethnic minority radio partnership, the government has played out 29 3-minute pre-recorded interviews with local NHS community members on vaccine confidence, running across 13 community radio networks in 14 different languages.^{[footnote 83]} This activity has also been carried on social media channels.

Additionally, vaccines confidence activity is planned to run on 42 multicultural stations in 12 languages. Phase 1 includes 5 stations in 3 languages producing 30 second TV ads with some of their best known talent. These video pieces went live on 18th Jan. A second phase has followed as more content has been produced.

Since the start of the pandemic, specialist government units have been working at pace to identify and rebut false information about coronavirus, including the vaccines. The cross-government Counter Disinformation Unit (CDU) brings together different monitoring teams across Whitehall including teams in the Home Office, Foreign Commonwealth and Development Office and the Rapid Response Unit (RRU).

RRU identifies important emerging narratives and, where they identify content that could breach platforms’ terms of service, are dangerous or misleading, they flag these to the platforms for review and removal via established processes. Monitoring covers all the major open platforms and the government has trusted flagging relationships in place with Twitter, Facebook/Instagram, and YouTube. The government has close relationships with Facebook and continues to work with them to identify potential solutions to mis- and disinformation across all their platforms.

To tackle mis- and disinformation among ethnic minorities, the government is regularly producing myth-busting content and utilising trusted platforms and messengers within communities and taking specific targeting approaches on social media channels (such as Facebook and Instagram which allows for better targeting). We are also using native language publisher sites and targeting specific media outlets (Asian Voice, Leader, The Nation, JC and Desi Express) as part of ongoing partnership work.

Translation continues to be a priority to reach those whose first language is not English and/or who have other accessibility needs. This includes translation of videos into British Sign Language and posters into Easy Read and Large Print as well as language translation. In addition to the translation of national assets, local authorities can request translations of their own assets.

Materials translated this quarter include a leaflet for parents advising what to do if their child contracts COVID-19 and guidance on self-isolation. This particular piece was translated into 26 community languages, based on insight from the Local Authority. Translated materials are also shared in editable formats so they can be used by as wide an audience as possible.

The government continued to reiterate health messaging around important calendar moments such as Diwali and Christmas. The approach has included work with faith leaders to advise, co-create and share communications on safe worship for these festivals.

Additionally, campaigns around universities and the COVID-19 app have included elements targeting ethnic minority people. Campaign approaches include press partnerships and the use of micro influencers to communicate important messages.

Innovative government-first partnerships with specialist ethnic minority media use regular, tailored advertorial content that helps get messages and guidance across.

The government has continued to prioritise collection of insight into diverse audiences to better understand and serve their communication needs. Through ongoing polling and focus groups, we now have a better understanding in areas such as media channel consumption and propensity to take up vaccines.

The government has been gathering insight on sentiment towards vaccinations, focusing on a variety of groups including ethnic minorities. When understanding the barriers to trust and uptake, it appears people including ethnic minorities are concerned about suspected side effects, ingredients used in the vaccines and safety.

Therefore, an integrated government campaign was launched to improve understanding and awareness of COVID-19 vaccinations among ethnic minority people and increase their confidence to take up the vaccination when offered it.

As part of the ‘government first’ partnerships with trusted community media outlets, the government is now working with more than 50 ethnic minority titles across 10 different languages (see Annex D). These titles have proven useful to reach the Asian, Afro-Caribbean, African, Bangladeshi, Roman Catholic, Methodist, Russian, Pakistani, Turkish, Jewish, Chinese and Polish communities. The reach of ethnic minority print titles is approximately 446,000 and content features leading experts from the national content as well as relevant local GPs to create bespoke double page spreads.

In addition, the government has partnered with 43 ethnic minority TV channels within a combined reach of 9 million and 14 community radio stations that broadcast in 13 different languages and reach 881,000 ethnic minority people every week.

Work is also in hand with the BBC World Service who support COVID-19 communications by producing videos on important questions from South Asian communities in Urdu, Punjabi, Tamil, Gujarati, and Sylheti, which are produced by the BBC bureau in Delhi and the UK BBC News team. These are shared on BBC websites and social channels.

To meet the needs of people with disabilities, marketing materials continue to be regularly translated into Easy Read, British Sign Language, Braille, Audio and Large Print. The partnership with the BBC continues, giving access to their BSL interpreter feed for live broadcasts of the press conference on government owned social media channels.

A letter about shielding that was sent to all clinically extremely vulnerable people as part of the Tier 4 announcement, was translated into 13 languages as well as Easy Read and Large Print.

Partnerships with respected community figures and organisations like places of worship are being used to help to build trust in the Test and Trace service and dispel existing myths and alternative narratives. To this end, on-site community-led testing, supported by community volunteers has been offered in these communities to learn and address the barriers to engagement and increase take up of testing.

The Test and Trace programme is also reaching diverse communities via social enterprises following a bottom up, community-based approach, to educate, inform and advance understanding of COVID-19 test/trace/isolate requirements as well as tackle concerns about the vaccines.

Highlights from the vaccines campaign (up to Jan 2021)

Organic content on NHS.UK channels has reached over 208 million people, generating 479K likes, 178K comments, 224K shares on Facebook and 5.2 million impressions on Twitter.

Digital advertising has generated 120 million impressions on Facebook, 34 million impressions on Twitter and 28.5 million impressions via digital display; moderators are hiding comments that propagate misinformation to limit its spread and Facebook adverts are being translated and served to those who have the channel set in another language.

Over 20 pieces of traditional media coverage and 40 social media posts featuring medical professionals, including Dr Ranj Singh on ITV’s This Morning, Dr Nighat Arif on BBC Breakfast and Dr Amir Khan on Lorraine and Islam Channel.

Cabinet Office weekly bulletin and toolkit including clear, accessible messages, creative content and Q&A sent out to over 100 local partners and over 70 national partners; additional weekly stakeholder outreach from DHSC to over 176 organisations from across the health, charity and adult social care sectors and from MHCLG to over 600 contacts that include more than 120 faith leaders and 343 Local Authorities.

Black Members of Parliament from the Conservatives and the Labour Party came together to produce a video to encourage people to take the vaccine. In the video MPs told moving personal stories of losing loved ones and warned against the spread of misinformation.

6.3 Next steps

The government will continue to tailor its communications strategy on vaccine roll out to reflect the latest evidence on vaccine uptake among ethnic minority groups.

The government will work closely with the new community champions to disseminate important public health messages, promote uptake of vaccines and tackle misinformation.

Government communications will reflect the findings of the qualitative research into people’s personal experiences of COVID-19 and will ensure that ethnic minorities are not treated as a single group and that public health messaging is not stigmatising.

1. https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

2. https://www.gov.uk/government/news/next-steps-for-work-on-covid-19-disparities-announced ↩

3. https://www.gov.uk/government/publications/covid-19-vaccination-uptake-plan/uk-covid-19-vaccine-uptake-plan ↩

4. https://www.ethnicity-facts-figures.service.gov.uk/ ↩

5. https://www.ukri.org/news/researching-factors-affecting-ethnic-minority-groups-during-covid-19/ ↩

6. https://www.gov.uk/government/publications/priority-groups-for-coronavirus-covid-19-vaccination-advice-from-the-jcvi-30-december-2020 ↩

7. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19infectionsinthecommunityinengland/december2020#likelihood-of-testing-positive-for-covid-19-from-nose-and-throat-swabs-by-ethnicity-in-england ↩

8. https://www.ons.gov.uk/peoplepopulationandcommunity/culturalidentity/ethnicity/articles/coronavirusandthesocialimpactsondifferentethnicgroupsintheuk/2020 ↩

9. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/whyhaveblackandsouthasianpeoplebeenhithardestbycovid19/2020-12-14 ↩

10. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19/latestinsights#ethnicity ↩

11. https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

12. https://www.gov.uk/government/publications/covid-19-pre-existing-health-conditions-and-ethnicity ↩

13. https://www.icnarc.org/Our-Audit/Audits/Cmp/Reports ↩

14. https://www.gov.uk/government/publications/covid-19-pre-existing-health-conditions-and-ethnicity ↩

15. Odds ratios can be treated as a percentage increase or decrease in risk of the outcome under investigation. An odds ratio in excess of 1 indicates raised risk while an odds ratio less than 1 indicates reduced risk compared with the reference population. ↩

16. http://eprints.gla.ac.uk/219444/1/219444.pdf ↩

17. https://www.gov.uk/government/publications/drivers-of-the-higher-covid-19-incidence-morbidity-and-mortality-among-minority-ethnic-groups-23-september-2020 ↩

18. https://www.medrxiv.org/content/10.1101/2020.12.21.20248475v1 ↩

19. https://doi.org/10.12688/wellcomeopenres.15007.1 ↩

20. https://www.gov.uk/government/publications/covid-19-reported-sars-cov-2-deaths-in-england/covid-19-confirmed-deaths-in-england-report#ethnicity ↩

21. https://www.medrxiv.org/content/10.1101/2021.02.03.21251004v1.full ↩

22. When referring to ASMRs for different ethnic groups for a specific wave, we quote the ASMR figures as published by ONS. When calculating % change between waves, we have used the annualised ASMRs for each wave. ↩

23. https://www.gov.uk/government/statistics/national-flu-and-covid-19-surveillance-reports ↩

24. https://www.medrxiv.org/content/10.1101/2021.02.03.21251004v1.full-text ↩

25. https://www.medrxiv.org/content/10.1101/2021.02.02.21250989v1.full.pdf ↩

26. https://www.medrxiv.org/content/10.1101/2021.02.03.21251004v1.full#T5 ↩

27. https://www.ethnicity-facts-figures.service.gov.uk/uk-population-by-ethnicity/national-and-regional-populations/regional-ethnic-diversity/latest ↩

28. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/956709/Weekly_Flu_and_COVID-19_report_w4_FINAL.PDF ↩

29. https://www.ethnicity-facts-figures.service.gov.uk/housing/housing-conditions/overcrowded-households/latest#by-ethnicity ↩

30. https://www.gov.uk/government/publications/housing-household-transmission-and-ethnicity-26-november-2020 ↩

31. https://www.gov.uk/government/publications/housing-household-transmission-and-ethnicity-26-november-2020 ↩

32. https://www.gov.uk/government/publications/react-2-study-of-coronavirus-antibodies-june-2020-results/react-2-real-time-assessment-of-community-transmission-prevalence-of-coronavirus-covid-19-antibodies-in-june-2020 ↩

33. https://www.nature.com/articles/s41586-020-2521-4 ↩

34. https://www.ethnicity-facts-figures.service.gov.uk/uk-population-by-ethnicity/demographics/people-living-in-deprived-neighbourhoods/latest#overall-most-deprived-10-of-neighbourhoods-by-ethnicity ↩

35. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsinvolvingcovid19bylocalareasanddeprivation/deathsoccurringbetween1marchand31july2020#rural-and-urban-areas ↩

36. https://www.medrxiv.org/content/10.1101/2021.02.03.21251004v1.full.pdf ↩

37. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/bulletins/coronaviruscovid19relateddeathsbyoccupationenglandandwales/deathsregisteredbetween9marchand25may2020#coronavirus-covid-19-related-deaths-by-occupation-data ↩

38. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/bulletins/coronaviruscovid19relateddeathsbyoccupationenglandandwales/deathsregisteredbetween9marchand28december2020 ↩

39. https://www.kingsfund.org.uk/publications/health-people-ethnic-minority-groups-england ↩

40. https://pubmed.ncbi.nlm.nih.gov/19622520/ ↩

41. https://library.oapen.org/bitstream/handle/20.500.12657/22310/9781447351269.pdf?sequence=4&isAllowed=y ↩

42. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/730917/local_action_on_health_inequalities.pdf ↩

43. https://jech.bmj.com/content/70/7/653 ↩

44. https://www.ethnicity-facts-figures.service.gov.uk/health/physical-health/health-related-quality-of-life-for-people-aged-65-and-over/latest#data-sources ↩

45. https://www.ethnicity-facts-figures.service.gov.uk/health/patient-experience/patient-experience-of-primary-care-gp-services/latest#by-ethnicity ↩

46. https://www.ethnicity-facts-figures.service.gov.uk/health/patient-experience/patient-satisfaction-with-gp-out-of-hours-services/latest ↩

47. https://www.ethnicity-facts-figures.service.gov.uk/health/patient-experience/inpatient-satisfaction-with-hospital-care/latest ↩

48. https://doi.org/10.1371/journal.pmed.1002515 ↩

49. https://pubmed.ncbi.nlm.nih.gov/23740930/ ↩

50. https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

51. https://pubmed.ncbi.nlm.nih.gov/23740930/ ↩

52. ONS are due to publish comprehensive national data on mortality rates by ethnicity later this year. ↩

53. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/730917/local_action_on_health_inequalities.pdf ↩

54. https://doi.org/10.1080/13557858.2014.921892 ↩

55. https://www.cancerresearchuk.org/health-professional/cancer-statistics/incidence/ethnicity#heading-Zero ↩

56. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/childhoodinfantandperinatalmortalityinenglandandwales/2018#inequalities ↩

57. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/733093/PHOF_Health_Equity_Report.pdf ↩

58. https://academic.oup.com/jpubhealth/article/29/2/191/1505208 ↩

59. https://bjcardio.co.uk/2018/09/diabetes-cvd-supplement-2-diabetes-and-cardiovascular-risk-in-uk-south-asians-an-overview/#:~:text=Like%20CHD%2C%20stroke%20is%20also%20more%20common%20in,in%20studies%20from%20the%20USA%20and%20India.%2028%2C29 ↩

60. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/908434/Disparities_in_the_risk_and_outcomes_of_COVID_August_2020_update.pdf ↩

61. https://www.medrxiv.org/content/10.1101/2021.01.15.21249756v2 ↩

62. https://www.medrxiv.org/content/10.1101/2020.07.27.20162354v1 ↩

63. https://www.gov.uk/government/publications/factors-influencing-covid-19-vaccine-uptake-among-minority-ethnic-groups-17-december-2020 ↩

64. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/952716/s0979-factors-influencing-vaccine-uptake-minority-ethnic-groups.pdf ↩

65. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandwellbeing/bulletins/coronavirusandthesocialimpactsongreatbritain/29january2021#attitudes-to-covid-19-vaccination-by-different-sub-groups-of-the-population ↩

66. https://www.imperial.ac.uk/media/imperial-college/institute-of-global-health-innovation/REACT-2-round-5-preprint.pdf ↩

67. https://github.com/opensafely/nhs-covid-vaccination-coverage/blob/master/released-outputs/opensafely_vaccine_report_overall.pdf ↩

68. https://www.medrxiv.org/content/10.1101/2021.02.11.21251548v2.full.pdf ↩

69. https://openpolicy.blog.gov.uk/about/ ↩

70. https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/articles/coronavirusandkeyworkersintheuk/2020-05-15 ↩

71. https://www.tuc.org.uk/sites/default/files/2020-06/Dying%20on%20the%20job%20final.pdf ↩

72. https://www.ifs.org.uk/inequality/chapter/are-some-ethnic-groups-more-vulnerable-to-covid-19-than-others/ ↩

73. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/925135/S0778_Drivers_of_the_higher_COVID-19_incidence__morbidity_and_mortality_among_minority_ethnic_groups.pdf ↩

74. https://www.gov.uk/government/publications/quarterly-report-on-progress-to-address-covid-19-health-inequalities ↩

75. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/914924/s0649-public-health-messaging-bame-communities.pdf ↩

76. https://www.gov.uk/government/publications/quality-improvement-plan-government-ethnicity-data/quality-improvement-plan-government-ethnicity-data ↩

77. https://www.kingsfund.org.uk/publications/ethnicity-coding-health-records ↩

78. https://osr.statisticsauthority.gov.uk/wp-content/uploads/2021/01/Ed-Humpherson-to-all-producers-Numbers-of-COVID-19-vaccinations-administered-in-the-UK.pdf ↩

79. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandwellbeing/bulletins/coronavirusandthesocialimpactsongreatbritain/29january2021 ↩

80. https://www.ethnicity-facts-figures.service.gov.uk/ ↩

81. https://www.ethnicity-facts-figures.service.gov.uk/covid-19 ↩

82. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/919156/1_BAME_Advisory_Group_report_accessible.pdf ↩

83. Live until 21 January 2021 ↩