Independent report

Chapter 2: disparities

Updated 10 January 2023

Introduction

Infectious disease epidemics and pandemics usually expose and exacerbate existing disparities in society, such as those associated with deprivation, ethnicity, sex, age and sexuality.^{[footnote 1]}, ^{[footnote 2]}, ^{[footnote 3]} The COVID-19 pandemic had some predictable and some less predictable disparities in health outcomes such as the striking age gradient in risk, and the risk of severe disease for people living with obesity.^{[footnote 4]}

Some health impacts are distinct to certain infections – for example, the heightened risk of HIV for men having sex with men in the 1980s, or the risk of severe disease among young adults as well as the very young and the elderly during the 1918 to 1919 influenza pandemic.^{[footnote 5]} Others appear repeatedly across different pandemics, such as more socio-economically deprived groups consistently experiencing greater risk of exposure to infection and worse health outcomes. ^{[footnote 6]}, ^{[footnote 7]}, ^{[footnote 8]}, ^{[footnote 9]}

Some disparities observed in the COVID-19 pandemic would be expected to arise from an airborne respiratory pathogen – such as increased spread among people living in crowded households or individuals working in face-to-face settings with inadequate ventilation or protective equipment and relative sparing of rural areas.^{[footnote 10]}

In addition to the direct health impacts, certain interventions put in place to control COVID-19 can themselves give rise to disparities – though the extent of the impact of COVID-19 control measures may never be fully understood due to lack of a clear counterfactual. For example, more deprived communities and younger people were disproportionately impacted by public health control measures in the short term, including closures to school and the hospitality sector. It is however difficult to say the size of the relative impact of not instigating these measures (and seeing potentially sustained high levels of community transmission) on these groups.^{[footnote 11]}

This chapter sets out how we understood what the key disparities were, and briefly sets out some efforts in response to this evidence – though this is by no means exhaustive and work to reduce disparities continues.

What knowledge was needed and why it was important

Evidence from previous pandemics indicated that it was important to understand differences in infection risk, disease severity and outcomes between groups. These may be linked, or separate; for example, the need to go to work may increase risk of acquiring disease but not severity, while living with obesity may increase risk of severe disease once acquired, but not of being infected.

Alongside this, it was also important to understand the differential impact among population groups of interventions introduced to try and control disease spread. For example, are the right communications getting to the right people, do people needing to isolate have the social, economic and practical support they need, and can everyone get adequate access to the necessary testing and clinical support? It was also essential to understand how different population groups responded to different communication channels, styles and languages, so that interventions could be adapted appropriately.

Disparities arising from the infection and the subsequent policy response will not always be immediately apparent and will instead emerge as the pandemic unfolds, and this was true for COVID-19.

How we found out the information

Our understanding of disparities related to SARS-CoV-2 exposure and COVID-19 outcomes rapidly evolved as the epidemic progressed across the UK. This was a result of the virus reaching increasing numbers of people and communities, and as research programmes, routine statistics and community engagement evolved to better capture the necessary data.

First wave

Early case reports and epidemiological studies on outbreaks provided some important early signals about potential disparities. As early as January 2020, reports from China indicated that COVID-19 led to worse outcomes among older patients and men.^{[footnote 12]} Over the next 2 to 3 months, additional data emerged, primarily from China and Italy, suggesting that people with certain underlying health conditions and immunosuppression were at increased risk of disease and death.^{[footnote 13]}, ^{[footnote 14]} Early data from China also suggested low skilled workers were at increased risk of progression to severe disease.^{[footnote 15]}

As cases began to appear in the UK, the First Few Hundred (FF100) enhanced surveillance protocol was commissioned, following World Health Organization (WHO) protocols and in line with previous pandemic response for MERS-CoV and H7N9 influenza.^{[footnote 16]}, ^{[footnote 17]} This provided basic demographic data and enhanced surveillance of clinical presentation on the first few hundred cases of SARS-CoV-2 infection, allowing for an initial detailed description of people affected.^{[footnote 18]} Early indications of key populations most affected were highlighted – for example, the increased clinical risk in people with underlying health conditions. However, it is worth noting that FF100 investigations are prone to biases (for example, where the first few hundred cases may be returning travellers with similar socio-economic status or health status). This is also covered in Chapter 1: understanding the pathogen.

Several surveillance systems and routine data sets were in place before the pandemic, such as the Second Generation Surveillance System (SGSS) laboratory monitoring and the Office for National Statistics (ONS) death certification. These systems indicated early on that exposure and infection risk were disproportionately high for those working in frontline care or other in-person service occupations, such as transport and cleaning. Although the systems were unable to provide detailed reasons for this, they were likely to be multifactorial and possibly include some non-work risk factors in addition to occupational ones.^{[footnote 19]} Some bespoke surveillance systems were also designed from scratch – for example, to count COVID-19 deaths in hospitals and COVID-19 attendances and admissions to NHS hospitals.

Hospital admission data then rapidly began to produce signals on potential disparities: by February 2020 there was evidence of increased risk of hospital admission for older adults, men and those with certain underlying health conditions.^{[footnote 20]} The regular publication of intensive care data also supported a rapidly growing understanding of ethnic disparities in the UK: in the first wave, statistics highlighted high rates of hospitalisations among patients of black and Asian ethnic groups compared to white ethnic groups.^{[footnote 21]} However, ethnic disparities were often confounded by deprivation and living in areas with high prevalence. As the pandemic went on, patterns of risk for both infection and severe disease changed as the epicentre shifted to areas with different ethnic makeup and as vaccines were rolled out with differing levels of uptake across different communities.

Testing data also supported understanding of disparities: in England, COVID-19 laboratory reporting forms included age and sex from the outset, and ethnicity information was then added by linking laboratory surveillance data with Hospital Episode Statistics data sets.

In order to properly monitor and report on characteristics linked to health disparities (such as ethnicity), there was a need to rapidly link data and enhance routine data sources with clinical and demographic information. This was achievable following rapid issuing of a Control of Patient Information (COPI) notice (for more details see Chapter 4: situational awareness, analysis and assessment). This expedited rapid data sharing between government organisations without requiring unduly long paperwork and approval processes which previously could have taken years.

There was also a need for in-depth reviews alongside these data sets, such as the report ‘Disparities in the risk and outcomes of COVID-19’, published in June 2020 by Public Health England (PHE).^{[footnote 22]} This report was largely undertaken on cases presenting to hospital with a clinical need where testing was concentrated. It highlighted important disparities by age, ethnic group, sex and occupation, likely to reflect disparities in both infection risk and clinical severity. It was not exhaustive and was unable at the time to adjust for some relevant factors in all analyses, such as underlying health conditions, which may affect some groups more than others. It highlighted however some important areas for further investigation, prompting a series of actions to address and mitigate this issue which were documented in reports published by the Equality Hub and Race Disparities Unit.^{[footnote 23]}

Public engagement exercises were used throughout the pandemic to understand the experiences and drivers of observed disparities in COVID-19 health outcomes. For example, an in-depth public engagement exercise with representatives of key affected groups alongside a rapid literature review and qualitative analysis culminated in the publication of another report ‘Understanding the Impact of COVID-19 on Black and Minority Ethnic (BAME) Communities’, which produced a series of recommendations on how to better understand and mitigate the impact of the pandemic on ethnic minority groups.^{[footnote 24]} This included a clear ask for improved data collection on ethnicity, occupation and faith in all routine clinical data and death certification. Alongside this, weekly calls between the CMO’s office and directors of public health helped highlight emerging issues in their communities.

Finally, several studies established in the early phases of the COVID-19 response provided an invaluable contribution to the understanding of COVID-19 disparities. These included the ONS COVID-19 Infection Survey, which provided weekly estimates of infection and immunity, and enabled detailed analyses of disparities such as occupation, ethnicity and deprivation.^{[footnote 25]} The Vivaldi study, meanwhile, collected qualitative and quantitative data on care homes to understand working conditions and the spread of infection and immunity in care home populations.^{[footnote 26]} Its findings have been used to inform the ongoing policy response, including vaccine recommendations. Other studies on specific groups and settings, such as for children and adults with learning disabilities, homeless shelters and prison populations, were helpful in exploring the impact of the pandemic on these groups.^{[footnote 27]}

The QCovid® tool, using population health data to predict outcomes from COVID-19 for different groups, also helped inform the response – for example, vaccination prioritisation. Although designed originally around likely clinical risk factors, it was one of the few tools to include socio-economic deprivation as a component of risk alongside clinical risk as data were refined. It was used slightly differently across the UK – this is explored in more detail in Chapter 8: non-pharmaceutical interventions.

Ongoing response

The regular and transparent publication of disparities data was helpful in maintaining a public and professional focus on disparities as they emerged and changed. Although some disparities data, such as hospital admissions by age and sex, were published from the outset of the pandemic, there was a need to expand and update both data collection and data publication. By the second wave the PHE weekly COVID-19 surveillance report had been expanded to include a wider range of disparities data, and other analyses and research also expanded to examine disparities. The publication of the PHE COVID-19 Health Disparities Monitoring for England (CHIME) tool from May 2021 onwards ensured regular reporting of COVID-19 disparities for a number of determinants and outcomes and is publicly available for use by a range of stakeholders.^{[footnote 28]} In common with most other surveillance systems during the pandemic, CHIME did not have access to data on underlying conditions so this limited the extent to which it could adjust for comorbidities in assessing disparities. Alongside these regular publications, the Scientific Advisory Group for Emergencies (SAGE) regularly reviewed evidence and data on disparities and published its minutes to support public discussion and response to these issues.^{[footnote 29]}

The surveillance landscape was regularly assessed and mapped to identify gaps in disparities data. As a general principle, healthcare and disease surveillance systems need to be designed at the outset with reporting forms that included information on key protected characteristics.^{[footnote 30]} This is to ensure that disparities linked to any of these characteristics could be assessed at the earliest stages of the pandemic. There was also an ongoing need to secure public trust in data gathering and usage, ensuring usage of data was transparently communicated.

Important factors in the COVID-19 pandemic

Infection risk

Certain occupational groups such as factory workers, healthcare workers, emergency service workers, social care workers and high contact professions, such as taxi drivers or security professionals, were shown to carry a heightened risk of exposure to infection. Living in urban and more deprived areas was an additional risk. In major cities, infection rates were initially higher than in rural settings, and more people reported participation in essential daily activities such as using public transport and attending work or education.^{[footnote 31]} Although to some extent this trend has persisted throughout the pandemic, urban areas benefitted from a great deal of national attention and consequent mitigation measures. Rural areas, which had largely been spared in earlier waves, came to experience high incidence in later waves due to lower immunity levels after most national public health control measures had lifted.

Crowded and multi-generational housing is a further risk factor commonly linked to infectious disease spread.^{[footnote 32]} Overcrowded housing is linked to socio-economic status and in the UK is more common in Bangladeshi, Pakistani and black African groups compared to white British.^{[footnote 33]} Importantly too, shared accommodation settings such as those for people experiencing homelessness and rough sleeping presented a significant risk of transmission for an already highly vulnerable population experiencing multiple existing socio-economic pressures and health needs.^{[footnote 34]} The ‘Everyone In’ initiative, launched in March 2020, aimed to provide safe accommodation for people experiencing homelessness and rough sleeping and was widely credited with saving lives during the pandemic.^{[footnote 35]}

Severe disease and mortality

Since the start of the pandemic, age has been the strongest risk factor for COVID-19 hospital admission and mortality,^{[footnote 36]} with older adults at high risk and children and young people at very low risk of severe outcomes.^{[footnote 37]} Mortality rates from COVID-19 in the most deprived areas of the country were more than double that found in the least deprived areas, with differences remaining after adjustment for age, sex, region and ethnicity. As a single group, ethnic minorities experienced higher all-cause death rates and death rates from COVID-19 compared to those of white British ethnicity, with relative differences varying throughout the pandemic and across different ethnic groups.^{[footnote 38]} In the working-age population, COVID-19 death rates were consistently and markedly higher for men than women throughout the pandemic.^{[footnote 39]}

Another group at particularly high risk for severe disease and premature mortality were those with a disability. In the first wave, 6 out of 10 deaths in England were among people who reported having a disability.^{[footnote 40]} Research based on the learning disability register found a persistent, marked increased risk in COVID-19 hospitalisation and mortality for people with a learning disability – though it is important to note that there are major limitations with the learning disability register as a robust assessment tool, with wider coding for learning disability, and that not all analyses adjusted for underlying health conditions.^{[footnote 41]}

Co-morbidities such as diabetes, severe asthma and obesity were identified as risk factors for poor outcomes, and were more prevalent in more deprived and in some ethnic minority groups. Linked primary care records of over 17 million adults with over 10,000 deaths between February and December 2020 found that while comorbidity did explain some of the different death rates by ethnicity, people from black and South Asian ethnic groups were both more likely to test positive and more likely to die from COVID-19 during the first wave compared with people from white ethnic groups after adjustment for deprivation, age, sex and comorbidity.^{[footnote 42]} Analysis of the second wave found that while differences in testing positive and higher death rates among South Asian ethnic groups remained, they were far less stark for black ethnic groups.

Disentangling the principal drivers was often complex because of the overlapping nature of many of the risk factors. For example, some South Asian populations might have higher probability of being in contact professions such as taxi driving or care work, higher rates of diabetes, more multigenerational households and being in an area of enduring transmission such as in the north-west of England. Some populations may use care and testing differently or face barriers in their access. Working out which was a risk factor and which was a confounding factor was inevitably complex and some residual confounding was likely.

Impact of public health measures

High case rates during the pandemic led to pressures on health and care services which in turn impacted different population groups in need of health and care support. Measures put in place to mitigate transmission, too, impacted interactions with health and care services for many – for example, visiting restrictions. This is covered in more detail in Chapter 10: improvements in care. Non-COVID-19 clinical harms were worse for some groups. For example, there was a greater reduction in routine elective admissions for care home residents compared to the general population, and routine referrals to hospital care fell 90% for children and young people in the first wave.^{[footnote 43]}, ^{[footnote 44]}

Many people saw a deterioration in mental health during the pandemic; the impact was particularly felt in some groups, such as women who reported worse mental health during the pandemic than men.^{[footnote 45]} Disparities in mental health outcomes in unemployed people and those experiencing financial insecurity widened during the pandemic.^{[footnote 46]} The public health response to the pandemic had wider impacts on the economy, wellbeing and education. Children and young people missed significant amounts of face-to-face education with impacts including lost learning, poor mental health and a reduction in the number of safeguarding referrals.^{[footnote 47]}

Widespread closures in sectors such as hospitality, leisure and tourism had significant economic impacts for individuals employed in these sectors, a greater proportion of whom were women. People in ethnic minorities were also more likely to work in insecure and casual forms of employment which were impacted by pandemic control measures. While the Coronavirus Job Retention Scheme (‘furlough’) provided some protection against unemployment, individuals on furlough experienced a 20% reduction in wages and this was more common for people on low-income wages and part-time workers.^{[footnote 48]} Rural and coastal areas were disproportionately impacted by some of the public health measures used to control spread, with these areas experiencing:

• an increased impact on hospital waiting times
• a reliance on the tourism and hospitality sector
• high levels of digital exclusion and an ageing population ^{[footnote 49]}

Areas of enduring transmission, such as Leicester and the north-west of England, were also disproportionately impacted by both continual transmission and long-running measures to bring this down – for example, in disruption to education.

The reasons for these disparities are complex and involve a range of social, economic, behavioural and biological risks.^{[footnote 50]} Disparities were the result of a complex interaction between existing disparities, the progression of the epidemic across the country (for example, which areas saw early seeding of infection), and the measures taken to control disease spread. For some communities, a relative lack of trust in government or the health service resulted in mistrust of national communications, which was compounded by disparities exposed by the pandemic. At times responses and communications were not appropriately tailored to different communities. This was sometimes exacerbated by interventions directly aimed at certain higher risk groups, leading to actual or potential stigmatisation by implying certain groups were more vulnerable to COVID-19 or more likely to transmit the virus. Tailoring messages for the highest risk groups without increasing stigma can be a very difficult balance to navigate in epidemics and pandemics, and was particularly important in the earlier stages (for example, for the Chinese community). It was seen previously – for example, in HIV and more recently in Monkeypox.

What was done in response

This sets out some elements of the response but is by no means exhaustive. Efforts to minimise disparities sat across a number of organisations and individuals and continue to evolve today.

Following publication of the PHE report on COVID-19 disparities in risks and outcomes, the Cabinet Office Race Disparity Unit was tasked to lead cross-government work to address the report findings, with the activities undertaken summarised in a series of reports.^{[footnote 51]}

Actions to address disparities initially focused on reducing the risk of infection, for example, the government published guidance on how to make workplaces more secure for individuals unable to work from home, including specific practical guidance for occupations at higher risk of exposure such as taxi drivers. Guidance and infographics for the public were translated into the most commonly spoken languages, and communications campaigns worked closely with the third sector to ensure local dissemination into communities.

Throughout the pandemic, different testing programmes were implemented to address certain disparities. This included mass asymptomatic testing programmes in care homes, the NHS and across the education sector as well as targeted community testing in areas of high or enduring transmission. Targeted community testing programmes were delivered through local authorities to benefit from in-depth knowledge of local community needs, trusted voices and detailed local data.^{[footnote 52]}

Other efforts to tackle COVID-19 disparities were focused on building vaccine confidence and promoting vaccine uptake among those groups that were more hesitant about vaccination. This required detailed discussions to unpick where the issues were. Delivery of the mass vaccination programme and targeted work with specific communities has been a result of a partnership approach between national and local government, health agencies, and the voluntary and community sector. One key component of this response was the Community Champions scheme launched in January 2021 which enabled councils and voluntary organisations to develop local networks of trusted local champions to provide advice about COVID-19 and the vaccine programme.^{[footnote 53]}

Discussion

Disparities in COVID-19 arose because of differences in infection risk, risk of severe disease or mortality, non-COVID-19 clinical harms and the wider impacts of public health measures to control the pandemic. The pattern of disparities highlighted the need to consider, as much as possible, disparities according to the following determinants:

1. Protected characteristics: as defined by the Equality Act (2010):

• age
• disability
• gender reassignment
• marriage and civil partnership
• pregnancy and maternity
• race, religion or belief
• sex and sexual orientation

2. Socio-economic circumstances such as:

• deprivation
• occupation (particularly key workers)
• geographical region

3. Inclusion health groups: those who have been socially excluded typically experience multiple overlapping risk factors for poor health (such as poverty, violence and complex trauma), experience stigma and discrimination, and are not consistently accounted for in routine data sets. In the UK, the concept of inclusion health has typically encompassed people experiencing homelessness, Gypsy, Roma and Traveller communities, vulnerable migrants and sex workers, among others.^{[footnote 54]}

It was essential to gather data and information about the existence and drivers of disparities in this pandemic – both quantitative and qualitative – and this required multiple different methodologies. Key informant interviews and focus group discussions, and more generally early engagement with communities, were vital to effectively tailoring interventions and anticipating future challenges in implementing any large-scale intervention. It was a resource-intensive method but has held great value. Access to major data sets (for example, via ‘OpenSafely’) has also enabled continuous surveillance and research on clinical and health outcomes from COVID-19, though there were occasional issues with sharing, linkage and timeliness of data (this is covered in more detail in Chapter 4: situational awareness, analysis and assessment). In the future, this could be better supported by joint working with and between local government, the health service (in particular data and digital teams), central government and academia.

A routine approach to evaluation and research on the direct and indirect benefits and harms of the public health response on local population groups and communities was also important from the start of the pandemic. This could help identify disparities more rapidly and facilitate the rapid adaptation of interventions to better meet the needs of specific population groups and minimise harms.

The data and information needed to understand disparities was often sensitive and was being asked of communities with relatively low trust in government organisations and understandable concerns about privacy and the use of their data. It has therefore been important, as it will continue to be beyond this pandemic, to engage closely with communities and to work with trusted organisations to understand disparities and avoid extractive methods of research in favour of close engagement and coproduction. This is likely to be true of any future pandemic or epidemic.

It was also important to empower and adequately resource local areas to adapt and respond to the specific needs of their communities, designing and implementing approaches with the communities most impacted. The long-term impacts of these disparities are yet to be fully felt, and an inclusive pandemic recovery programme will be key to ensuring that the same populations disproportionately impacted during the pandemic will not suffer ongoing disparities throughout the subsequent socio-economic recovery.^{[footnote 55]}, ^{[footnote 56]}

Our response to future pandemics will be strengthened by understanding these long-term effects and by improving our understanding of the key drivers of health and inequalities, and of the different needs of different communities. This will enable local and national policymakers to improve community resilience between pandemics to better mitigate harms in the future.

The findings of this pandemic have led to a renewed effort to address pervasive inequalities in health in some areas – for example, in the work of the NHS in recovery including equity audits of waiting lists and the Duty to Reduce Inequalities on the emerging integrated care boards in England. The pandemic has reinforced the message seen in many previous pandemics that those already marginalised, socio-economically disadvantaged and suffering poorer health outcomes are likely to be at increased risk during a pandemic. Routine data sets (particularly for health and care), surveillance systems, research and planning exercises therefore need to involve these groups while keeping flexible to evolving evidence on the specific risk factors for any new pathogen in the future.

Reflections and advice for a future CMO or GCSA

Point 1

This pandemic, in common with many others, reflected and in many cases exacerbated existing inequalities.

Understanding how the combination of existing inequalities and pathogen-specific vulnerabilities affect individuals across the population was essential to inform the policy and public health responses.

Point 2

Research on where the disparities were, what their causes were and how best to reduce them needed to begin from the outset of the pandemic.

Some signals only come when the epidemic reaches a particular stage or hits a particular area. Some disparities also changed with the changing epidemic (for example, as waves hit different areas of the country).

Point 3

A wide range of qualitative and quantitative research methods were needed to understand disparities.

These included:

• population-level surveillance
• research directly with affected groups
• surveys and in-depth reviews alongside routine data sets

Properly completed demographic and other fields in a range of data sets, and linkage of data, were particularly important in understanding disparities – and this can be strengthened in ‘peacetime’.

Point 4

Continual dialogue with local communities was important in understanding risks and vulnerabilities, and to co-design effective responses at a hyper-local level that may not be picked up in larger, national data sets or research.

References

1. Hayward, S; Harding, R; McShane, H, et al. Factors influencing the higher incidence of tuberculosis among migrants and ethnic minorities in the UK. F1000Research 2018; 7:461 ↩

2. Gwenda Hughes and Nigel Field. The epidemiology of sexually transmitted infections in the UK; impact of behaviour, services and interventions. Future Microbiology, Vol.10, No 1. 19 Jan 2015 https://doi.org/10.2217/fmb.14.110 ↩

3. Kumar, S and Quinn S. Health Inequalities and Infectious Disease Epidemics: A Challenge for Global Health Security. Biosecurity and Bioterrorism. 2014;12(5):263-273 ↩

4. Williamson, E; Walker, A; Bhaskaran, K; et al. OpenSAFELY: factors associated with COVID-19 death in 17 million patients. Nature. 2020;584(7821):430-436 ↩

5. Taubenberger, J.K and Morens, D. 1918 Influenza; the Mother of All Pandemics. Emerging Infectious Disease. 2006. Jan; 12(1):15-22 ↩

6. Suhrcke, M; Stuckler, D; Suk, J; et al. The impact of economic crises on communicable disease transmission and control: A systematic review of the evidence. Plos One. Published: June 10, 2011 ↩

7. Mamelund S and Dimka J. Social inequalities and infectious diseases. Scandinavian Journal of Public Health. 2021;49(7):https://journals.sagepub.com/doi/full/10.1177/1403494821997228 ↩

8. Semenza, J; and Giescke, J. Intervening to reduce inequalities in infections in Europe. American Journal of Public Health. May 2008. 98, 787-792 https://doi.org/10.2105/AJPH.2007.120329 ↩

9. Jouni Paavola. Health impacts of climate change and health and social inequalities in the UK. Environmental Health 16 (Suppl 1), 113 (2017). https://doi.org/10.1186/s12940-017-0328-z ↩

10. Aldridge, R; Pineo, H; Fragrasy, E; et al. Household overcrowding and risk of SARS-CoV-2: analysis of the Virus Watch prospective community cohort study in England and Wales [version 1; peer review: 1 approved with reservations]. Wellcome Open Research. 2021;6:347 ↩

11. Powell, A; Francis-Devine; and Clark, H. Coronavirus: impact on the labour market. House of Commons Library. [Internet]. [cited 7 September 2022]. Available from: CBP-8898.pdf (parliament.uk). ↩

12. Verity, R; Okell, L; Dorigatti, I; et al. Estimates of severity of coronavirus disease 2019: a model-based analysis. Lancet Infectious Disease 2020; 20: 669-77. ↩

13. Huang, C; Wang, Y; Li, X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The Lancet. 2020. Vol 395, Issue 10223, 15-21, pages 497-506. ↩

14. Docherty, A; Harrison, E; Green, C; et al. Features of the 20,133 UK patients in hospital with COVID-019 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational study. BMJ 2020; 369, Published 22 May 2020 ↩

15. Shi Y, Yu X, Zhao H, et al. Host susceptibility to severe COVID-19 and establishment of a host risk score: findings of 487 cases outside Wuhan. Crit Care 24, 108 (2020). https://doi.org/10.1186/s13054-020-2833-7 ↩

16. Public Health England. ‘The First Few Hundred (FF100)’ Enhanced Case and Contact Protocol v6.3. Epidemiological protocols for comprehensive assessment of Early Middle East Respiratory Syndrome coronavirus cases and their close contacts in the United Kingdom. [Internet]. Published December 2015. [cited 7 September 2022].Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/772618/2015_2016_FF100_Protocol_MERSCoV_V6_3_2015527.pdf ↩

17. Public Health England. ‘The First Few Hundred (FF100)’ Enhanced Case and Contact Protocol v6.3. Epidemiological protocols for comprehensive assessment of Early Influenza A (H7N9) cases and their close contacts in the United Kingdom. [internet] [cited 7 September 2022]. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/360190/2012_13_FF100_Protocol_H7N9_ver_12.pdf ↩

18. Boddington, N; Charlett, A; Elgohari, S, et al. Epidemiological and clinical characteristics of early COVID-19 cases, United Kingdom of Great Britain and Northern Ireland. Bulletin World Health Organisation. 2021 Mar 1; 99(3): 178-189. Published online 2020 Nov 30. doi: 10.2471/BLT.20.265603 ↩

19. EMG transmission Group. COVID-19 Risk by occupation and workplace. 11 February 2021. [Internet]. [cited 7 September 2022].Available from: COVID-19 risk by occupation and workplace (publishing.service.gov.uk) ↩

20. ICNARC. ICNARC report on COVID-19 in critical care: 27th March 2020. 2020 https://www.icnarc.org/DataServices/Attachments/Download/b5f59585-5870-ea11-9124-00505601089b ↩

21. Intensive Care National Audit and Research Centre (ICNARC). ICNARC report on COVID-19 in critical care [Internet]. April 2020 [cited 17 October 2022]. Available from: https://www.icnarc.org/DataServices/Attachments/Download/c31dd38d-d77b-ea11-9124-00505601089b ↩

22. Public Health England. Disparities in the risk and outcomes of COVID-19. [Internet]. August 2020. [cited 7 September 2022]. Available from: https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

23. Cabinet Office Race Disparity Unit. Final report on progress to address COVID-19 health inequalities. [Internet]. Published 3 December 2021. [cited 7 September 2022]. Available from: Final report on progress to address COVID-19 health inequalities - GOV.UK (www.gov.uk) ↩

24. Public Health England. Beyond the Data: Understanding the impact of COVID-19 on BAME groups. June 2020. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/892376/COVID_stakeholder_engagement_synthesis_beyond_the_data.pdf ↩

25. Deaths involving COVID-19 by local area and socioeconomic deprivation: deaths occurring between 1 March and 17 April 2020 ↩

26. University College London. VIVALDI study. 2022. Available from: VIVALDI Study, UCL Institute of Health Informatics, University College London ↩

27. Lewer D, Braithwaite I, Bullock M, Eyre M, White P, Aldridge R, Story A, Hayward A. ‘COVID-19 among people experiencing homelessness in England: a modelling study. Lancet Respiratory Health, 2020 Dec. Vol 8, Iss 12. DOI: https://doi.org/10.1016/S2213-2600(20)30396-9 ↩

28. Public Health England. COVID-19 Health Inequalities Monitoring for England (CHIME). 2022. Available from: https://analytics.phe.gov.uk/apps/chime/ ↩

29. See, for example, published minutes for SAGE 39 at which paper on high-risk institutional settings was reviewed. SAGE Transparency data: SAGE 39 minutes: Coronavirus (COVID-19) response, 28 May 2020. Available from: https://www.gov.uk/government/publications/sage-39-minutes-coronavirus-covid-19-response-28-may-2020/sage-39-minutes-coronavirus-covid-19-response-28-may-2020#high-risk-institutional-settings ↩

30. Department of Health. Public Health Surveillance. Towards a Public Health Surveillance Strategy for England. Published December 2012.[Internet]. [cited 7 September 2022].Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/213339/Towards-a-Public-Health-Surveillance-Strategy.pdf ↩

31. Beale, S; Braithwaite, I; Navaratnam, A; et al. Deprivation and exposure to public activities during the COVID-19 pandemic in England and Wales. Journal of Epidemiology and Community Health. 2022;76:319-326. Available from: Deprivation and exposure to public activities during the COVID-19 pandemic in England and Wales. Journal of Epidemiology & Community Health (bmj.com) ↩

32. Public Health England. Disparities in the risk and outcomes of COVID-19. [Internet]. August 2020. [cited 7 September 2022]. Available from: https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

33. Gov.uk. Overcrowded households. 2020. Available from: Overcrowded households - GOV.UK Ethnicity facts and figures (ethnicity-facts-figures.service.gov.uk) ↩

34. Hayward, A and Storey, A. COVID-19 in inclusion health populations. Prepared for SAGE by UCL Collaborative Centre for Inclusion Health. 2022. Available from: https://www.gov.uk/government/publications/ucl-collaborative-centre-for-inclusion-health-covid-19-in-inclusion-health-populations-4-june-2020 ↩

35. House of Commons Library. Research briefing: Coronavirus: Support for rough sleepers (England). 2021. Available from: https://commonslibrary.parliament.uk/research-briefings/cbp-9057/ ↩

36. Public Health England. Disparities in the risk and outcomes of COVID-19. [Internet]. August 2020. [cited 7 September 2022]. Available from: https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

37. Ward, J; Harwood, R, Smith, C et al. Risk factors for PICU admission and death among children and young people hospitlisaed with COVID-19 and PIMS-TS in England during the first pandemic year. Nature Medicine. 20 December 2021. ↩

38. Public Health England. Disparities in the risk and outcomes of COVID-19. [Internet]. August 2020. [cited 7 September 2022]. Available from: https://www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes ↩

39. Public Health England. COVID-19 Health Inequalities Monitoring for England (CHIME). 2022. Available from: https://analytics.phe.gov.uk/apps/chime/ ↩

40. Office for National Statistics. Updated estimates of coronavirus (COVID-19) related deaths by disability status, England: 24 January to 20 November 2020. 2022. Available from: Updated estimates of coronavirus (COVID-19) related deaths by disability status, England - Office for National Statistics (ons.gov.uk) ↩

41. Williamson, E; McDonald, H; Bhaskaran, K; et al. Risks of covid-19 hospital admission and death for people with learning disability: population based cohort study using the OpenSAFELY platform. British Medical Journal. 2021;374:n1592. Available from: Risks of covid-19 hospital admission and death for people with learning disability: population based cohort study using the OpenSAFELY platform. The BMJ ↩

42. Williamson, E; Walker, A; Bhaskaran, K; et al. OpenSAFELY: factors associated with COVID-19 death in 17 million patients. Nature. 2020;584(7821):430-436 ↩

43. Grimm, F; Hodgson, K; Brine, R; and Deeny, S. Hospital admissions from care homes in England during the COVID-19 pandemic: a retrospective, cross-sectional analysis using linked administrative data. International Journal of Population Data Science. 2021;5(4) doi: 10.23889/ijpds.v5i4.1663. Available from: Hospital admissions from care homes in England during the COVID-19 pandemic: a retrospective, cross-sectional analysis using linked administrative data. International Journal of Population Data Science (ijpds.org) ↩

44. Morris, J and Fisher, E. Growing problems, in depth: The impact of Covid-19 on health care for children and young people in England. The Nuffield Trust. 2022. Available from: Growing problems, in depth: The impact of Covid-19 on health care for children and young people in England. The Nuffield Trust ↩

45. Office for Health Improvement and Disparities (OHID). Gender Spotlight [Internet]. In: COVID-19: mental health and wellbeing surveillance report. November 2021 [cited 17 October 2022]. Available from: https://www.gov.uk/government/publications/covid-19-mental-health-and-wellbeing-surveillance-spotlights/gender-covid-19-mental-health-and-wellbeing-surveillance-report ↩

46. The Parliamentary Office of Science and Technology (POST). Mental health impacts of the COVID-19 pandemic on adults [Internet]. July 2021 [cited 17 October 2022]. Available from: https://researchbriefings.files.parliament.uk/documents/POST-PN-0648/POST-PN-0648.pdf ↩

47. Viner R, Russell S, Saulle R, Croker H, Stansfield C, Packer J, Nicholls D, Goddings AL, Bonell C, Hudson L, Hope S. School closures during social lockdown and mental health, health behaviors, and well-being among children and adolescents during the first COVID-19 wave: a systematic review. JAMA Pediatr. 2022 [cited 17 October 2022] 176(4):400–409. doi:10.1001/jamapediatrics.2021.5840. Available from: https://jamanetwork.com/journals/jamapediatrics/article-abstract/2788069 ↩

48. Resolution Foundation. The effects of the coronavirus crisis on workers [Internet]. May 2020 [cited 17 October 2022]. Available from: https://www.resolutionfoundation.org/publications/the-effects-of-the-coronavirus-crisis-on-workers/ ↩

49. Nuffield Trust. Rural, remote and at risk: Why rural health services face a steep climb to recovery from Covid-19 [Internet]. December 2020 [cited 17 October 2022]. Available from: https://www.nuffieldtrust.org.uk/research/rural-remote-and-at-risk ↩

50. Bambra C, Riordan R, Ford J, Matthews F. The COVID-19 pandemic and health inequalities. J Epidemiol Community Health. 2020 Nov 1;74(11):964-8 [cited 17 October 2022]. Available from: https://jech.bmj.com/content/jech/74/11/964.full.pdf ↩

51. Cabinet Office Race Disparity Unit. Final report on progress to address COVID-19 health inequalities. [Internet]. Published 3 December 2021. [cited 7 September 2022]. Available from: Final report on progress to address COVID-19 health inequalities - GOV.UK (www.gov.uk) ↩

52. Department of Health and Social Care. Press Release: Targeted community testing extended to 10 more areas [Internet]. December 2020 [cited on 17 October 2022]. Available from: https://www.gov.uk/government/news/targeted-community-testing-extended-to-10-more-areas ↩

53. Public Health England. Community champions: A rapid scoping review of community champion approaches for the pandemic response and recovery [Internet]. August 2021 [cited 17 October 2022]. Available from: https://www.gov.uk/government/publications/community-champion-approaches-rapid-scoping-review-of-evidence ↩

54. Public Health England. Inclusion Health: Applying All Our Health [Internet]. May 2021 [cited 17 October 2022]. Available from: https://www.gov.uk/government/publications/inclusion-health-applying-all-our-health/inclusion-health-applying-all-our-health ↩

55. Suhrcke M, Stuckler D, Suk JE, Desai M, Senek M, McKee M, Tsolova S, Basu S, Abubakar I, Hunter P, Rechel B. The impact of economic crises on communicable disease transmission and control: a systematic review of the evidence. PloS one. 2011 Jun 10;6(6):e20724 [cited 17 October 2022]. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0020724 ↩

56. UCL Institute of Health Equity. Build Back Fairer: The COVID-19 Marmot Review [Internet]. December 2020 [cited 17 October 2022]. Available from: https://www.health.org.uk/publications/build-back-fairer-the-covid-19-marmot-review ↩