Intended for healthcare professionals

Analysis Racism in Medicine

Transforming the health system for the UK’s multiethnic population

BMJ 2020; 368 doi: (Published 11 February 2020) Cite this as: BMJ 2020;368:m268

Read all of the articles in our special issue on Racism in Medicine

  1. Sarah Salway, professor of public health1,
  2. Daniel Holman, research fellow1,
  3. Caroline Lee, research associate2,
  4. Victoria McGowan, research associate3,
  5. Yoav Ben-Shlomo, professor of clinical epidemiology4,
  6. Sonia Saxena, professor of primary care5,
  7. James Nazroo, professor of sociology6
  1. 1Department of Sociological Studies, University of Sheffield, Sheffield, UK
  2. 2Cambridge Institute of Public Health, University of Cambridge, Cambridge, UK
  3. 3Institute of Health and Society, Newcastle University, Newcastle, UK
  4. 4Bristol Medical School, Population Health Sciences, University of Bristol, Bristol, UK
  5. 5School of Public Health, Imperial College London, London, UK
  6. 6Cathie Marsh Institute for Social Research, University of Manchester, Manchester, UK
  1. Correspondence to: S Salway s.salway{at}

The UK health system must take urgent action to better understand and meet the health needs of migrants and ethnic minority people, say Sarah Salway and colleagues

Ethnic diversity is a global phenomenon resulting from historical and contemporary movements of people. However, healthcare policy makers, practitioners, and researchers have been slow to wake up to this reality. We urgently need to improve our understanding of, and responses to, the health needs of mobile and ethnically diverse populations.

As a country with a colonial past, a long history of migration, well established minority ethnic groups, and high investment in health related research, the UK should be leading the way in evidence informed, equitable healthcare. In reality, however, the UK has significant shortcomings. We believe the health system’s failure to respond to ethnic diversification reflects a deeper, politically led, ambivalence towards the notion of multiethnic UK. Policy makers, practitioners, and researchers can and should challenge the persistent marginalisation of this agenda.

Failure of policy and practice

There is widespread evidence that UK policy responses to ethnic diversity are ambivalent, fragmented, confused, and often harmful. This is despite the apparently strong legal framework of the 2010 Equalities Act.

Initiatives such as the Race Disparity Unit, established in 2018 by the former prime minister Theresa May, suggest a desire to tackle inequalities. However, other policy strands undermine equality. For instance, the “community cohesion” policy adopting the language of “British values” has legitimised concerns about imagined threats to cultural identity and security posed by ethnic minority communities.1 Hostility from nationalist politicians and the press encourages the scapegoating of migrants and ethnic minority people as the root of society’s problems.2

Health policy and practice do not exist in a vacuum. It is therefore unsurprising that we find a similarly patchy and ambivalent picture. Recent policy documents relating to dementia and women’s mental health include welcome reference to the inequalities experienced by ethnic minority groups. However, they do little to identify remedial action.34

Other policies do not mention ethnic diversity and inequality. Such omission perpetuates a “one size fits all” mentality and ignores persistent exclusionary processes.5 The UK’s well established health inequalities agenda has repeatedly failed to embed attention to migration and ethnicity.6 The central role of racism (interpersonal, institutional, and cultural) is consistently overlooked when considering the physical and mental health of migrant and ethnic minority people (box 1).68

Box 1

Ethnicity, race, and racism

  • Ethnicity can be described as a form of collective identity that draws on notions of ancestry, cultural commonality, geographical origins, and shared physical features. Ethnic identities are not fixed or natural. They are social constructs fluid across space and time. Individuals may self identify with several or none of the ethnic categories used in government statistics or research. Processes of ethnic identification are also externally imposed. Ethnic categories acquire meaning in particular contexts. Societal structures and ideologies reinforce ethnic boundaries and the illusion that differences between ethnic groups are immutable7

  • The term race is used less often than ethnicity in UK health policy and research. However, these terms are used variably across the world. Both concepts are social constructions, and both invoke physical as well as cultural distinctions. Both are linked to processes of exclusion and differential access to power, resources, opportunities, and status

  • Racism has a central role in shaping the health of minority ethnic and migrant people. Processes of racism are entrenched in laws, policies, and practices. They are reproduced within societal institutions and organisations. The circulation of images, language, and symbols within everyday conversation, media, and policy perpetuates an ideology of inferiority and difference. Interlocking structural and cultural processes result in differential access to health promoting resources, exposure to health risks, and access to healthcare


Some important health problems that disproportionately affect ethnic minority groups are ignored by national policy (for example, the higher risk of hepatitis B among people who have migrated from east Asia).9 Other health policies can stigmatise ethnic minority populations rather than promote culturally competent action on inequity. One example is safeguarding legislation against female genital mutilation, which has led to negative stereotyping of the Somali community in Bristol.10

Research reveals large ethnic inequalities in healthcare. Lower uptake and poorer satisfaction with care have been documented among ethnic minority groups across diverse NHS contexts.1112 Maternity and mental health services show particularly worrying and persistent ethnic inequalities in experiences and outcomes.1314 We also know that healthcare practitioners often feel ill equipped and poorly supported to meet the needs of ethnically diverse patient groups.15

Increasingly stringent rules around entitlement to NHS care for migrants have also prompted concern.161718 There are, nevertheless, some positive examples, such as the Desmond diabetes programme adaptation, which responds to diversity in the patient’s preferred language, dietary, and religious practices.19 Unfortunately, scale-up of such innovations is patchy. Spread is often hampered by short term funding, unrealistic targets, expectations of cost savings, and failure to share learning.19

Inadequate data

Past research has shown that progress on this agenda requires much more investment in generating and applying evidence,2021 and our recent work confirms this picture. Using desk based reviews, interviews, group discussions, and deliberative stakeholder workshops, we found a combination of data gaps, low demand for evidence, and a failure to act on available knowledge. The resultant reinforcing cycle perpetuates the marginalisation of ethnic minority and migrant health.

Participants in our workshops voiced concerns about inadequacy of data. Poor data availability and quality undermines our ability to describe and understand health and healthcare among ethnic groups and by migrant status. A recent Public Health England technical report confirms substantial data gaps,22 as does the Race Disparity Audit.23 The Health Survey for England,24 one of our most important resources, has not focused on ethnic minority people since 2004. Other data collection initiatives with the potential to increase understanding of processes linking ethnicity to health have ceased (such as the Citizenship Survey).25

In England, joint strategic needs assessments are central to establishing local authority health profiles and priority action plans. We reviewed assessment documents from 32 local authority areas, selected to provide geographical distribution and with varied demographic profiles, and found a combination of data gaps and missed opportunities to use data. Some local authorities had taken targeted “deep dive” assessments to understand their local population. However, ethnicity and migration were not embedded in the assessment processes. This was true even in areas with long established ethnic minority populations and high ethnic diversity. Racism was rarely identified as an important determinant of poor health.

Despite some improvement, the completeness and accuracy of ethnicity recording within routine health data systems also remains patchy.2627 NHS ethnic categories have not been revised with census updates. Migration status is not routinely recorded. The conduct of health equity audits has dwindled in recent years. Most services simply do not routinely consider whether they are meeting the needs of different ethnic and migration status groups within their catchment populations. Equality concerns continue to be distant from the healthcare quality agenda.28

The picture is even starker when we look for evidence to inform action on unmet need and health inequalities. We examined the research funded by four of the National Institute for Health Research (NIHR) funding streams (public health research, health services and delivery research, research for patient benefit, and health technology assessment) from 1995 to 2017. It could be argued that all research intended to inform the design and delivery of health services for the UK’s multiethnic population should consider ethnicity. In fact, we found that only 8-10% of funded research projects mentioned ethnicity related terms in their title or abstract. Furthermore, out of 2658 funded projects, just 38 had a clear focus on health needs linked to ethnicity or migration.

Participants in our deliberative stakeholder workshops have identified several factors contributing to this low volume. These include a lack of demand for evidence from policy makers and limited visibility of this field in leading medical journals. Limited competence and confidence among researchers and low representation of ethnic minority researchers within academia also contribute. Participants also thought that charities and social science funding streams were more ready to fund such research than NIHR.

In advocating for more research, quality is key. Earlier commentaries and reviews have cautioned that poorly conducted research can do more harm than good.2930 Workshop participants reiterated the importance of underpinning all research in this field with conceptual rigour; meaningful involvement of patients and the public; and concrete plans for moving knowledge into action.

Much biomedical and health services research fails to adequately recognise the contextual and multidimensional nature of ethnic (and migrant) identities. It often falls back on untested narrow cultural or biological explanations for patterns of health and healthcare, rather than uncover a more complete causal picture. There is also a need to acknowledge that ethnic categories encompass diverse groups of people with highly variable health related assets and opportunities. Similarly, many interventions aimed at meeting the needs of migrant or ethnic minority groups lack sound theoretical underpinning.29 Crucially, the role of racism must be more consistently addressed if we are to develop effective interventions at individual and societal levels.8

Stronger national leadership

There are signs that some national and local leaders in the health system are beginning to take ethnic inequality seriously. This collection of articles in The BMJ ( is clearly one such indicator. Integrated care systems present an opportunity to enhance equity, routinely identify those who are not receiving services, and hold leaders to account. Local innovations are appearing around the country, such as drop-in GP clinics for newly arrived migrants,31 yet more can be done to learn from and scale up such solutions. More doctors and healthcare practitioners should be encouraged to advocate for underserved migrant and minority patient groups. More practitioners from ethnic minority groups are required at senior levels of the NHS, though there is evidence of slow improvements. 32

Closing the gaps

Recent publication of relevant Public Health England resources223233 and the Race Disparity Audit23 are welcome signs that the need to investigate ethnic inequality is increasingly recognised. In addition, concerted effort is needed to plug data gaps and reinstate routine equity audits in order to identify who is not benefiting from services. We can learn from local authorities that are undertaking strong assessment work to increase understanding of local health needs. For example, in Nottingham and Leicester there has been good engagement with ethnic minority groups to identify health concerns.3435 These authorities also developed clear recommendations to act on gaps in data and to implement longer term strategies around quality standards, monitoring and evaluation, financial planning, and specific provision and coordination between services.

Improvements to data collection must go hand in hand with assurance of data protection. Recent data sharing between NHS Digital, the Department of Health and Social Care, and the Home Office for immigration enforcement presented substantial risks to the health and wellbeing of migrant and settled ethnic minority communities and must not resume.

UK health research can learn from elsewhere. In the US, all clinical research funded by the National Institutes of Health (NIH) must include women and minority groups (when appropriate to the research question). Additionally, the National Institute on Minority Health and Health Disparities is focused on eliminating disparities.36

Recent UK developments that have the potential to influence the way evidence is generated include the Equality Diversity and Inclusion in Science (EDIS) network37 and the diversity and inclusion workstream of INVOLVE, a national advisory group set up to promote public involvement in research.38 The results of research projects examining key challenges should be widely disseminated. These include projects on the meaningful involvement of ethnic minority people,39 understanding and tackling discrimination in health services,40 and data linkage to improve understanding of ethnic health inequalities.41

We need to increase the visibility and status of research in this area. The demand for evidence among decision makers must be boosted. Effective channels for routinely applying evidence to policy and practice are also needed. The recently launched NIHR applied research collaboration (ARC) for East Midlands identifies “ethnicity and health inequalities” as a priority; we hope this may influence the other 14 regional ARCs. We should also look to countries such as Norway, where research-policy partnership structures facilitate ongoing dialogue around these health needs.42

Rather than countering discriminatory processes of wider society, the UK health system often mirrors the forces that undermine the health of migrants and ethnic minority people. We overlook, misconstrue, and respond poorly to the health needs of these groups. An inadequate knowledge base contributes to this unacceptable situation. We need radical action to increase the creation of high quality research evidence and data at local, regional, and national levels. Such knowledge must be routinely expected and used to inform action. Stronger national leadership is required. This must be coupled with greater involvement of ethnic minority people and sustained support for local innovators who can lead the way.

Key messages

  • The UK population is increasingly diverse as a result of both immigration and natural growth

  • Health policy and healthcare for minority groups is patchy, sometimes stigmatising, and rarely culturally sensitive

  • Important gaps in routine data, national surveys, and commissioned research mean there is inadequate evidence on how to meet the health needs of these groups

  • Well designed research with meaningful involvement of patients and the public is required to inform action

  • Stronger national leadership is needed plus support for local innovators who can lead the way towards an evidence informed, inclusive, and equitable health system


We thank all those who participated in interviews, group discussions, and workshops, as well as the members of our patient and public involvement group, for their valuable contributions to the project.


  • For more articles in The BMJ’s Racism in Medicine special issue see  

  • Contributors and sources: This article arose from an NIHR School for Public Health Research funded project led by SSal (grant reference number PD-SPH-2015). The project included a review of policy documents at local and national government levels; a review of research grants awarded by NIHR funding schemes; a series of interviews and group discussions with public health and healthcare practitioners; and two deliberative workshops (one with researchers and national level policy makers, and one with mainly local level practitioners and members of the public). The authors are based in the UK and span a range of disciplines, including anthropology, sociology, public health, epidemiology, social statistics, and primary care. SSal conceived the project and drafted the manuscript and is guarantor. SSal, DH, CL, and VMcG were directly involved in all aspects of the project. YBS and SSax gave technical support to all aspects of the project. YBS and JN contributed to one of the deliberative workshops. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

  • Patient and public involvement: A public involvement group was convened to provide critical commentary on the design and conduct of the project. This comprised 13 people who self identified as having a migrant or minority ethnic identity. Group members participated in two workshop style meetings (one half day and one full day) over the course of the project.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; externally peer reviewed.