Intended for healthcare professionals


The language of ethnicity

BMJ 2020; 371 doi: (Published 23 November 2020) Cite this as: BMJ 2020;371:m4493
  1. Kamlesh Khunti, professor of primary care diabetes and vascular medicine1,
  2. Ash Routen, research associate1,
  3. Manish Pareek, associate clinical professor in infectious diseases2,
  4. Shaun Treweek, professor of health services research3,
  5. Lucinda Platt, professor of social policy and sociology4
  1. 1Diabetes Research Centre and Centre for Black Minority Health, University of Leicester, Leicester, UK
  2. 2Department of Respiratory Sciences, University of Leicester, Leicester, UK
  3. 3Health Services Research Unit, University of Aberdeen, Aberdeen, UK
  4. 4London School of Economics and Political Science, London, UK
  1. Correspondence to: K Khunti kk22{at}

Collective terms BAME and BME should be abandoned

The disproportionate impact of covid-19 on ethnic minority groups12 has attracted global attention, causing journalists, broadcasters, politicians, the public, and academics to grapple with the most appropriate way to describe people of minority ethnic backgrounds.34 This issue is critical if we are to identify, understand, and resolve the enduring inequalities in life chances affecting these communities.

Various collective terms have been used to describe ethnic minority groups, including BAME (black, Asian, and minority ethnic), BME (black and minority ethnic), ethnic minority, non-white, and people of colour. BAME and BME, both commonly used in the UK,5 are problematic as they indiscriminately combine people from different geographical, behavioural, social, and cultural backgrounds. They also focus on skin colour. Few minority people identify with these acronyms,3 and in one Twitter poll only 13% of 7775 respondents, selected BAME or BME as an appropriate term.6 So what terms should we use?

Some historical context may be helpful. Before the American civil rights movement of the mid-20th century, the term black was derogatory.7 Although activists and academics subsequently used black politically as a collective term covering all “non-white” populations8—a use maintained, for example, by the domestic violence organisation Southall Black Sisters—others associated it with people of solely African heritage.

As a result, it was argued that the use of black did not recognise the experience of South Asians in the UK.9 References to black populations were then reframed as black and Asian, and later black, Asian, and minority ethnic (BAME). Though widely used, this term has no direct counterpart in routine data collection, which relies on disaggregated categories of ethnic identity developed and regularly reviewed by the UK Office for National Statistics (ONS).1011

The ONS groupings have also been criticised,12 but BAME and BME are particularly problematic because they are often used to label people who do not identify as such.3 The term BAME also identifies black and Asian communities as inherently different from other ethnic minorities and may sustain anti-black or anti-Asian sentiments.13 Although it is unclear who “minority ethnic” includes, it is widely understood to exclude disadvantaged minorities typically coded as white, such as Gypsy, Roma, and Travellers, who despite disproportionately poor health 14 are not recognised as an ethnic group in the NHS data dictionary.15


Defining individuals as non-white does not adequately describe the cultural, social, and religious nuances that define ethnicity.16 South Asian and black people are no more homogeneous than white people in their health status, service use, socioeconomic position, 17 or social and cultural experiences.161819 For example, Black Caribbean people have a higher mortality risk from covid-19 infection than Black African people—inviting us to look beyond ethnicity for the source of such inequalities.20

However problematic ethnic categories are, scientists need them to enable systematic and comparable scientific study to examine and address social inequalities and injustice. Researchers must also avoid terminology such as race that wrongly implies inherent biological differences between ethnic groups. Disaggregating minority populations allows studies to acknowledge different histories and social and economic experiences in comparisons of outcomes.

The 18 ONS categories are a good start despite limitations such as combining all black African nationalities and ethnic groups. But even with greater granularity, analysing minority populations separately could still result in inappropriate conclusions through erroneous assumptions of common shared experience.

Where granularity isn’t possible (in studies with small sample sizes for example) we suggest that “ethnic minority groups” is a more appropriate collective term than BAME or BME, placing the focus on all minority groups regardless of skin colour.

Researchers and policy makers should use the most appropriate ethnicity label needed to effectively interpret or implement their findings. Where sample sizes are insufficient for analysis of discrete groups, researchers should carefully consider the validity and implications of combining people with heterogeneous cultural histories, experiences, health and socioeconomic profiles.

The discontinuation of BAME and BME should be supported by scientific journals and others that publish research and policy relevant to the health and social wellbeing of ethnic minority groups. This could include adopting expected standards of ethnicity categorisation, appropriate to the local context.1121

There will never be a perfect term to encompass ethnic diversity, but researchers should avoid reinforcing perceptions of homogeneity where none exists, or excluding groups that do not fit within BAME terminology. Even ethnic minority groups is an imperfect collective term, and researchers should be prepared to break it down further to ensure that their findings benefit those who need them most. Language does matter.


  • Competing interests: KK is director of the Centre for Black Minority Health and is a trustee of the charity South Asian Health Foundation and co-chair of its diabetes working group. KK is chair of the SAGE ethnicity subgroup and member of Independent SAGE. LP is a member of the SAGE ethnicity subgroup.

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