In approaching the issue of ethnicity and diversity in health we need to be aware of a complex and contentious history in the evolution of ideas and terminology.  Traditional anthropology defined four major human ‘races’, usually described as ‘Caucasian’ (‘white’ or European), ‘Negroid’ (Black or African), ‘Mongoloid’ (Asian, Chinese or Indic), and ‘Australoid’ (that is, the group of people described as ‘Aboriginal’ to Australia). These groups assumed that race was a bio-scientific concept explaining significant biological differences between populations. This concept of race is now firmly discredited by modern genetics. Over 99% of the genetic make up of human beings is common to all ethnic groups. Those differences that do exist between people and populations are minor and largely reflect superficial physical characteristics (‘phenotypes’) such as facial features, hair or skin colour. In this sense the division of people into ‘races’ reflects social decisions rather than having any real scientific justification, but is based on fallacious genetic/biological associations.

Culture is a complex social phenomenon and its definition problematic. It consists of the shared beliefs, values and attitudes that guide the behaviour of group members. The concept of ‘ethnicity’ is more complex, but recognises  that people identify themselves with a social grouping on cultural grounds including language, lifestyle, religion, food and origins. The basis of ‘ethnicity’ is thus often a tradition of common descent or intermarriage and shared culture or history. It is essential to recognise that, in a world of migration and mixing, cultures and societies are dynamic rather than fixed.

The table below compares the concepts of race, culture and ethnicity.

When considering the causes of ill health and approaches to its prevention or care, it is necessary to consider the individual at risk, or the group to which they belong, in a holistic manner. The problem is using categories that most effectively describe key factors relevant to, for example, susceptibility to poor health or health outcomes. One of the objectives of CEEHD is to compile literature which discusses this, and information on this will be found elsewhere on the website.

 Defining 'ethnic group'

 The UK Race Relations Act 1976 defined a 'racial group' as 'a group of persons defined by reference to colour, race, nationality or ethnic or national origins...' 'Ethnicity' and 'ethnic group' became more formally defined in UK law by a House of Lords decision (Mandla v Lee 1983) as relating to those with 'a long shared history and a distinct culture'. Other 'relevant' characteristics were 'a common geographic origin or descent from a small number of common ancestors; a common language; a common literature; a common religion and being a minority within a larger community'. 

For practical purposes, there is little alternative to using a selection of labels and categories. The ethnic groups identified by the Office of National Statistics in the decennial UK Census are usually adopted. The following table gives the categories used in the 1991 census and those asked in the Census in 2001. While the 2001 census uses the term 'Ethnic Group', it also makes it clear that this is seen as a matter of 'cultural background'. 

Excellent discussions of the history of the Census question, and its use, can be found elsewhere on the Web.  One of the best is to be found at the QuestionBank at the web page:
      http://qb.soc.surrey.ac.uk/topics/ethnicity/ethnicintro.htm

The most recent census questions reflect changes such as a tendency for some people of African-Caribbean origins born in Britain to determine their own identity as 'Black British'. The 2001 census also asked people about their religion, which may make it easier to make projections of the numbers of people from the main religious groups, and anticipate the needs they may bring to the health service for religious observance, diet and counselling.

Categories of ethnic group recorded in the UK Censuses of 1991 and 2001

 (Adapted from ONS forms: reproduced with permission)

In the 2001 census, there was also a question on people’s religious affiliation – although it was not compulsory. This is shown below, for information.

Figure: Question 10 of the 2001 census

 Identifying ethnicity for health studies research

In terms of epidemiological research, the tendency has been to rely upon the commonly recorded variable ‘place of birth’, normally available on death certificates. However, this has been problematic. At the time of the 1991 Census, over half the population in the ‘Black’ categories (54% Black Caribbean, 84% Black Other, and 36% Black African) were UK-born, as were half of those giving their ethnic group as Pakistani, 42% of ‘Indians’ and 37% of ‘Bangladeshis’. It is now estimated that less than 40% of the black and minority ethnic population can be identified by birthplace, and increasingly few by the birthplace of their parents.

Members of CEEHD have developed a diagrammatic model which shows how many of the health-related key concepts involved in the category ‘ethnicity’ link together, which may be helpful: