Intended for healthcare professionals

Education And Debate

Style Matters: Ethnicity, race, and culture: guidelines for research, audit, and publication

BMJ 1996; 312 doi: (Published 27 April 1996) Cite this as: BMJ 1996;312:1094

The following guidelines on the use of ethnic, racial, and cultural descriptions in published research were produced with the help of Drs Kwame McKenzie and Natasha Crowcroft (see also editorial). We hope that authors will use them in describing patients individually and in groups and that both authors and readers will find them helpful

The terminology of race, ethnicity, and culture is likely to be the source of continuing debate and will change because of fashion and political imperatives. So that researchers can compare their results now and in future, authors should try to use accurate descriptions rather than catch all terms in common use. Authors should describe in their methods section the logic behind their “ethnic” groupings. Terms used should be as descriptive as possible and reflect how the groups were demarcated. For example, “black” as a group description is less accurate than “self assigned as black Caribbean (Office of Population Censuses and Surveys category)” and “Asian” less accurate than “UK born individuals of Indian ancestry” or “French born individuals of Vietnamese ancestry.”

Categorisation for research or audit should relate to the type of hypothesis under investigation. For instance, genetic make up is the most powerful determinant of biological difference and the most accurate measure of risk; race has limited biological validity. Ethnicity and culture are socially determined variables of limited use in biological research, though they are useful in health services research. All the variables are confounded by socioeconomic status.

If it is unknown which of ethnicity, race, or culture is the most important influence then an attempt should be made to measure all of them.

A range of information is best collected:

  • Genetic differences (using relevant genetically determined polymorphism)

  • Self assigned ethnicity (using nationally agreed guidelines enabling comparability with census data)

  • Observer assigned ethnicity (using OPCS or other national census categorisation or the researchers' own logically argued categories)

  • Country or area of birth (the subject's own, or parents' and grandparents' if applicable)

  • Years in country of residence

  • Religion.

Culture is difficult to measure and describe, and there is scope for a large list of variables, for example, language and diet, which reflect our diversity and may be of importance in a particular study. Any cultural dividing line is arbitrary, but this list gives the minimum that should be expected of researchers.