This website places cookies on your device to help us improve our service to you. To find out more, see our Privacy and Cookies statement.

Ethnic minorities: introduction

Ethnicity refers to a common group identity based on language, culture, religion or other social characteristics. This means that people define their own ethnicity, that everyone (and not just those in minorities) has an ethnicity and that a person's ethnic identity may change over time. More information on defining ethnicity and race is available.

In the 2011 census, non-white minority ethnic groups in England and Wales were estimated to account for 14.1% of the population (source: Census 2011 release 2.1 but on the latest available Scottish figure only 4% (212,000) of the Scottish population (source: Scottish Census 2011 - release 2A). In Scotland the largest minority ethnic groups are of Pakistani and of white Irish origin, but there may be local concentrations of specific groups such as asylum seekers or migrant workers. More details of the ethnic composition of the Scottish population and Analysis of Equality results from the 2011 Census are available from the 2011 Scottish census.

Country of birth has been used as a guide to ethnic group, though it gives only an approximate indication. For example, Fischbacher et al (2007) (see Key references) analysed mortality by country of birth for people living in Scotland between 1997 and 2003. They found that men and women born in Pakistan had similar overall mortality to the Scottish-born population.

Data on live births in Scotland by country of birth of the mother can be produced by National Records Scotland. This may give more recent information than that based on the census, but may not reflect the ethnicity of minority ethnic persons whose mothers were born in the UK.

Analyses of disease risk among ethnic minorities need to distinguish between relative and absolute differences. In the figures from the paper by Fischbacher above, Scottish figures are compared separately to people born in England and in Scotland. People born in Pakistan and living in Scotland did not have a lower risk of all-cause mortality in comparison to the English population, but in comparison to the Scottish population their risk was clearly lower (Table 1). This reflects the significantly higher risk of all-cause mortality in the Scottish born population compared to English born. For CHD and Stroke the mortality risk for people born in Pakistan and living in Scotland was higher than the risk for the English, but about the same as the risk for the Scottish-born population.

Table 1: Deaths and Standardised Mortality Ratio (SMR), including 95% confidence interval, for Pakistan-born residents of Scotland (25 years and older), Jan 1997-Mar 2003 using Scottish-born residents of Scotland as reference.

  Men Women
  Deaths SMR Deaths SMR
Coronary heart disease 64  101 (77,128) 26 102 (66,149)
Stroke 19 104 (62,162) 13 83 (44,142)
All cause 171 63 (54,73)* 110 71 (58,85)*

* two tailed significance testing, p

Source: Fischbacher, C. M., Steiner, M., Bhopal, R., Chalmers, J., Jamieson, J., Knowles, D., & Povey, C. 2007, "Variations in all cause and cardiovascular mortality by country of birth in Scotland, 1997-2003.[erratum appears in Scott Med J. 2008 May;53(2):66]", Scottish Medical Journal, vol. 52, no. 4, pp. 5-10.

These figures emphasise that minority ethnic groups should not be assumed to have worse health across the board. They also emphasise that, in general, the health needs of ethnic minorities should be assessed using absolute measures, such as numbers of deaths, rather than relative ones, such as Standardised Mortality Ratio (SMR).

Published data (Using routinely collected data to compare hospital admission rates by ethnic group in Scotland (2016) revealed similar findings for hospital admission rates for CHD in the first national analysis of routinely-collected ethnicity data in hospital admissions records. Rate ratios for admissions for each ethnic group relative to the White Scottish group were lower among White Polish and Chinese groups, with higher rates in some Asian groups, White British and White Other. These results demonstrated that the completeness and quality of ethnicity coding in routinely-collected health data was of sufficient quality to allow publication and wider use. 


Page last updated: 03 November 2017

© Scottish Public Health Observatory 2014