A study of ethnic variations in mortality by socio-economic measures

Yuan Huang Lee, Office for National Statistics

[Project number 20063]

Variations in health among ethnic groups and the general population have been widely reported. Explanations for the differences in health and mortality have been linked to biological, cultural, discriminatory and socio-economic factors. Research by Marmot et al in 1984 reached the conclusion that class and socio-economic factors do not have an effect on the health variations among ethnic groups. However, later studies by Nazroo (1997), Evandrou (2000), have shown that socio-economic factors are of importance when explaining ethnic variations in health. Data used for the studies in the past have included the General Household Survey, Census and the Fourth National Survey on Ethnic Minorities. By using the LS to look at mortality data for ethnic groups, it is hoped to describe ethnic variations in mortality and to further develop analysis of the factors which lead to variations in health of ethnic groups.

The project is expected to be small scale and completed in 6 months.

The aims of the study are:

  1. To see if the Longitudinal Study is able to provide a reliable set of standardised mortality ratios for ethnic groups; and
  2. To see whether if any variation exists and if this variation can be explained by other factors when controlling for socio-economic and social status measures.

Standardised Mortality Ratios will be calculated for periods in the 70s, 80s and 90s.

Initial analysis will determine whether we can derive ethnic group for data available at 81, as there currently is no information on ethnic group for this period. Then, by using the wide range of variables available from the LS to explore any confounding factors on variations in mortality, these include length of time in country and other measures of social status, for instance, housing tenure, car ownership and education.

With data on mortality from the different decades, trends in mortality can be analysed and used for comparison against existing data sources. Although the LS is a useful source of information, it is expected that small numbers within the ethnic groups may be problematic.

Top of page