Ethnicity, mortality and self-reported LLTI and general health at the 2001 Census

Dermot O'Reilly and Michael Rosato, Queen's University, Belfast

(Project no. 30080)

This research aims to use the ONS-LS to examine variations in the levels of limiting long-term illness (LLTI) and general health (GH) across the different ethnic groups in England and Wales at the time of the 2001 census, adjusting for a range of other demographic and social factors, and to examine the relationship between LLTI and GH and subsequent mortality to see if this differs systematically across these ethnic groups.

While in the past the assessment of population health focused on mortality data as primary indicator, it is now felt that morbidity data can be more relevant to present day public health concerns. To fill this information gap in the United Kingdom a question on LLTI was included in the 1991 UK Census which proved central in both constructing resource allocation formulae for health service funding, and also as a tool for research, for example, in adjusting for 'need' in studies of social inequality in access to medical care. A second question, on GH in the preceding year, was included in the 2001 Census, with the expectation that this would further improve the understanding of needs for health services and resource allocation.

However, although these self-reported measures are accepted as independent predictors of mortality and health care use, questions remain about their subjectivity since the recorded levels depend not only on the 'objective' health status of an individual, but also on their perception of this and their propensity to report what health problems they have. There is now a growing concern that ethnicity is one such modifying factor leading to systematic differences in the validity of these measures amongst the different ethnic groups.

Self reported health (SRH) measures are very important as parameters both in public health analysis and in government initiatives to identify areas of deprivation throughout the UK - allowing development of strategies to reduce the burden of this on both individuals and civil society in general. For government the census is the most important source of these data, and over the past fifteen years they have been used with great success. However, in studies using SRH measures in conjunction with more detailed instruments of health measurement the picture is usually more inconsistent: self reported health measures are regularly interpreted as not being consistent across categories of various parameters such as SES, age, or even gender - calling into question their efficacy/validity and reliability as separate from these mediating factors. The proposed study will test the consistency of SRH measures in relation to the ethnicity data available at the 2001 Census.