Inequality in mortality rates

Steven Wyatt, Yvonne Thomas and Tony Snell, Birmingham and the Black Country Strategic Health Authority

[Project number 20098]

Reducing health inequalities is a key government objective for the National Health Service. Little is known however about the nature and scale of health inequalities in Birmingham and the Black Country. To this end the Medical Director and Public Health Lead have commissioned an analysis into inequalities in mortality rates over time and across a number of dimensions of inequality - gender, region of birth, deprivation, geography and ethnicity.

Directly standardised mortality rates have been calculated for all causes and specific ICD10 classification between 1996 and 2002 for a range of population subgroups. These rates have been used to quantify the level of inequalities (both absolute and relative) between subgroups using a population weighted version of the Index of Disparity.

Given however, that ethnicity is not recorded as part of the death registration process; it has not been possible to directly calculate mortality rates for ethnic groups. A method of estimating these mortality rates has been developed using 'postcode of usual residence' from death registration files combined with census data about the geographical distribution of ethnic groups at output area level.

The analysis will be written up into a report and used locally to raise awareness, prompt action and support management decision making on tacking health inequalities.

Longitudinal Study data is requested in order to corroborate estimated mortality rates for ethnic groups using the methodology described above. Counts of deaths by ethnic groups estimated using this methodology would be compared to those observed in the LS in Birmingham and the Black Country between 2002 and 2004. If wide variations between the counts are observed then the methodology will be reviewed. If there is concordance or systematic variations between the counts then this information will be made available to readers of the report to assist in the interpretation of the analysis.