Does selective migration cause the increasing socio-economic gradient in ill-health? An analysis using the ONS Longitudinal Study for England and Wales
Dermot O'Reilly and Sheelah Connolly, Queen's University Belfast
[Project number 30036]
The current method of assessing whether inequalities in health within populations may be changing is to compare the health status of areas, or groups of areas, over time. However, this methodology ignores the potential impact of migration and especially selective migration. Selective migration may be the cause of increasing inequalities between areas even if the health of the individuals within these areas have not changed. In order for this to occur there would have to be a net movement of healthier people from deprived areas or a net movement of less healthy people to deprived areas. Furthermore there must be no significant gradient constraint; that is people moving from more to less deprived areas must have a level of health approximately equal to or better than the group they are joining. The evidence relating to selective migration is mixed, though one important factor appears to be the level of analysis, with studies looking at a local geographical level more likely to find evidence of selective migration than those concerned with a regional level.
The proposed study aims to address the question of whether selective
migration is responsible for the increasing socio-economic gradients in
ill-health in England and Wales between 1991 and 2001. The Carstairs index
for each electoral ward for 1991 will be used to group LS members into
deciles of deprivation ranging from most affluent to deprived. These categories
will be used to compare changes in mortality and morbidity in the same
groups of areas between 1991 and 2001. If inequalities are found to be
increasing over time, we propose to determine how much is related to the
selective movement of healthy or unhealthy individuals. This will be done
by re-calculating morbidity and mortality rates for each decile for 2001
when individuals are put back to their place of residence in 1991. Such
an analysis allows the migration effect to be isolated. Finally it is
intended to compare the health experience of migrants to that of non-migrants.
The longitudinal study (LS), based on a continuing one per cent sample of the population of England and Wales is a high quality multi-cohort research database with ongoing linkage of routinely collected vital statistics data with the census returns for 1971, 1981, 1991 and 2001. Its size and representativeness greatly enhances robustness and confidence in analyses and findings.
In this study, we aim to investigate the relation between migration and health in an attempt to discover if this relationship could have an impact on the monitoring of health inequalities by area over time. Deprivation scores at electoral ward level will be used to divide England and Wales into deciles ranging from most affluent to most deprived. Information on migration and health will be extracted from the 1991 and 2001 Census. Limiting long-term illness and standardised mortality ratios (in conjunction with mortality data) will be used as an indicator of health, while residence in a different decile in 2001 than 1991 will define a migrant. The health experience of migrants will be compared to that of non-migrants. The impact of selective migration will be assessed by replacing all migrants to their deprivation decile of residence of residence in 1991.