Exploring the ‘Scottish Excess’ with eDatashield and linkage of an adjusted Index of Multiple Deprivation
Ball, W., Kyle, R. & Atherton, I. (2017) UK Administrative Data Research Network Annual Research Conference, Royal College of Surgeons, Edinburgh, UK, 1 - 2 June 2017 [SLS][ONS LS]
Background Scotland has worse health outcomes compared to England and Wales. Not all of this variation can be explained by deprivation and it may explain less now than in the past. However, much of this analysis has relied on outdated measures of deprivation. Directly comparing data from the constituent countries of the United Kingdom is also difficult due to differences in the way Indices of Multiple Deprivation (IMD) are constructed in each country. This study addresses the methodological limitation of current research by linking routinely collected data from England/Wales and Scotland with an adjusted UK Index of Multiple Deprivation (IMD) to conduct joint analysis with an up to date and directly comparable measure of deprivation.
Methods An adjusted UK IMD has been developed using two comparably measured domains from the English and Scottish IMDs. This will be linked to individual level census data from both the ONS Longitudinal Study (ONS LS) and the Scottish Longitudinal Study (SLS) to provide a consistent measure of area deprivation. Confidentiality and disclosure control prevent combined analysis of these longitudinal datasets. However, the eDatashield process enables pooled analysis by passing summary statistics between studies to ensure that the datasets are comparable and analysis is equivalent to working with individual level data.
Descriptive and correlational analyses will explore the presence and extent of health inequalities within and between Scottish, English and Welsh populations. If a 'Scottish excess' is discerned, logistic regression will be used to establish how much of this variation can be accounted for by deprivation, adjusting for individual and household socio-economic indicators.
Discussion Pooled analysis of individual level census data from England, Wales and Scotland linked to an adjusted UK IMD provides a novel approach to exploring inequalities across the UK. This paper sets out the study and shares preliminary findings.