Investigating the contribution of socio-economic position to ethnic inequalities in severe COVID-19 outcomes: population-based mediation analyses of national linked Scottish data

Abstract

We quantified the extent to which socio-economic position (SEP) contributed to ethnic inequalities in severe COVID-19 outcomes (hospitalization or death) in Scotland. We used linked 2011 Scottish Census and health records to assess whether ethnic inequalities were mediated by different SEP measures: area deprivation, educational status, household composition, and multigenerational household. We considered disaggregated ethnicities ‘White Scottish’, ‘White British or Irish’, ‘Other White’, ‘South Asian’, ‘African, Caribbean, or Black’, and ‘Other’. We applied marginal structural models to estimate causal pathways. Of the 3 297 205 individuals analysed, 38 213 (1.2%) had severe COVID-19 outcomes. South Asians had elevated risk of severe COVID-19 compared to White Scottish (hazard ratio: 1.7; 95% confidence interval: 1.5–1.9), while White British or Irish (hazard ratio: 0.7; confidence interval: 0.6–08) and other White (hazard ratio: 0.8; confidence interval: 0.7–0.9) had reduced risk. When holding area deprivation constant, the risk of severe COVID-19 declined by 16.5% for South Asians and 49.2% for White British or Irish; but increased for other White (75.4%). When holding education constant, the risk of severe COVID-19 reduced by 24.8% for White British or Irish and 20.6% for other White; but increased by 74.6% for South Asians. Only a slight change in risk was observed for the South Asians after holding household size and multigenerational household constant. Risk estimates for African, Caribbean or Black, and other groups were underpowered. SEP measures differed substantially in the extent to which they mediated ethnic inequalities in severe COVID-19. This highlights the necessity of addressing multiple dimensions of SEP that drive ethnic inequalities.

Rights

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Cite as

Downloadable citations

Identifiers

DOI

https://doi.org/10.1093/eurpub/ckaf078

Repository URI

https://strathprints.strath.ac.uk/92991/