Explaining ethnic differentials in COVID-19 mortality: cohort study.

Ethnic inequalities in coronavirus disease 2019 (COVID-19) hospitalizations and mortality have been widely reported but there is scant understanding of how they are embodied. The UK Biobank prospective cohort study comprises around half a million people who were aged 40-69 years at study induction between 2006 and 2010 when information on ethnic background and potential explanatory factors was captured. Study members were prospectively linked to a national mortality registry. In an analytical sample of 448,664 individuals (248,820 women), 705 deaths were ascribed to COVID-19 between 5th March, 2020 and 24th January, 2021. In age- and sex-adjusted analyses, relative to White participants, Black study members experienced around five times the risk of COVID-19 mortality (odds ratio; 95% confidence interval: 4.81; 3.28, 7.05), while there was a doubling in the South Asian group (2.05; 1.30, 3.25). Controlling for baseline comorbidities, social factors (including socioeconomic circumstances), and lifestyle indices attenuated this risk differential by 34% in Black study members (2.84; 1.91, 4.23) and 37% in South Asian individuals (1.57; 0.97, 2.55). The residual risk of COVID-19 deaths in ethnic minority groups may be ascribed to a range of unmeasured characteristics and requires further exploration.