The Public Health England report “Disparities in the risk and outcomes of COVID-19” and the subsequent “Beyond the data: Understanding the impact of COVID-19 on BAME groups” was based on the analysis published by the Office of National Statistics (ONS) “Deaths involving COVID-19, England and Wales: deaths occurring in April 2020″.
This resulted in the general conclusion that the Black, Asian, and Minority Ethnic (BAME) population were more likely to die from Covid-19 than the White British population.
The ONS report collected data from 2 sources which both showed that the BAME population represented ~ 17% of deaths ( 83% White) from Covid-19; this is compared to the total population split of 13% ( 2011 Census). Hence the conclusion about the disproportionality of deaths in the BAME population.
The NHS England data for the period to 12 May 2020 supported this same percentage for BAME deaths ( ~ 17%)
From the same NHS England data source from 12 May – 19 June, the split is markedly different.
This shows the increase in the last 5-6 weeks ( in effect the 2nd half of the pandemic, so far) for the BAME population is ~7% of the deaths – a significant reduction from the previous 17%
The interpretation of the original data informing the risk level of the BAME population was flawed. As quoted in the latest Public Health England (PHE) report “Beyond the data: Understanding the impact of COVID-19 on BAME groups”
“These analyses did not account for the effect of occupation, comorbidities or obesity. These are important factors because they are associated with the risk of acquiring COVID-19, the risk of dying, or both. Other evidence has shown that when comorbidities are included, the difference in risk of death between ethnic groups among hospitalised patients is greatly reduced.”
The analysis also doesn’t take account of any genetic factors e.g. the relative ability of people with different skin colours to generate Vitamin D. Also, :
“Mortality data has primarily relied on hospital reported deaths; not including care home deaths could artificially inflate BAME deaths as more white British older adults reside in nursing and residential homes”
Ethnicity is not routinely recorded, and is not on the death certificate; this is the data source for the ONS reporting. The ONS had to track back to the 2011 census to obtain ethnicity of each individual
This quote from the same PHE report is most pertinent:
“In their view ( 4000 people consulted from within BAME population), COVID-19 did not create health inequalities, but rather the pandemic exposed and exacerbated longstanding inequalities affecting BAME communities in the UK.”
The mass media, fuelled by limited tentative data from expedited reports, has created a level of anxiety in the BAME community which is unnecessarily exaggerated. There is no evidence that BAME people are, fundamentally, more at risk from Covid-19 than the wider population.
To the extent that the data focuses attention on underlying social issues, and inspires action, then it is a good outcome – but we should be wary of force-fitting data to suit a political agenda.