Comment by Andrew Brown
Atheists are less at risk than average from Covid-19 and Jews more at risk. But the effects on all other religious groups are moderate. In particular, the higher absolute death rate among Muslims in Britain turns out not to have anything to do with their religion but is a function of poverty and ethnicity.
These are the headline conclusions from the Office of National Statistics study of the role that faith has played in the death toll of the pandemic.
The absolute figures suggest a much higher death toll for several faith groups. Muslim and Jewish men in particular have died at more than twice the rate of Christian men of the same age, and Muslim and Jewish women at nearly twice the rate of their Christian counterparts. Since the report relies on the religious affiliations reported in the 2011 census, Christians are overwhelmingly the largest group and are used as a baseline in most comparisons.
But when these absolute death rates are controlled for socio-economic status and a range of other demographic factors such as educational attainment, much of the disparity vanishes.
Muslim men are still more likely to die than their Christian counterparts, and slightly more likely than either Hindus or Sikhs, but the disparity for women is now within the margin of error.
Add in a further control for ethnicity and the disparity vanishes altogether.
What remains, though, is disturbing. This analysis suggests that skin colour is itself a health risk factor in Britain today. Low social status is known to be bad for health and it is possible that ethnicity both signals and aggravates low social status. Other possible factors are cultural: some diseases such as diabetes are much more common in South Asians in Britain.
Even when all the possible confounding factors are stripped out, the Jewish population remains an outlier. Jewish men are still nearly twice as likely as Christians to die of the disease. One possible explanation of this would be that the deaths are concentrated among the ultra-orthodox.
Because the survey works with death figures that start on 2 March, before the lockdown, it is possible the that practice of frequent prayer in small prayer rooms or synagogues helped to spread the disease among that population. But this is speculation, since the figures are not broken down by style of worship or other cultural or theological factors.