54 International The EconomistNovember 21st 2020
21Chicago. Many houses, shops and factories
are boarded up or abandoned. Fear of gun
violence keeps children indoors. Africa
Food and Liquor and Quick Food Mart offer
few fresh vegetables—mostly cabbage—
but shelves stacked with sweets, fizzy
drinks and booze. People assume most
black deaths in Chicago are the result of
gun violence, but the primary cause of ear-
ly death in neighbourhoods like these is
heart disease, says David Ansell, a doctor at
Rush University Medical Centre. Across
America black men under 50 are twice as
likely as white men to die of heart disease.
In Brazil skin colour is a good proxy for
social factors too, says Fatima Marinho, an
epidemiologist in São Paulo. Sandra Maria
da Silva Costa lives in a favela in Rio. She is
46 “but I look 56.” Even before catching co-
vid-19 in April, she suffered from a litany of
health problems, including in her lungs.
Her lungs worsened in September when,
after stealing some meat and milk, she
spent a month in prison, where she re-
ceived no health care. Both her parents
died last year. They never spoke about rac-
ism or exclusion, says Ms da Silva Costa;
they simply accepted that they would not
get proper health care. “We’re black, poor
and jobless,” she says. “We’re invisible.”
Wealth and education matter even in
countries where people are treated more
fairly. People who live in the areas of Eng-
land and Wales that count among the most-
deprived 10% are twice as likely to die of co-
vid-19 as those in the least-deprived areas.
All ethnic minorities except Indians and
Chinese are more likely to live in such
places than whites. Pakistanis are more
than three times as likely as white Britons
to do so and Bangladeshis twice as likely.
Two things help explain the dispropor-
tionate impact of covid-19 on ethnic mi-
norities. First, and most important, they
are more likely to be exposed to the virus.
In many Western countries minorities are
more likely to work in jobs that put them
into regular and close contact with the
public, increasing their risk of infection.
They are also more likely to live in cities, in
deprived areas and in crowded, multi-
generational homes, all of which increase
their exposure. Second, when they catch
the virus they are more likely to die of it
than white people. That is probably be-
cause pre-existing conditions, such as dia-
betes and heart disease, which increase the
risk of dying of covid-19, are more common
among ethnic minorities.A virus that discriminates
Such factors go a long way to explaining the
disproportionate impact of covid-19 on
non-white people in Britain, according to
the country’s Office for National Statistics.
But not entirely. Bangladeshi men are
three-and-a-half times more at risk of dy-
ing of covid-19 than white men of the sameage. After controlling for geography (this
group is twice as likely to live in densely
populated areas), this ratio fell to 2.3. After
controlling for factors such as poverty and
exposure at work, it fell to 1.9. But even after
including self-reported health concerns
and pre-existing conditions, their risk was
still almost one-and-a-half times that of
white men of the same age (see chart).
Even Sir Michael concedes that it is in-
creasingly clear that socioeconomic condi-
tions do not fully explain racial disparities
in health. In a recent report he and col-
leagues found that in several countries in
the Americas, such as Colombia and Brazil,
the worse health of black people cannot
fully be explained by conventional socio-
economic measures. The differences are
greater for men than women in America.
For black American women the life expec-
tancy gap narrows significantly when con-
trolling for education and income. “But for
men a sizeable unexplained gap remains,”
says Sir Michael. Some of that disparity can
be explained by high homicide rates
among black American men. They are also
more likely to die of aids(though this af-
fects relatively few men, it kills them when
they are young and so has a significant im-
pact on average life expectancy). But that
does not fully explain the gap.Puzzling patterns
Cancer is a good example of a disease the
prevalence of which cannot be explained
by socioeconomic factors alone. In Britain
black people have much higher rates of
stomach and prostate cancer than other
groups. Asian women are more likely than
any group to contract mouth cancer. South
Asian women are the least likely to get cer-
vical cancer. White Britons have the high-
est rates of cancer overall. Understanding
why certain groups are more likely to getdifferent cancers hints at the complex in-
teraction of social factors and biology that
may be at work.
People’s risk of dying of particular dis-
eases tends to reflect underlying condi-
tions that make them more vulnerable,
their access to a doctor or the treatment
they will receive. Black women in America
are no more likely than white women to get
breast cancer but much more likely to die
from it. Ethnic minorities made up 11% of
covid-19 hospital admissions in Britain in
May but 36% of those receiving intensive
care. Hospitalised South Asians were the
group least likely to survive, whereas there
was no difference between black and white
people, according to one study.
Disparities exist in other areas, too. Few
things predict more accurately whether a
woman will survive childbirth than the
colour of her skin. In America black wom-
en are three times, and native Americans
two-and-a-half times, as likely to die from
pregnancy-related causes as white women.
Even after controlling for education, dif-
ferences persist. Covid-19 could compound
this. In Britain’s first wave 55% of pregnant
women hospitalised with the virus were
from black and other ethnic-minority
groups (they represent 14% of the popula-
tion). In Brazil black pregnant women hos-
pitalised with covid-19 have been around
twice as likely to die as white ones.
What might explain such gaps? First,
pre-existing conditions. In the rich world
the leading cause of death related to child-
birth is heart disease, responsible for over a
third of deaths. Prevalence is higher among
black women. Second, access. In America
89% of white women receive prenatal care
in their first trimester, compared with 75%
of black women. This means missed op-
portunities for early diagnosis of problems
in pregnancy. Third, unequal treatment. InCounting the cost
England, death rate involving covid-19, by ethnic group and sex
Relativetothewhitepopulation,March2nd- July28th2020,logscaleSource:OfficeforNationalStatisticsAdjustedforage... andgeography... andsocioeconomics... andhealthstatusChineseMixed/multiple
ethnic groupsIndianPakistaniBlack CaribbeanBangladeshiBlackAfrican4x
higherTwice
as highSame
rateMenChineseMixed/multiple
ethnic groupsIndianPakistaniBlack CaribbeanBangladeshiBlackAfrican4x
higherTwice
as highSame
rateWomen