10 | New Scientist | 22 August 2020
THE idea of herd immunity
has had a bumpy ride as the
coronavirus pandemic has played
out. It was initially touted in some
countries as a viable strategy
for dealing with the spread of
covid-19, before being dismissed.
Today, some headlines celebrate
the fact that many places might
have achieved herd immunity
including Britain and pockets of
London, New York and Mumbai.
But others warn that millions will
die before we get there.
The true picture is far messier,
partly because scientists don’t
even agree on what herd
immunity is, let alone how it
might be achieved. So how will
we know when populations are
protected against the coronavirus?
While the definition of herd
immunity depends on who you
ask, let’s assume that it refers to
a situation in which enough of
a population is immune to a
pathogen that it no longer spreads
throughout a community. Those
who might be susceptible to it are
indirectly protected thanks to the
immune responses of others.
These immune responses might
have developed after a person was
infected with a pathogen or after
being vaccinated against it.
Our experience of other viruses
show how herd immunity can
develop. Seasonal viruses like
the common cold often sweep
through a population until
enough people have encountered
them and built up a protective
immune response for them to
stop spreading. Widespread
use of the MMR vaccine led to
herd immunity for measles in
some countries.
But herd immunity doesn’t
necessarily last. Viruses can evolveand change. And if vaccination
rates drop, viruses can make
a comeback.
When it comes to the
coronavirus, there are even
more challenges. For a start,
we don’t know what proportion
of a population would need to
be immune to generate herd
immunity. This figure is typicallyestimated using the basic
reproduction number – or
R number – of a virus, which
represents how many other
people a person who has the
virus will go on to infect.
Most estimates say that
between 60 and 70 per cent of
a population would need to be
immune to the coronavirus to
stop its spread in a community.
That is based on an R number of
somewhere between 2.5 and 3.But even if these levels were
reached, it doesn’t mean that the
virus won’t spread at all. It could
still be passed from an infected
person to someone who is
vulnerable, given the right setting.
“It’s a population-level statistic,”
says Julian Tang at the University
of Leicester, UK. It is difficult to
know when herd immunity has
been reached, but a decline in
cases would be a good indicator.
The problem is that humans
don’t tend to behave as a herd. The
calculation for the herd immunity
threshold assumes that people are
equally spaced and mixing equally
with each other. In other words,
“no walls, barriers, cars or glass
windows in the way”, says Tang.
In reality, of course, people live
in varied environments, and
travel and interact with others
to varying degrees.
Estimates of the R number
vary, too, and change over time.
Our behaviour can also influence
the figure. Lockdowns, social
distancing and other infection
control measures will havelowered the R number in many
regions, for example.
This line of thought led Gabriela
Gomes at the University of
Strathclyde, UK, and her colleagues
to recalculate the herd immunity
threshold. The team developed
a mathematical model based
on the simple herd immunity
calculation, but tried to factor
in variation between individuals.Already immune
In theory, this variation takes into
account the differences across a
population in susceptibility to the
virus, and the chances of people
coming into contact with it. The
more variation there is, the lower
the threshold for herd immunity,
says Gomes. By her team’s
calculations, only around 10 to
20 per cent of a population needs
to be immune to the virus to
achieve herd immunity (medRxiv,
doi.org/10/d6sh).
By that estimate, some places
have already achieved herd
immunity. More than 10 per cent
of the population of Madrid is
thought to have developed some
immunity to the virus, for
example. “In Europe, I think most
countries are close to having that
status,” says Gomes. This would
suggest that the epidemic is
currently at its peak in places like
Madrid, she says.
Samir Bhatt at Imperial College
London disagrees. “I personally
don’t believe we’re anywhere near
herd immunity,” he says. He points
to regions with exceptionally large
outbreaks, such as Bergamo in
Italy, where 57 per cent of people
tested between April and June had
antibodies to the virus. “If 10 per
cent were sufficient, why did they
get up that high?” Bhatt asks.
One issue is that estimates
for the threshold vary so wildly,
JUAanywhere from 10 to 70 per cent.NJOM
AR
TIN/EPA-EFE/SHUT
TE
RSTO
CKViolinist Ara Malikian
performs in Madrid
during the pandemic10-20%
Infection rate needed to
reach herd immunity,
according to one studyNews Coronavirus
TransmissionJessica HamzelouCan herd immunity ever happen?
Claims that some countries may have achieved the threshold oversimplify the issue