The EconomistJune 27th 2020 Science & technology 652
1that might threaten human health. The
second is the monitoring of blood samples
and other indicators from people living in
places where new diseases are most likely
to emerge. The third is a concerted pro-
gramme that employs all the data thus col-
lected to get a head-start in the develop-
ment of drugs and vaccines that might be
used to meet an emerging disease halfway.
More than 1,400 pathogens are known
to infect human beings. Only a fifth of
these are viruses. But viruses are the cause
of more than two-thirds of new human dis-
eases, which is why both the discussion
and the effort are focused on them.
Being simple organisms that can repro-
duce quickly, viruses undergo much more
rapid evolution than other types of patho-
gens. Hence their ability to adapt to novel
hosts. Simian immunodeficiency virus
(siv), for example, moved from monkeys to
chimpanzees and, separately, to gorillas
before one of the chimpanzee strains got
into people and became hiv-1, which went
on to cause the aidspandemic (see Books
& arts). Influenza viruses that break out
into human beings are routinely found to
have cycled through pigs or chickens first.
And sars-cov, as the virus which caused
the sarsoutbreak in 2002 is now known,
started in bats before (it is widely believed)
migrating to civets.Going viral
Ecologically, human beings as a species are
particularly likely to be on the receiving
end of this process. Few wild animals
spend as much time cooped up with mem-
bers of their own and other species as do
herd animals and their herders. So when
people domesticated animals and began to
live in large, fixed settlements they gave vi-
ruses many opportunities to jump back
and forth between species, a process called
viral chatter. Many common diseases date
back to the early days of domestication,
and the population densities it brought
with it. The species involved were not all
domesticated. Smallpox appears to have
come from rats. But the animals involved
were ones that thrived in human company.
Zoonoses have continued ever since. Of
more than 330 diseases which emerged be-
tween 1940 and 2004, over 60% were zoo-
notic. Of those over 70% originated in
wildlife. For viruses the proportions were
69% and 87% respectively (see chart on
next page).
This process seems to be accelerating.
As human populations grow, previously
wild areas are settled. That brings people
into contact with sources of infection they
would not otherwise have encountered.
And, having encountered them, it is also
easier to pass them on. Modern transport
means that if a disease gets into people liv-
ing in this frontier between civilisation
and the wilderness it can quickly make itsway to a local metropolis and thence, cour-
tesy of lorries, trains and planes, to others
anywhere in the world.
The idea that these risks deserve sys-
tematic appraisal and monitoring surfaced
after the emergence, in 2005, of the h5n1
strain of avian influenza. This was first de-
tected in 1996, when it killed some geese in
Guangdong province, China. The following
year it infected 18 people associated with a
poultry market in neighbouring Hong
Kong, six of whom died. But for most of the
subsequent decade the virus was restricted
to farmed birds on the Chinese mainland.
In 2004, however, a highly pathogenic
strain emerged and began to spread across
South-East Asia, killing tens of millions of
birds. By the middle of 2005 this version of
the virus had infected wild geese, which
took it into Europe, India and Africa. That
year, 98 people were infected, and 43 of
them died—a death rate severe enough for
David Nabarro, then co-ordinator of the
un’s response to influenza, to issue a warn-
ing that an unchecked h5n1outbreak could
kill up to 150m people. In 1968 a less patho-
genic strain of flu, which had originated in
the same area, killed 1m people when it
spread around the world. In 1957 a still-ear-
lier relative killed 1.1m. h5n1was consider-
ably more lethal than either.
In the end, forms of h5n1that could
spread easily from person to person never
arose. But they came close. In 2012 Yoshi-
hiro Kawaoka of the University of Wiscon-
sin-Madison and Ron Fouchier of Erasmus
University in Rotterdam undertook to dis-
cover how many mutations would have
been required to make h5n1transmissible
between people via the droplets expelled in
a sneeze. They found that changes at just
five points in the genome would have done
the trick. Two of these mutations were later
shown to exist already in wild populations
of the virus.
Fortunately, the other three never hap-
pened. The disease was brought under con-trol among farmed birds, though it still cir-
culates at a low level in wild populations,
and human fatalities remained in the tens,
rather than the tens of millions. This near
miss may have spurred complacency
among laymen. Scientists’ dire warnings
had come to naught. Among virologists
and epidemiologists, though, it was a call
to action—and one which came at an op-
portune time. The cost of sequencing the
dnaand rnain which viruses store their
genes was, in the second half of the 2000s,
falling at extraordinary speed (see Graphic
detail). That made virus hunting possible
on a previously unimaginable scale.Predict and survive
In 2009 Dennis Carroll, an infectious-dis-
ease expert at usaid, America’s interna-
tional development agency, who had led
that agency’s response to the h5n1out-
break, set up predict. This project investi-
gated and catalogued potential disease
threats to people living near wildlife, with a
particular focus on viruses. A few years lat-
er Jeremy Farrar, an infectious-disease doc-
tor who was then at Oxford University’s
clinical research unit in Ho Chi Minh City
and now heads the Wellcome Trust, a large
medical-research charity, created Vizions.
This project tracked pathogens circulating
in people and animals living close together
in farms and markets across Vietnam.
predictran for just over a decade. Sci-
entists working with local teams in 30
countries collected around 170,000 sam-
ples from people and wild animals, mainly
non-human primates, bats and rodents. In
the process they discovered 1,200 new vi-
ruses belonging to families known to have
the potential to infect people and cause
epidemics. Among these were more than
160 potentially zoonotic coronaviruses.
This, though, just scratched the surface.
On the basis of how much they found dur-
ing their early work Dr Carroll and his col-
leagues made a statistical estimate that, allDisease Y
Predicted emerging-disease hotspotsSource:EcoHealthAllianceRisk of disease emerging Lower Higher