My Web-links

Please click on each of the web-links listed on the right hand side below, to explore your horizons.

Friday, January 31, 2020

Wuhan virus - email from Victor Gnanadurai

Made in China
Will the Wuhan virus become a pandemic?
Probably. But public health services can help determine how severe it turns out to be
Two things explain why a new infectious disease is so alarming. One is that, at first, it spreads
exponentially. As tens of cases become hundreds and hundreds become thousands, the
mathematics run away with you, conjuring speculation about a health-care collapse, social and
economic upheaval and a deadly pandemic. The other is profound uncertainty. Sparse data and
conflicting reports mean that scientists cannot rule out the worst case—and that lets bad
information thrive.
So it is with a new coronavirus, known as 2019-ncov, which has struck in China. The number of
reported cases grew from 282 on January 20th to almost 7,800 just nine days later. In that time
four reported cases outside mainland China have multiplied to 105 in 19 territories. Doubt
clouds fundamental properties of the disease, including how it is passed on and what share of
infected people die. Amid the uncertainty, a simulation of a coronavirus outbreak by Johns
Hopkins University in October, in which 65m people lost their lives, was put about as a
prediction. It is not.
Those are the right questions, though: will the new virus become a global disease? And how
deadly will it be? A definite answer is weeks or months away, but public-health authorities have
to plan today. The best guess is that the disease has taken hold in China and there is a high risk
that it spreads around the world—it may even become a recurrent seasonal infection. It may
turn out to be no more lethal than seasonal influenza, but that would still count as serious. In the
short term that would hit the world economy and, depending on how the outbreak is handled, it
could also have political effects in China.
The outbreak began in December. The repeated mingling of people and animals in China
means that viral mutations that infect humans are likely to arise there; and mass migration to
cities means that they are likely to spread between people. This virus probably originated in bats
and passed through mammals, such as palm civets or ferret badgers, ending up in Wuhan’s wet
market, where wild animals were on sale. Symptoms resemble flu, but can include pneumonia,
which may be fatal. About 20% of reported cases are severe, and need hospital care; about 2%
of them have been fatal. As yet, there is no vaccine or antiviral treatment.
The greatest uncertainty is how many cases have gone unrecorded. Primary health care is
rudimentary in China and some of the ill either avoided or were turned away from busy
hospitals. Many more may have such mild symptoms that they do not realise they have the
disease. Modelling by academics in Hong Kong suggests that, as of January 25th, tens of
thousands of people have already been infected and that the epidemic will peak in a few
months’ time. If so, the virus is more widespread than thought, and hence will be harder to
contain within China. But it will also prove less lethal, because the number of deaths should be
measured against a much larger base of infections. As with flu, a lot of people could die
nonetheless. In 2017-18 a bad flu season saw symptoms in 45m Americans, and 61,000
deaths.

Scientists have started work on vaccines and on treatments to make infections less severe.
These are six to 12 months away, so the world must fall back on public-health measures. In
China that has led to the biggest quarantine in history, as Wuhan and the rest of Hubei province
have been sealed off. The impact of such draconian measures has rippled throughout China.
The spring holiday has been extended, keeping schools and businesses closed. The economy
is running on the home-delivery of food and goods.
Many experts praise China’s efforts. Certainly, its scientists have coped better with the Wuhan
virus than they did with sars in 2003, rapidly detecting it, sequencing its genome, licensing
diagnostic kits and informing international bodies. China’s politicians come off less well. They
left alone the cramped markets filled with wild animals that spawned sars. With the new virus,
local officials in Wuhan first played down the science and then, when the disease had taken
hold, enacted the draconian quarantine fully eight hours after announcing it, allowing perhaps
1m potentially infectious people to leave the city first.
That may have undermined a measure which is taking a substantial toll. China’s growth in the
first quarter could fall to as little as 2%, from 6% before the outbreak. As China accounts for
almost a fifth of world output, there will probably be a noticeable dent on global growth. Though
the economy will bounce back when the virus fades, the reputation of the Communist Party and
even of Xi Jinping may be more lastingly affected (see article). The party claims that, armed with
science, it is more efficient at governing than democracies. The heavy-handed failure to contain
the virus suggests otherwise.
Outside China such quarantines are unthinkable. The medical and economic cost will depend
on governments slowing the disease’s spread. The way to do this is by isolating cases as soon
as they crop up and tracing and quarantining people that victims have been in contact with—
indeed, if the disease burns out in China, that might yet stop the pandemic altogether. If, by
contrast, that proves inadequate, they could shut schools, discourage travel and urge the
cancellation of public events. Buying time in this way has advantages even if it does not
completely stop the disease. Health-care systems would have a greater chance to prepare for
the onslaught, and to empty beds that are now full of people with seasonal flu.
Despite all those efforts the epidemic could still be severe. Some health systems, in Africa and
the slums of Asia’s vast cities, will not be able to isolate patients and trace contacts. Much
depends on whether people are infectious when their symptoms are mild (or before they show
any at all, as some reports suggest), because such people are hard to spot. And also on
whether the virus mutates to become more transmissible or lethal.
The world has never responded as rapidly to a disease as it has to 2019-ncov. Even so, the
virus may still do great harm. As humans encroach on new habitats, farm more animals, gather
in cities, travel and warm the planet, new diseases will become more common. One estimate
puts their cost at $60bn a year. sars, mers, Nipah, Zika, Mexican swine flu: the fever from
Wuhan is the latest of a bad bunch. It will not be the last. ■
This article appeared in the Leaders section of the print edition under the headline "Will the
Wuhan virus become a pandemic?"

No comments:

Post a Comment