As coronavirus spreads through China and threatens a pandemic, another contagion is moving even faster around the world: disinformation about the disease. Some of the stories surging through social media are relatively harmless retellings of unconfirmed studies, such as one showing that infection moved into humans from snakes; some are politically loaded conspiracy theories, for example the allegation that the virus escaped from a germ warfare lab in Wuhan; and some offer erroneous and potentially harmful advice, like the idea that you can prevent infection by spraying yourself with alcohol or chlorine solution.
The mathematics of contagion varies little from coronavirus to computer virus, financial risk to fake news. Now, with perfect timing, a good guide has arrived to pull together scientific knowledge about the way things spread and how to block (or encourage) their transmission.
Adam Kucharski, an epidemiologist at the London School of Hygiene and Tropical Medicine, completed The Rules of Contagion before the novel coronavirus disease Covid-19 emerged in Wuhan - and he sensibly resisted rushing into a last-minute update or revision to take account of the new epidemic, which is developing so fast that anything he wrote about it would have been out of date before publication.In any case his book prepares the ground comprehensively for readers to make sense of what is happening today, by distilling the wisdom gathered by studying previous epidemics over more than a century.
Although John Snow is widely regarded as the father of modern epidemiology for his landmark study in 1854, mapping cases of cholera in London to show that contaminated water rather than "bad air" spread the disease, Kucharski's hero is another Briton, Ronald Ross (1857-1932), who devoted 50 years to understanding how mosquitoes transmit malaria. Ross demonstrated mathematically that malaria could be eliminated by reducing mosquito populations below a critical level at which the disease would fade away, without having to wipe out every insect.
Many of Ross's peers were unconvinced but mosquito control did eventually reduce infection sufficiently to remove malaria from continental Europe by the 1970s. His dynamic view of the way a disease proceeds - depending on the numbers who are susceptible to infection, actually infected and immune (because they have recovered or been vaccinated) - has been at the heart of epidemiology ever since. Indeed, Ross extended his ideas into a prescient "Theory of Happenings" that covered not only disease but "even commerce, politics and statesmanship", as he put it, though the wider world would not take this extension seriously until long after his death.
Another important concept from the early 20th century is "herd immunity", developed by the British medical statistician Major Greenwood (Major was his first name; his military rank was Captain). Just as Ross showed that malaria could be controlled without killing every last mosquito, herd immunity makes it possible to wipe out infections without vaccinating everyone in the population.
A third key to understanding epidemics was formulated surprisingly recently. In the 1970s the German mathematician Klaus Dietz outlined a quantity known as the reproduction number R, representing the average number of people that an infectious individual will infect in a specific population.
If R is less than one, the number of cases declines over time. For many viruses in the early stages of an outbreak, including pandemic flu, Ebola and (it seems) Covid-19, R is around two. Coincidentally, Kucharski writes, Facebook researchers found R=2 for their fastest propagating online content such as the 2014 ice-bucket challenge. One of the most infectious germs known causes measles, with R as high as 20 in a fully susceptible group - which is why vaccination rates above 95 per cent are required to achieve sufficient herd immunity to stop measles.
If you are pushing a product or political message, you want R to be as large as possible. If you are fighting a virus, whether biological or electronic, you have to minimize R. In the case of Covid-19 a vaccine is unlikely to be available until next year. Until then health authorities must rely on other measures, such as quarantining those infected and tracing their contacts.
To reduce online exposure to medical misinformation, the tech giants are collaborating to push reliable sources to the top of internet sites. For instance, if you search for coronavirus on Google, help and information from the World Health Organization feature prominently; on Twitter in the UK, the latest government advice is the first thing you see.
Although epidemics share many common features that help to structure our efforts to control them, each one also has distinctive aspects from which we can learn for the future. Never before, for example, has information about a new disease spread around an interconnected world more rapidly than during the 2020 coronavirus epidemic.
"In outbreak analysis the most significant moments aren't the ones where we're right," Kucharski concludes. "It's those moments when we realize we've been wrong." For Covid-19 such moments are yet to come.