No Business Like Snow Business
In March, the London School of Hygiene and Tropical Medicine hosted a birthday party with a difference. The Birthday Boy couldn’t make it as he was 200 years old, but he’s the closest thing epidemiologists have to a folk hero. His name was John Snow, and he was the first person to figure out that mapping disease outbreaks might give us useful clues about where they start, and what we can do to stop them.
Born in one of the poorest areas of York, Snow was apprenticed to a surgeon in Newcastle at 14. By the time he first heard about cholera in the UK, he was working just a few miles from the first victim - a Sunderland man called William Sproat. The Victorians already knew about cholera in a milder form, but most people thought it was a problem of the climate in British colonies. The Asiatic cholera which arrived in 1831 was different, swift and horrific. Many victims died in convulsions, lying in a pool of their own watery poo, within a day of infection.
But it wasn’t until Snow set up a practice in London’s Frith Street that cholera played a starring role in his life. The big theory of the time was that disease was caused by miasma - foul smelling or bad air. But Snow was an anaesthetist, he knew all about gas and he thought it was a dodgy view. If cholera was something people were breathing in, why did it affect their bowels and not their lungs? In the middle of the second epidemic of 1848, Snow published a pamphlet stating that cholera was an infection of the bowels passed through drinking water contaminated with the faeces of cholera victims - a kind of miasma but spread through water, not air (it was long after Snow’s death that Robert Koch identified the cholera bacterium as the source of the disease, in 1883).
There were thousands of pamphlets on cholera published that year, so Snow’s didn't make much of an impression. The London Medical Gazette, one of it’s few reviewers, said it lacked persuasion and needed experimental proof.
Snow’s chance came in 1854. On August 31st, several people in his neighbourhood were struck down with cholera in a single night. In the next 4 days, 127 people died of the disease in the surrounding streets, and Snow saw the opportunity for his Grand Experiment. So what did he do?
He made a map. He obtained a list of cholera deaths and addresses, and carried out door to door enquiries on September 6th. He found that 83% of victims took their water from the same source - a public pump on nearby Broad Street. Snow took his information to the local parish board meeting, the next day the pump handle was removed and the rest is…. well, a mixture of myth and history. Chances are the epidemic was already subsiding by the time the pump handle was removed, but it might have stopped the epidemic a bit faster than otherwise.
Infectious disease in the wealthiest countries is no longer a big threat. Public sanitation, antibiotics and vaccines mean that if anything we’re not scared enough - misguided media panic lead the people of Swansea to fear vaccination more than disease, and now they’re suffering a measles epidemic for the first time in a generation. But on the whole, “herd immunity” has added much to our quality of life.
But that doesn’t mean disease mapping is obsolete. In 2000 the World Health Organisation said that in developing nations, infectious diseases were still one third of the disease burden. And even in wealthier countries, new threats emerge all the time.
On November 16th 2002 a Guangdong man died of a nasty flu-like disease shortly after flying to Hong Kong, and many followed along the air routes of East Asia and beyond. International air travellers panicked. The disease was SARS, and within 4 months the WHO had issued a global alert.
There’s a special protocol when a disease like SARS shows up. The WHO establish the syndrome - the collection of symptoms we’re looking at to be sure reports are of the same disease and not different things. Then we can confirm an epidemic is actually happening, work out its extent, define who’s at risk and work out a strategy that’ll hopefully prevent it spreading.
Both local and international teams worked hard with governments to control infection and investigate the cause of SARS. Healthcare workers struggling to pinpoint it and prevent the spread died in their efforts.
Geo-informatics played an important role. The science of gathering geographic information about how disease spreads is still really important 200 years on from Snow. Maps help us see changes and trends in what can be huge sets of data, and they help us to make better informed decisions. The WHO has its own epidemiological mapping system that models environmental and climatic conditions which might trigger outbreaks, too. All very clever stuff.
The WHO published a daily world map of reported SARS cases almost from the start. But one of the most impressive mapping projects came not from them, but the Hong Kong Yellow Pages. Their internet maps came with strong zooming abilities, which could take the user to street, and even building level. A location based service, users could not only find their five closest infected buildings, but find details of the nearest district hospitals should the worst happen. In a closely populated hub city like Hong Kong, it was invaluable. A mobile phone company even used its customer’s location data to send SMS messages warning customers if they were getting close to an infected area.
In fact, mobile phones have continued to prove very useful in mapping disease. Last November, an iPhone app called ClickClinica was launched by a team at Liverpool University. As well as being a handy reference point for clinical guidelines, the app offers doctors the chance to record and upload their patient’s symptoms and the treatment given. Use of ClickClinica has grown steadily worldwide, and gives junior doctors a handy way to report infectious diseases - a job for which they’re generally responsible but rarely have the know-how.
But why use a doctor anyway? Nowadays, disease mapping technology doesn’t even need a hospital to report a case or analyse the spread of infection. Search engines and social websites have been unlikely allies to epidemiologists looking at flu outbreaks. While traditional surveillance systems for flu epidemics relied on data published once a week with a 2 week reporting lag, Google have tried to develop a program that analyses flu related search terms so they can detect outbreaks far sooner than the current system - with only a day’s reporting lag. The sooner we detect an illness, the sooner we can contain and stamp on it. Twitter has a lot more “background noise” than a search engine, and the use of disease related terms can be much harder to monitor, but scientists are working on a system that will get rid of a lot of the chit-chat and help track disease by usage patterns and time lines.
John Snow, with his door to door surveys and bars on maps, probably couldn’t even begin to understand how we map disease in a tech savvy world. But if he could have hopped in a time machine and been a guest at his own birthday party, I like to think he’d approve. In fact, I like to think he’d only have had one question.
Should he go Apple, or Android?