http://www.economist.com/node/21538656AS THE influenza season splutters into life across the northern hemisphere, millions will head to their computers in search of information, advice and remedies. Since 2008 Google has used these inquiries to track influenza-like illnesses (ILIs)—as symptoms not backed up by a definitive viral test are officially known—among its users around the world. Google Flu Trends displays whizzy graphs and colourful maps showing the intensity and progress of each seasonal epidemic.
This approach is not perfect, though. In order to stay accurate, Google has to tweak its algorithms regularly, to match the incidence of illness in the world. For this, it relies on data provided by America’s Centres for Disease Control and Prevention, and similar institutions in other countries. But different countries have different reporting cultures. Belgium, for example, typically reports five times as many ILIs as its neighbour, the Netherlands (employees’ need for a doctor’s certificate to take more than one day of medical leave is probably to blame), and even England and Scotland—supposedly part of the same United Kingdom—cannot agree on what constitutes a flu epidemic. The system is also prone to false alarms. When the H1N1 swine-flu pandemic stole headlines in the summer of 2009, Google searches went through the roof long before most people fell ill.
John Edmunds, of the London School of Hygiene and Tropical Medicine, and his colleagues hope they might be able to do better by recruiting volunteers in advance, for a “citizen science” project designed to discover how much influenza is really out there, how it spreads and how effective flu vaccines are. Citizen science is an increasingly popular technique that involves recruiting a large number of laymen into a study, and using them as observers and, sometimes, as data processors.
The Flu Survey is the first such project to ask people across Europe to sign up and report cold- or flu-like symptoms over the winter. It will operate in ten countries, posing identical questions; the organisers hope to recruit 50,000 participants. It will also, Dr Edmunds hopes, capture the kind of information that epidemiologists crave but that Google Flu Trends cannot provide. These elusive data include the age and sex of those affected, and whether they are part of a risk group, such as being asthmatic. The project should help, too, to catch new strains presenting novel symptoms, such as the unexpected diarrhoea that was associated with swine flu.
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
