Malaria on the MAP: the Malaria Atlas Project

Maps have always been critical in the planning of military campaigns: knowing where your enemy is, and how powerful, is essential to mounting an effective response. The same is true of the fight against disease. Simon Hay, Professor of Epidemiology in Oxford's Department of Zoology, is co-leader of the Malaria Atlas Project, a global project that is assembling spatial data relevant to malaria treatment and control, and representing it as a series of global and country-by-country maps.

'For 40 years we had no means of knowing how geographically widespread malaria was, or how severe', says Professor Hay. 'With the appearance in the 1980s of resistance to chloroquine, which had begun to get the disease under control, it became increasingly urgent to collect this data.' The project, co-founded in 2005-6 by Professor Hay and Professor Bob Snow of the KEMRI-Wellcome Research Programme in Nairobi, built on work Professor Snow had undertaken to map malaria in Africa.

The bedrock of the atlas is the parasite rate survey. Collaborators working in 85 malaria-endemic countries take blood samples from randomly-selected individuals in communities across each country and note the percentage that are infected at each map location. So far MAP has collated 25,000 community surveys for falciparum malaria, which causes the most deaths, and 10,000 for vivax malaria, which is less deadly but causes a high level of disease.

The new maps have many advantages over the hand-drawn disease maps that appeared in textbooks in the 1960s. They have the accuracy derived from modern satellite positioning technology, but they are also grounded in evidence that is freely available to others: the database is open-access. Modern statistics is also crucial to their usefulness. 'We provide detailed uncertainty data', says Professor Hay. 'For example, the data we have on the Democratic Republic of Congo, which is a war zone with little research infrastructure, is much more sparse that what we have from Indonesia or Kenya, where we have excellent contacts with established research groups through the Oxford Tropical Network. You can see these levels of uncertainty reflected on the maps.'

The maps are essential tools not only for health researchers but for national and international agencies. 'They help us to do useful epidemiological things, such as predict the numbers of likely cases', says Professor Hay. 'That leads to the economic impact: how many bednets will be needed, how many drugs and so on.' Recently he and his colleagues have used the data to audit spending on malaria by both national governments and international agencies, enabling them to identify regional imbalances between the level of need and the level of resource.

For the future, Professor Hay is excited by the potential of social media to contribute to this type of mapping project. 'There are 150m tweets a day', he says, 'and up to 10 per cent of them include geopositional information. It could be a quicker way to identify flu outbreaks, for example, than anything we have at the moment.'