On World Malaria Day, Chris and Xand Van Tulleken, working with the aid agency Merlin, highlight the devastating link between conflict and rates of malaria infection.
When we were studying tropical medicine in London, a favourite trick of one professor was to ask students what we thought the deadliest animal in the world might be.
One Ghanaian colleague suggested a lion riding on the back of an elephant, eating and trampling everything in sight.
Conflict causes 30% of malaria deaths in Africa. Picture: Jacqueline Koch.
But the required answer was the humble mosquito - or more precisely, the anopheles mosquito, the carrier and transmitter of malaria, a disease which kills over one million people every year.
This rather clever answer isn't entirely true.
Anopheles mosquitoes are found in harmless abundance in many places on earth. Malaria, once widespread even in Kent, has been eradicated from Europe and North America, as well as many parts of Asia.
Why then does it persist with such deadly effect in some parts of the world?
Poverty and weak health systems contribute hugely to the problem, but another, largely overlooked, factor is responsible for nearly 30% of all malaria deaths in Africa - conflict.
In 2007, 26 million people were driven from their homes by conflict. The effects of climate change - and conflict over limited resources like water, food and land - mean that every year, larger numbers of people are likely to be displaced.
When people flee conflict, they don't take hospital records with them. They don't take demographic data or disease patterns or any of the other details needed to tackle malaria.
They are often settled on land which has been abandoned because of the risk of malaria, or forced to live in over-crowded camps with limited health services, water, food and shelter.
In this vacuum of information and mass displacement, malaria is at its most deadly: frequently, more people die from the disease than the actual violence.
The British medical aid agency, Merlin, has been running emergency mobile clinics in Kenya's Rift Valley province since post-election violence forced hundreds of thousands to flee their homes.
More than 32,000 terrified people flooded into Nakuru district in the Rift Valley alone. Medics working on the ground simply don't have records of drug resistance levels or immunity for the displaced families now living in camps.
Cases of malaria are normally quite low in Nakuru, but with so many new people in the area, Merlin is aware that the chance of disease patterns changing is high, as is the risk of an outbreak.
Experience shows that there is no single effective solution to controlling malaria; mosquitoes, resistance to drugs and people's immunity all vary greatly from place to place.
Insecticide-treated bed nets have a vital role to play in preventing malaria, as Gordon Brown's recent pledge of $200m to fund mass net distribution demonstrates.
Drugs used in one part of the world may be useless in another.
But nets don't work so well if, like many displaced people, you have no bed, and no home.
Likewise, destruction of mosquito breeding sites can control the disease, but first you must know whether the local bugs breed in dirty, sunlit ground water or clean water in dark places.
Applying insect repellents to skin helps, but only if you know when the mosquitoes are likely to bite; anopheles gambiensis bites indoors at night (so bed nets work well), anopheles bellator bites outdoors at dusk.
Diagnosis poses similar problems.
Parachecks are rapid-test malaria kits, much like pregnancy tests. I used them to monitor for a malaria outbreak in Darfur; they were quick and easy for local staff to learn to use. My colleague who performed the tests was so proficient that he was popularly, and respectfully, known as Monsieur Paracheck.
These tests are not however appropriate in all settings and their usefulness depends on the number of people affected with the disease, the types of malaria, and the diagnostic information needed.
Treatment, again, varies. Drugs which can be effective within hours in one part of the world, may have such high resistance elsewhere to render them useless.
Malaria prevention, diagnosis and treatment require intensive, local information gathering which is often extremely difficult when people are still migrating or when violence is rife.
But all are essential to understand quickly and implement early if, as predicted, the trend for mass displacement caused by conflict continues to rise.
Insecticide on burkas
Solutions have to be tailored to specific circumstances.
Even in desperately vulnerable, displaced communities, it is possible to greatly reduce the number of people who die from the disease with well-designed, locally effective control programmes
An effective programme for Afghan refugees in Pakistan, rested on the discovery that the malaria mosquitoes there mostly feed on animals, and the displaced communities often live with their livestock so are constantly exposed to bites.
"Research showed that by coating the livestock with insecticide, malaria rates plummeted", explains Merlin's malaria advisor, Dr Ahmed Fayaz.
"But there were also unexpected benefits: the animals gained weight and milk production increased. These welcome side-effects ensured farmers continued to use the insecticide which protected them from malaria."
Sadly, these methods won't work in Africa, where mosquitoes tend to feed on people.
Nonetheless it is ideas founded in local knowledge such as these, or the technique of applying insecticide to women's burkas in Muslim countries, which can help save lives.
Malaria is a curable illness and a preventable disease. Even in desperately vulnerable, displaced communities, it is possible to greatly reduce the number of people who die from the disease with well-designed, locally effective control programmes.
As conflict forces more people from their homes, the need to put malaria control at the heart of any humanitarian response has never been more urgent.
Chris and Xand Van Tulleken presented the Channel 4 series Medicine Men Go Wild. Xand lectures on international public health at University College London, and Chris is a Senior House Officer in Infectious Diseases.