Living conditions may significantly increase a child's risk of malaria attacks, a study has suggested.
The researchers looked at protection measures
Wellcome Trust researchers found household differences in a Kenyan village accounted for around a third of the variations in attack rates.
In Public Library of Science they said practical measures, such as insecticide use, were more important than gene resistance.
Malaria kills around two million people a year, mainly in sub-Saharan Africa.
Genes versus environment
Dr Margaret Mackinnon, a researcher at the University of Edinburgh along with colleagues at the Kenya Medical Research Institute/Wellcome Trust Programme in Kilifi, carried out a five-year study of over 3,500 children living in the east of the country.
The first stage of the study looked at mild cases of malaria among 640 under-10s who lived in 77 households.
Just under a quarter of the variation in malaria fevers between the children was linked to genetic differences - but 29% of variation was linked to differences in household conditions.
The researchers also looked at a second group of 2,900 under-fives.
Around 1,000 of these were admitted to hospital at least once, with half suffering from malaria.
Again, there were differences between the households of those who were ill and those who were not.
Dr Mackinnon said: "It's quite remarkable to see the big variation in cases of malaria among children living almost side by side.
"Some suffered twice the amount of infections as those they played and even ate with.
"In this study we teased apart how much this was due to sharing genes versus sharing the same house.
"It turns out that living in the same home is more important than having the same genes."
She added: "We don't yet know exactly what makes the difference between a good or a bad house.
"But a lot probably depends on whether there is a mosquito-breeding site in the back yard, the quality of the building and whether insecticides or other repellents are used.
"Identifying and improving factors that put some homes at much lower risk than others would go a long way towards relieving the burden of disease in children living under such conditions."
She said, while studies of genes offering protection against malaria could lead to the development of new drugs or vaccines in the long-term, until then making sure homes carried as little of a risk as possible was key.
Dr Ian Hastings, of the Liverpool School of Tropical Medicine, said prevention methods such as insecticide treated nets and environmental measures had been shown to protect against malaria infection.
But he said the difference between households may not be due to differences in their use.
He told the BBC News website: "We know that houses will differ a lot in how many mosquitoes come in and bite people. Houses in the same village can differ 20-fold.
"An African village is not like a village in the Cotswolds. People can live two or three kilometres apart."
A second study in PLOS found babies born to mothers who had suffered from malaria during pregnancy were at an increased risk of the disease themselves.
And a third reported researchers from the Pasteur Institute in Paris are making progress in the development of a vaccine against malaria.
The team found antibodies in the blood of vaccinated volunteers were able to restrict the growth of the malaria parasite in laboratory studies.