Scientists have developed a climate-based model to predict outbreaks of dengue fever with 60 per cent accuracy up to 40 weeks in advance.

The model, developed by researchers from the University of Miami (UM) and the University of Costa Rica, was created using data from Costa Rica but could be used in any dengue-prone area in Africa, Asia, Latin America and the Caribbean, the researchers say.

About two-thirds of the world's population resides in areas infested with the Aedes aegypti and Ae. albopictus mosquitoes, which transmit dengue fever (DF) and its more deadly complication dengue haemorrhagic fever (DHF). Between 50 and 100 million cases occur each year, mostly in tropical and subtropical areas.

The new model predicts outbreaks using data on sea-surface temperature coupled with changes in vegetation. These are linked to evaporation and humidity near the ground, where mosquitoes breed.

Similar models have been produced for malaria but predicting dengue has proved more difficult as few countries have the years of data on dengue cases needed to implement such a model. But countries could access information on vegetation and sea temperatures from meteorological centres and satellite images for example.

Douglas Fuller, principal investigator for the research and associate professor at UM, told SciDev.Net: "If we can alert authorities to the increased likelihood of dengue outbreaks based on [previous] climatic conditions, they will be better prepared. Thus, our model could be used as the basis for an early warning system similar to famine or hurricane warning systems."

The model was tested using data from DF and DHF cases in Costa Rica, successfully forecasting a major epidemic that occurred in 2005.

"It has also been tested using data from Singapore and Trinidad. In both cases the model reproduced epidemics very well," says Fuller.

Dziedzom De Souza, researcher at the Noguchi Memorial Institute for Medical Research at the University of Ghana, says the study shows promising applications.

"However, it would be good to see a follow-up study on its application in other countries and the ease with which it is applied since most countries — especially low to middle income countries — lack the database infrastructure required for the application of such models,".

Source: SciDev.Net
Date: 18 June, 2009