Authors
L NelliH FergusonH Ranson3; S W Lindsay2; S Nfale1; A Tiono1; A S SanouJ Matthiopoulos 1 Centre National de Recherche et de Formation sur le Paludisme, Burkina Faso; 2 Durham University; 3 Liverpool School of Tropical Medicine; 4 University of Glasgow Discussion
High levels of insecticide resistance (IR) in the primary mosquito vectors is hypothesized to be a major contributor to persistent levels of malaria transmission. As part of a long-term interdisciplinary programme, we are developing a spatial modelling framework aimed at quantifying the contribution of IR within vector populations on local patterns of malaria incidence. While much of the data required to formulate this model is being collected, we are developing the model framework definition. Initial work has focused on a simulation approach to generate a simulated dataset of malaria infection reported across the whole study area, as a function of environmental factors. A subsequent modelling approach based on Bayesian statistics was used to test whether the model was able to correctly predict a range of hypothetical relationships between IR, environmental variables and local malaria incidence, against a simulated background of realistic levels of uncertainty in the measurements. Preliminary results indicate that our approach seems promising for being able to pick up subtle effects of IR on malaria incidence – if they exist – even within the type of patchy and incomplete surveillance data that is typically collected, and provides a useful framework for a more quantitative analysis of the impacts of IR on malaria control efforts – as is needed to anticipate the potential consequences and how to respond to them. 
