Negative Spatial Association Between Lymphatic Filariasis and Malaria in Africa

Louise A. Kelly-Hope, Vector Research Group, Liverpool School of Tropical Medicine, Liverpool, UK
Peter J. Diggle, Medical Statistics Unit, Lancaster University, UK & Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health
B.S. Rowlingson, Department of Mathematics & Statistics, Lancaster University
J.O. Gypapong, Health Research Unit, Ghana Health Service, Ministry of Health, Accra, Ghana
D. Kyelem, Division of Preventive Medicine, Ministry of Health, Ouagdougou, Burkina Faso
M. Coleman, Vector Research Group, Liverpool School of Tropical Medicine, Liverpool, UK
M.C. Thomson, International Research Institute for Climate Prediction, Earth Institute at Columbia University
V. Obsomer, Prince Leopold Institute of Tropical Medicine, Antwerpen, Belgium
S.W. Lindsay, School of Biological and Biomedical Sciences, University of Durham, Durham, UK
J. Hemingway, Vector Research Group, Liverpool School of Tropical Medicine
D.H. Molyneux, Lymphatic Filariasis Support Centre, Liverpool School of Tropical Medicine, Liverpool, UK

Abstract

Human lymphatic filariasis (LF), caused by Wuchereria bancrofti, is a disabling parasitic disease endemic throughout sub-Saharan Africa. A detailed inter-country study in West Africa using a grid sampling technique for the rapid assessment of LF distribution has demonstrated that W. bancrofti prevalence varies considerably throughout Benin, Burkina Faso, Ghana and Togo. Here we show, using geographic information systems (GIS) and spatial statistics that a robust negative association between LF and malaria prevalence exists. This is a surprising finding, given that in rural West Africa both diseases are transmitted by mosquitoes of the Anopheles gambiae complex. A key factor distinguishing the two disease epidemiologies may be the different distributions and vector competence of the An. gambiae cytoforms. Alternatively interspecies competition between the parasites, with W. bancrofti being more dominant where malaria transmission is more seasonal, may drive the spatial distributions. Differentiating between the various hypotheses may become crucial as large-scale LF and malaria control programmes are implemented in West Africa that may impact on the epidemiology of both diseases.