Risk factors and spatial patterns of hookworm infection among schoolchildren in a rural area of western Co ˆte d’Ivoire Giovanna Raso a,b,c , Penelope Vounatsou a , Laura Gosoniu a , Marcel Tanner a , Elie ´zer K. N’Goran b,d , Ju ¨rg Utzinger a, * a Department of Public Health and Epidemiology, Swiss Tropical Institute, P.O. Box, CH-4002 Basel, Switzerland b Centre Suisse de Recherches Scientifiques, 01 BP 1303, Abidjan 01, Co ˆte d’Ivoire c Molecular Parasitology Laboratory, Australian Centre for International and Tropical Health and Nutrition, Queensland Institute of Medical Research, Brisbane, Qld 4029, Australia d UFR Biosciences, Universite ´ d’Abidjan-Cocody, 22 BP 770, Abidjan 22, Co ˆte d’Ivoire Received 24 June 2005; received in revised form 8 September 2005; accepted 9 September 2005 Abstract This study is aimed at investigating the risk factors for hookworm infection among schoolchildren in a rural area of western Co ˆ te d’Ivoire and predicting and mapping the spatial distribution of infection. We used demographic and socio-economic data from a cross-sectional survey of 6– 16-year-old schoolchildren from 56 schools. Infection with hookworm was determined by microscopic examination of stool samples employing the Kato–Katz technique and an ether-concentration method. Environmental data were derived from satellite images and digitised maps. Bayesian variogram models were applied to investigate the variation of hookworm infection in relation to demographic, socio-economic and environmental factors. The overall hookworm infection prevalence, based on the pooled microscopic diagnoses, was 43.3% and ranged from 5.4 to 79.1% in the schools surveyed. Bivariate analyses showed that sex, age, socio-economic status, elevation, rainfall and land cover were significantly associated with the spatial distribution of hookworm infection. The final multivariate spatial model consisted of the covariates age, sex, socio-economic status, elevation and land cover. When assuming non-stationary underlying spatial dependency, the results of the model suggested that spatial correlation depended on the location only marginally. We conclude that, at the current resolution, it seems more reasonable to target interventions based on well-established epidemiologic risk factors, rather than on spatial factors. q 2005 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Bayesian geo-statistics; Co ˆte d’Ivoire; Geographic information system; Hookworm (Ancylostoma duodenale and Nector americanus); Non-stationary spatial correlation; Risk prediction and mapping 1. Introduction Infections with hookworm (Ancylostoma duodenale and/or Necator americanus) occur worldwide, with the highest prevalences found in poor areas of the developing world. Important risk factors include the lack of sanitary waste disposal and access to clean water, often coupled with inadequate hygiene practices (de Silva et al., 2003; Hotez et al., 2004). The main public health significance of hookworm disease arises from morbid sequelae, such as poor iron status and iron- deficiency anaemia, which has been shown to have adverse effects on children’s growth and cognitive development, school performance, course and outcome of pregnancy and worker productivity (Albonico et al., 1999; Crompton, 2000; Gilgen et al., 2001; Brooker et al., 2004; Ezeamama et al., 2005). It is estimated that 740 million people are infected with hookworms (de Silva et al., 2003) and the global burden might be as high as 22.1 million disability-adjusted life years lost each year (WHO, 2002). An estimated 65,000 people die each year due to hookworm disease (WHO, 2002). The life cycles of A. duodenale and N. americanus involve transmission to humans either through contact with L3- contaminated soils, with L3 penetrating the skin (both A. duodenale and N. americanus), or when L3 is ingested (A. duodenale). The larvae migrate through the bloodstream to the lungs, move up the respiratory tract to enter the oesophagus and finally reach the small intestine where they mature and mate. An adult female worm produces thousands of eggs each International Journal for Parasitology 36 (2006) 201–210 www.elsevier.com/locate/ijpara 0020-7519/$30.00 q 2005 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2005.09.003 * Corresponding author. Tel.: C41 61 284 8129; fax: C41 61 284 8105. E-mail address: juerg.utzinger@unibas.ch (J. Utzinger).