SEARCH

SEARCH BY CITATION

Keywords:

  • Schistosoma mansoni;
  • hookworm;
  • risk factors;
  • urban agriculture;
  • Côte d'Ivoire
  • Schistosoma mansoni;
  • ankylostome;
  • facteurs de risque;
  • agriculture urbaine;
  • Côte d’ Ivoire
  • Schistosoma mansoni;
  • anquilostoma;
  • factores de riesgo;
  • agricultura urbana;
  • Costa de Marfil

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives  To identify risk factors for Schistosoma mansoni and hookworm infections in urban farming communities, and to investigate small-scale spatial patterns of infection prevalence.

Methods  A cross-sectional survey was carried out in 113 farming households (586 individuals) and 21 non-farming households (130 individuals) from six agricultural zones in the town of Man, western Côte d'Ivoire. Heads of households were interviewed on common agricultural activities, land and water use, education attainment, socioeconomic status and sanitation facilities. Household members provided stool specimens that were processed by the Kato-Katz technique and a formol-ether concentration method and diagnosed for S. mansoni, hookworms and other soil-transmitted helminths and intestinal protozoa. Bayesian statistics were employed for spatial analyses.

Results  The prevalences of S. mansoni and hookworm in farming households were 51.4% and 24.7%, respectively. Risk factors for a S. mansoni infection comprised living in close proximity to the Kô River, water contact with irrigation wells and ponds and low education attainment. Living in zones of smallholder irrigated rice plots or large rice perimeters, using water from domestic wells, and low socioeconomic status were risk factors for a hookworm infection. We found significant spatial heterogeneity between agricultural zones, with the highest infection prevalences of S. mansoni and hookworm in the zone where there was a large rice perimeter.

Conclusions  In this urban setting, both S. mansoni and hookworm infections were related to specific agricultural activities. Health education and active participation of urban farmers for the control of schistosomiasis and soil-transmitted helminthiasis is recommended.

Objectifs  Identifier les facteurs de risque associés aux infections par Schistosoma mansoni et par l'ankylostome dans des communautés urbaines d'agriculteurs et investiguer les profils spatiaux de la prévalence des infections sur une petite échelle.

Méthodes  Etude transversale sur 113 familles d'agriculteurs (586 individus) et 21 familles de non agriculteurs (130 individus) provenant de six zones agricoles dans la ville de Man, dans l'ouest de la Côte d'Ivoire. Les chefs de familles ont été interviewés sur les pratiques agricoles en cours, l'utilisation des terres et de l'eau, l’éducation, le statut socio-économique et les installations sanitaires. Des échantillons de selles ont été collectés chez les membres des familles et ont été traités par la technique de Kato-Katz et par une méthode de concentration au formol-éther, puis diagnostiqués pour S. mansoni, l'ankylostome et autres helminthes transmis à partir du sol ainsi que pour des protozoaires intestinaux. Des statistiques bayésiennes ont été utilisées pour les analyses spatiales.

Résultats  Les prévalences de S. mansoni et de l'ankylostome dans les familles d'agriculteurs étaient de 51,4% et 24,7%. Les facteurs de risque pour l'infection àS. mansoni comprenaient: la vie à proximitéétroite du fleuve Kô, le contact avec l'eau des puits d'irrigation et celle des étangs et le bas niveau d’éducation. La vie dans les zones à petites exploitations de parcelles irriguées de riz ou à grands périmètres de riz, l'utilisation de l'eau des puits domestiques et le bas statut socio-économique étaient des facteurs de risque pour l'infection à l'ankylostome. Nous avons trouvé une hétérogénéité spatiale significative entre les zones agricoles, avec les plus hautes prévalences pour l'infection àS. mansoni et à l'ankylostome dans la zone à grands périmètres de riz.

Conclusions  Dans la zone urbaine étudiée, les infections àS. mansoni et à l'ankylostome étaient associées à des activités agricoles spécifiques. L’éducation sanitaire et la participation active des fermiers urbains au contrôle de la schistosomiase et de l'helminthiase transmise à partir du sol est recommandée.

Objetivos  Identificar factores de riesgo para infecciones por Schistosoma mansoni y anquilostoma en comunidades agrícolas urbanas, e investigar patrones espaciales a pequeña escala sobre la prevalencia de infección.

Métodos  Estudio croseccional de 113 hogares agrícolas (586 individuos) y 21 hogares no-agrícolas (130 individuos) de seis zones agrícolas en la población de Man, al oeste de Costa de Marfil. Se entrevistó a los cabezas de familia sobre las prácticas agrícolas más comunes, el uso del agua y la tierra, el nivel educativo, el estatus socio-económico y las servicios sanitarios. Los miembros del hogar proveyeron muestras coprológicas que fueron procesadas mediante la técnica Kato-Katz yla técnica de concentración por formol-éter, siendo diagnosticados para S. mansoni, anquilostoma y otros helmintos y protozoos intestinales transmitidos por el suelo. Se utilizó la estadística Bayesiana para realizar un análisis espacial.

Resultados  La prevalencia de S. mansoni y anquilostoma en hogares agrícolas fue del 51.4% y 24.7%. Los factores de riesgo para infección por S. mansoni incluían el vivir en proximidad al río Kô, el contacto con el agua, con pozos de irrigación y un nivel de educación bajo. El vivir en zonas de minifundios con cultivos de arroz irrigados, o perímetros mayores de cultivo de arroz, el utilizar agua de pozos domésticos y un nivel socio-económico bajo eran factores de riesgo para infección por anquilostoma. Encontramos una heterogeneidad espacial significativa entre zonas agrícolas, con prevalencias de infección más altas para S.mansoni y anquilostoma en la zona en la que había un mayor perímetro de cultivo de arroz.

Conclusiones  En este asentamiento urbano, las infecciones por S. mansoni y anquilostoma estaban relacionadas con actividades agrícolas específicas. Se recomienda la educación sanitaria y la participación activa de agricultores urbanos para el control de esquistosomiasis y helmintiasis transmitido por tierra.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Soil-transmitted helminthiasis and schistosomiasis are endemic in most regions of the humid tropics (Bethony et al. 2006; Gryseels et al. 2006). More than one billion people are infected with any of the three major soil-transmitted helminths, namely Ascaris lumbricoides, hookworms and Trichuris trichiura (de Silva et al. 2003; Bethony et al. 2006), and an estimated 207 million people are infected with schistosomes (Steinmann et al. 2006). The global burden of soil-transmitted helminthiasis might be as high as 39 million disability-adjusted life years (DALYs), and hence approaching that of malaria (Hotez et al. 2006; Lopez et al. 2006).

Hookworm larvae prefer warm and partially shaded sandy soils, which are common in rural areas of the tropics and poor urban settlements (Brooker et al. 2004). An infection with hookworms is accomplished when the unprotected human skin is exposed to such soils (e.g. walking barefoot). Previous studies have shown that hookworm disease is common among farmers and vegetable growers, and is often associated with wastewater and night soils that are used to enhance agricultural output (Brooker et al. 2004; Ensink et al. 2005). Schistosomiasis is typically considered as a rural disease. Transmission occurs when humans contact freshwater bodies infested with cercariae that have been released by intermediate host snails (Gryseels et al. 2006). In view of schistosomiasis being closely linked to human water contact patterns, irrigated agriculture plays an important role (Huang & Manderson 1992; Bethony et al. 2004; Steinmann et al. 2006).

Available data suggest that there is a declining trend in the prevalence of soil-transmitted helminth infections in many parts of the developing world. The causes are multifactorial, including urbanization, economic advancement and abandoning agrarian lifestyles in rural settings (de Silva et al. 2003; Hotez et al. 2006). However, in many parts of Africa and Asia where urbanization progresses at a rapid pace (Utzinger & Keiser 2006), urban farming has become an important livelihood strategy (Bryld 2003). Urban agriculture is linked with farming-related health issues, for example, through domestic and industrial waste disposal, which in turn contaminate soils and water (Binns et al. 2003; Amoah et al. 2006). Only few studies have investigated the epidemiology of soil-transmitted helminthiasis and schistosomiasis in urban settings (Firmo et al. 1996; Ximenes et al. 2003; Fournet et al. 2004; Brooker et al. 2006b). There is a paucity of this kind of investigation focusing on urban farming communities.

In a preceding study, we mapped and characterized productive Anopheles breeding sites and investigated risk factors for malaria in urban farmers and their families living in different zones of irrigated crop systems in a district town of western Côte d'Ivoire (Matthys et al. 2006a,b). The objectives of this study were (i) to identify risk factors for hookworm and Schistosoma mansoni infections in these urban farming communities, and (ii) to assess micro-geographical heterogeneity of infection prevalence, using Bayesian spatial statistics.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study area and population

The study area and the population surveyed have been described before (Matthys et al. 2006a,b). In brief, the study was carried out between April 2004 and June 2005 in the town of Man, located in the western part of Côte d'Ivoire. The population of Man is estimated at 115 000 people. One-third of the households are engaged in urban agriculture, primarily along the banks of the Kô River and in small inland valleys in the south-western part of the town. Figure 1 shows the six selected agricultural zones with irrigated crop systems. Zones 1 and 3 consisted of mixed crop systems with vegetable gardens and traditional smallholder rice plots with one cropping cycle per year. Zones 4 and 6 were smallholder rice plots. Zones 5 and 7 consisted of large rice perimeters allowing two cropping cycles. In the latter zone, smallholder rice paddies were also present. Zone 2 is located furthest north at the outskirts of Man. Because of logistical problems, no parasitological survey was carried out in this zone, but questionnaires were pre-tested here.

image

Figure 1.  Spatial distribution of Schistosoma mansoni and hookworm mono-infection and co-infection prevalence for individuals from farming households, stratified by agricultural zone, in the town of Man, western Côte d'Ivoire in June 2005.

Download figure to PowerPoint

A total of 131 farming households were identified by directly contacting farmers during their work in the fields (Matthys et al. 2006b). We sought oral consent and then collected geographical coordinates of the houses and the main agricultural plots using a hand-held global positioning system (Garmin eTrex and 12XL; Garmin International Inc., Olathe, KS, USA). The name, age and sex of all household members were recorded (n = 1164) and assigned unique identifiers. An additional 34 households not engaged in urban agriculture were randomly selected from the same study area using the EPI survey approach (Lemeshow & Robinson 1985). The estimated population in these non-farming households was 272 people.

Questionnaire surveys

In October 2004, a questionnaire was administered to the heads of the farming households. Questions pertained to agricultural activities and land and water use (Matthys et al. 2006b). In June 2005, heads of both farming and non-farming households were interviewed on education attainment, socioeconomic status and sanitation facilities. Information was also collected from all household members on common water contact patterns, including types of water bodies contacted (i.e. Kô River, creek, pond, irrigation canal and wells), water-related activities (i.e. swimming/bathing, fetching water for irrigation, cleaning irrigation canals, washing clothes and fishing), and frequency of water contact patterns. As mentioned before, the questionnaires were pre-tested in zone 2. In addition, snail surveys were conducted in all agricultural zones alongside Anopheles breeding site assessments that have been described elsewhere (Matthys et al. 2006a). Intermediate host snails of S. mansoni (i.e. Biomphalaria pfeifferi) were transferred to a laboratory at the Université de Cocody-Abidjan, exposed to artificial light and checked for cercarial shedding.

Parasitological survey and laboratory investigations

Field and laboratory procedures have been detailed elsewhere (Matthys et al. 2006b). Briefly, each household was visited in the evening, the purpose of the parasitological survey explained and all household members invited to provide a stool specimen. Small plastic containers were handed out for stool collection. The following day, the filled containers were collected, attached with unique identifiers and transferred to a laboratory in Man.

A two-pronged diagnostic approach was employed. Firstly, a small amount of stool (1–2 g) was preserved in a tube filled with 10 ml sodium acetate–acetic acid formalin (SAF). These SAF-preserved stool samples were forwarded to a laboratory in Abidjan. Within 12 months, they were processed by a formol-ether concentration method and examined by experienced laboratory technicians for the presence of S. mansoni and soil-transmitted helminth eggs, and cysts or trophozoites of intestinal protozoa (Marti & Escher 1990). Secondly, two Kato-Katz thick smears were prepared from each stool specimen on microscope slides, using 42-mg plastic templates (Katz et al. 1972). Slides were cleared for 30–45 min and then examined under a light microscope by experienced laboratory technicians. Eggs of S. mansoni, hookworm, A. lumbricoides and T. trichiura were counted and recorded separately. For quality control, a random sample of 10% of the slides was re-examined the same day by a senior technician and discrepancies were discussed.

Data management and statistical analysis

Data were double-entered and cross-checked using the EpiData software version 3.1 of (EpiData Association; Odense, Denmark). Statistical analysis was performed in stata version 9 (Stata Corporation; College Station, TX, USA) and in WinBUGS version 1.4.1 (Imperial College and Medical Research Council; London, UK). Maps and shortest straight-line distances from houses to the Kô River were obtained in ArcMapTM version 9.0 (Environmental Systems Research Institute; Redlands, CA, USA).

Study participants with complete parasitological data (i.e. ≥1 Kato-Katz thick smear plus SAF reading) and questionnaire data were included in the final analysis. Age was grouped into five classes: (i) <10, (ii) 10–14, (iii) 15–24, (iv) 25–39 and (v) ≥40 years. An infection with S. mansoni or hookworm was defined as the presence of at least one egg either in the Kato-Katz thick smear or the SAF-conserved stool sample. Intestinal protozoan infections were defined by the presence of intestinal protozoa in the SAF reading. For participants with two Kato-Katz thick smears, the arithmetic mean egg count was calculated. Schistosoma mansoni infection intensity was classified as follows: light [1–99 eggs per gram of stool (epg)], moderate (100–399 epg) and heavy (≥400 epg). For hookworm infections, intensities were grouped as follows: light (1–1999 epg), moderate (2000–3999 epg) and heavy (≥4000 epg) (WHO 2002).

The socioeconomic status of a household was derived by principal component (PC) analysis (PCA), using selected housing characteristics (e.g. type of wall) and household assets owned (e.g. bicycle). Wealth quintiles were derived from the first PC according to a widely used methodology (Gwatkin et al. 2000; Filmer & Pritchett 2001), readily adapted to the local context (Raso et al. 2006a). Pearson's χ2, Fisher's exact test and Wilcoxon signed rank test were used, as appropriate, to compare proportions or medians between groups.

Logistic regression models were fitted to investigate risk factors for an infection with S. mansoni or hookworm. Interactions between age, socioeconomic status, education and agricultural zone were examined using the likelihood ratio test. Explanatory variables were entered in the Bayesian spatial and non-spatial multiple logistic regression models in case they were significant at a 15% significance level. Household-specific random effects with an exchangeable correlation structure were introduced to take into account the between-household variation (inline image). Likewise, random effects related to the location of the household were employed to model geographical correlation. The spatial correlation was assumed to be an exponential function of the distance, i.e. inline image, including the shortest straight-line distance between households k and l (dkl), the geographical variability known as sill (inline image) and a smoothing parameter that controls the rate of correlation decay with distance (ρ). The minimum distance is represented by 3/ρ in meters at which spatial correlation between house locations is below 5% and is known as the range of geographical dependency. We fitted two different multiple logistic regression models for S. mansoni and hookworm infection prevalence. While the first model did not take into account any random effects, the second model considered between-household variation. Whilst gamma prior distributions were used for inline image and inline image, a uniform prior distribution was employed for ρ. The model parameters were fitted applying Markov chain Monte Carlo simulation (Gelfand & Smith 1990). As a goodness of fit measure, we used the deviance information criterion (DIC) (Spiegelhalter et al. 2002). The model with the smallest DIC was considered the best fitting one. Further details on model specifications have been presented elsewhere (Matthys et al. 2006b).

Ethical considerations and treatment

Institutional approval of the study protocol was granted by the Swiss Tropical Institute (Basel, Switzerland) and the Centre Suisse de Recherches Scientifiques (Abidjan, Côte d'Ivoire). The study received ethical clearance by the Ministry of Public Health in Côte d'Ivoire.

Study participants infected with soil-transmitted helminths were treated with a single 400-mg oral dose of albendazole and participants with a S. mansoni infection were given a single 40-mg/kg oral dose of praziquantel, adhering to WHO guidelines (WHO 2002).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study compliance and sociodemographic characteristics

Overall, 1000 individuals from 159 households participated in our cross-sectional surveys. However, 124 individuals had no or incomplete questionnaire data, and 160 people were absent during our parasitological survey or provided insufficient amounts of stool to prepare two Kato-Katz thick smears. The final study cohort therefore consisted of 716 people from 134 households (586 individuals from 113 farming households and 130 individuals from 21 non-farming households). Sex and age distributions were similar for farming and non-farming households. Among these 716 individuals, we were able to prepare an additional SAF-conserved stool sample for examination of intestinal protozoa (510 individuals from farming and 90 in non-farming households).

The demographic and socioeconomic profiles of the farming households have been described before (Matthys et al. 2006b). Briefly, two-thirds of the household heads were illiterate, and very few received specific agricultural training. The socioeconomic status of the farmers in the different agricultural zones was similar. Half of the households had their own latrines (with a slab made of cement or terra soil), and one-third of the households shared a latrine with their neighbours. The remaining households (17%) had no latrines. The majority of households were inhabited by less than 10 people, but in one out of 20 households over 20 people were registered.

Parasitic infections in farming and non-farming households

The results from the cross-sectional parasitological survey are summarized in Table 1, stratified by farming and non-farming households. The overall infection prevalence of S. mansoni and hookworm was 50.1% and 22.9%, respectively. The prevalence of hookworm infections in farming households was significantly higher than that in non-farming households (24.7%vs. 14.6%; odds ratio (OR) = 0.52 for non-farming households, 95% confidence interval (CI) = 0.31–0.88). Infections with T. trichiura were found in 3.4% and A. lumbricoides in 1.1% of the participants.

Table 1.   Number (%) of individuals infected with Schistosoma mansoni, soil-transmitted helminths and intestinal protozoa, stratified by farming and non-farming households in the town of Man, western Côte d'Ivoire
Parasite Overall (n = 716)Farming households (n = 586)Non-farming households (n = 130) OR† 95% CIP-value‡
  1. †OR, crude odds ratio (OR = 1.00 for farming households).

  2. P-value based on likelihood ratio test.

  3. §The prevalence of intestinal protozoan infections is based on the SAF-conserved stool specimens examined by an ether-concentration method.

S. mansoni359 (50.1)301 (51.4)58 (44.6)0.760.52, 1.120.164
Soil-transmitted helminths
 Hookworm164 (22.9)145 (24.7)19 (14.6)0.520.31, 0.880.010
 Trichuris trichiura24 (3.4)19 (3.2)5 (3.8)1.190.44, 3.260.734
 Ascaris lumbricoides8 (1.1)7 (1.2)1 (0.8)0.640.08, 5.260.662
Intestinal protozoa§(n = 600)(n = 510)(n = 90)   
 Entamoeba coli320 (53.3)278 (54.5)42 (46.7)0.730.47, 1.140.170
 Blastocystis hominis225 (37.5)188 (36.9)37 (41.1)1.200.75, 1.890.445
 Endolimax nana124 (20.7)113 (22.2)11 (12.2)0.490.25, 0.950.024
 Entaemoeba histolytica/Entaemoeba dispar109 (18.2)103 (20.2)6 (6.7)0.280.12, 0.660.001
 Iodamoeba buetschlii101 (16.8)90 (17.6)11 (12.2)0.650.33, 1.270.190
 Chilomastix mesnili50 (8.3)45 (8.8)5 (5.6)0.610.23, 1.280.278
 Entamoeba hartmanni37 (6.2)34 (6.7)3 (3.3)0.480.15, 1.610.192
 Giardia duodenalis32 (5.3)32 (6.3)0   

Entamoeba coli (prevalence 53.3%) and Blastocystis hominis (37.5%) were the most common apathogenic intestinal protozoa. We found an overall prevalence of Entamoeba histolytica/Entamoeba dispar of 18.2%, with a three-fold higher prevalence observed in farming compared with non-farming households, accounting for a highly significant difference (OR = 0.28 for non-farming households, 95% CI = 0.12–0.66). Whilst the prevalence of Giardia duodenalis in farming households was 6.3%, not a single infection was detected in non-farming households, resulting in a highly significant difference (Fisher's exact text, P = 0.005).

Concomitant infections with two or more intestinal parasites were observed in 69.6% of the individuals from farming households compared with 55.6% from non-farming households. Absence of intestinal parasites was noted in 13.1% of the individuals from farming families compared with 20.0% from non-farming families (χ2 = 6.94, degree of freedom (d.f.) = 2, P = 0.031).

Schistosoma mansoni and hookworm data

Table 2 summarizes the prevalence of S. mansoni and hookworm infections among individuals from farming and non-farming households, stratified by sex, age, socioeconomic status, education level and agricultural zone. In farming households, males had a significantly higher S. mansoni infection prevalence than females (58.8%vs. 44.9%; χ2 = 11.30, d.f. = 1, P = 0.001), whereas no significant sex difference was found in non-farming households. In both farming and non-farming households, an infection with S. mansoni was significantly associated with age, with a peak in the prevalence found in 15–24 years olds. In farming families there was a significant association with S. mansoni infection and education attainment; the prevalence of S. mansoni infection among college and high school attendants was 38.9% whereas 56.6% of the people not attending school were infected (χ2 = 9.76, d.f. = 2, P = 0.008). The observed prevalence of S. mansoni did not differ significantly between the six agricultural zones, neither for farming nor for non-farming household members.

Table 2.   Infection prevalence of Schistosoma mansoni and hookworm among individuals from farming and non-farming households, stratified by sex, age, socioeconomic status, education level and agricultural zone in the town of Man, western Côte d'Ivoire
ParameterFarming households (n = 586)Non-farming households (n = 130)
TotalS. mansoni no. (%)P-value†Hookworm no. (%)P-value†TotalS. mansoni no. (%)P-value†Hookworm no. (%)P-value†
  1. P-value based on Pearson's chi-square-test.

  2. P-value based on Fisher's exact test.

Sex
 Male272160 (58.8) 77 (28.3) 5324 (45.3) 6 (11.3) 
 Female314141 (44.9)0.00168 (21.7)0.0637734 (42.2)0.89913 (16.9)0.378
Age (years)
 <1014748 (32.7) 14 (9.5) 346 (17.7) 4 (11.8) 
 10–149359 (63.4) 29 (31.2) 2210 (45.5) 3 (13.6) 
 15–2412180 (66.1) 39 (32.2) 2818 (64.3) 9 (32.1) 
 25–397645 (59.2) 24 (31.6) 2516 (64.0) 1 (4.0) 
 ≥4014969 (46.3)<0.00139 (26.2)<0.001218 (38.1)0.0012 (9.5)0.067‡
Socioeconomic status
 Poorest14580 (55.2) 39 (26.9) 00 0 
 Very poor14483 (57.6) 30 (20.8) 32 (66.7) 0 
 Poor9951 (51.5) 28 (28.3) 4017 (42.5) 5 (12.5) 
 Less poor9643 (44.8) 17 (17.7) 4621 (45.7) 4 (8.7) 
 Least poor10244 (43.1)0.11031 (30.4)0.1644118 (43.9)0.896‡10 (24.4)0.213‡
Education level
 No school318180 (56.6) 77 (24.2) 5625 (44.6) 8 (14.3) 
 Primary school17886 (48.3) 53 (29.8) 5023 (46.0) 5 (10.0) 
 College/high school9035 (38.9)0.00815 (16.7)0.0602410 (41.7)0.9406 (25.0)0.240‡
Agricultural zone
 15528 (50.9) 8 (14.6) 188 (44.4) 3 (16.7) 
 36840 (58.8) 22 (32.4) 0    
 46627 (40.9) 21 (31.8) 3414 (41.2) 2 (5.9) 
 513365 (48.9) 47 (35.3) 3117 (54.8) 10 (32.3) 
 618293 (51.1) 28 (15.4) 2910 (34.5) 2 (6.9) 
 78248 (58.5)0.26619 (23.2)<0.001189 (50.0)0.5762 (11.1)0.031‡

With regard to hookworm infections, males in farming households had a higher prevalence than females, but the difference was not statistically significant (28.3%vs. 21.7%; χ2 = 3.46, d.f. = 1, P = 0.063). Prevalence of infection showed a highly significant age-relationship in farming households with highest prevalences (31.2–32.2%) found in the three age groups 10–14, 15–24 and 25–39 years (χ2 = 26.10, d.f. = 5, P < 0.001). In contrast to S. mansoni, no significant association was found between hookworm infections and education attainment, but hookworm infection prevalence showed significant spatial heterogeneity between agricultural zones. The highest prevalence in both farming and non-farming households was observed in zone 5 (35.3% in farming households and 32.3% in non-farming households).

Spatial distribution of Schistosoma mansoni and hookworm infection among farming households

Figure 1 shows the micro-spatial distribution of S. mansoni and hookworm mono-infections and co-infections for individuals from farming households. Schistosoma mansoni mono-infections were elevated in zones 1, 6 and 7, located in the western part of the study area. The highest levels of hookworm mono-infections were found in zones 4 and 5, located in the eastern part of the study area. The prevalence of S. mansoni-hookworm co-infections ranged from 8.2% (zone 6) to 26.5% (zone 3).

Schistosoma mansoni and hookworm infection intensities

Schistosoma mansoni infection intensities were significantly different between farming and non-farming households (χ2 = 10.05, d.f. = 2, P = 0.007). Heavy S. mansoni infections, defined as ≥400 epg, were found in 24.1% of the individuals from farming households, compared with only 5.4% in non-farming households. With regard to hookworm infection intensities, 11.3% of the study participants from farming households had moderate-to-heavy infections (defined as ≥2000 epg), compared with 5.3% individuals from non-farming households.

Household clustering of Schistosoma mansoni and hookworm infection

Figure 2 shows household-clustered S. mansoni and hookworm infections from 94 and 59 farming households, respectively, with at least one household member infected. The results are further stratified by agricultural zone. After calculating the arithmetic mean (according to Brooker and colleagues (2006b)), egg counts of S. mansoni and hookworm for the respective sex and age-class and the standard deviation (SD) for each individual was computed from this mean. The data were averaged by household, and standardized with a mean of 0 and a SD of 1 (Behnke et al. 2000). The average level of infection is denoted by the horizontal zero line. Schistosoma mansoni and hookworm infections showed pronounced aggregation in a few households. Zone 6 showed the highest averaged S. mansoni egg counts per family. Meanwhile, the averaged hookworm infections were below the mean in this zone.

image

Figure 2.  Household clustering of Schistosoma mansoni and hookworm infection from six agricultural zones in the town of Man, western Côte d'Ivoire in June 2005.

Download figure to PowerPoint

Risk factors for Schistosoma mansoni and hookworm infections

Demographic, socioeconomic and farming-related risk factors for an infection with S. mansoni and hookworm are summarized in Tables 3 and 4. Key findings from the multivariate spatial random effects models are highlighted. The risk of a S. mansoni infection was age and sex-related. Males were at a higher risk of a S. mansoni infection than females, as were individuals aged geqslant R: gt-or-equal, slanted 10 years when compared with children below the age of 10 years. The socioeconomic status was not significantly associated with a S. mansoni infection. A higher education level of the household head was a protective factor (OR = 0.37, 95% Bayesian credible interval (BCI) = 0.19–0.63 for household heads attending college/high school).

Table 3.   Results of bivariate, non-random model and spatially-explicit random effects multivariate logistic regression model. Outcome: Schistosoma mansoni infection; explanatory variables: demographic and education, socioeconomic, household characteristics and agricultural parameters. Medians were used to summarize the posterior distribution of the parameters of the Bayesian model
Explanatory variablesBivariate modelBayesian multivariate model†
OR95% CIP-value‡OR95% BCI
  1. BCI, Bayesian credible interval.

  2. †Spatially-explicit model with household and location-specific random effects.

  3. P-value based on likelihood ratio test.

  4. §Not included in the multivariate model.

Demography
 Sex
  Male1.00  1.00 
  Female0.570.41, 0.790.0010.460.29, 0.67
 Age (years)
  <101.00  1.00 
  10–143.582.08, 6.17 4.772.34, 8.81
  15–244.022.42, 6.70 5.812.94, 10.52
  25–392.991.69, 5.31 5.702.61, 11.04
  ≥401.781.11, 2.85<0.0012.321.21, 4.08
Socioeconomic status
 Poorest1.00  1.00 
 Very poor1.110.69, 1.76 1.940.97, 3.52
 Poor0.860.52, 1.44 1.070.50, 2.01
 Less poor0.660.39, 1.11 1.240.57, 2.39
 Least poor0.620.37, 1.030.1090.930.42, 1.78
Education level
 No school1.00  1.00 
 Primary school0.720.50, 1.04 0.690.40, 1.10
 College/high school0.490.30, 0.790.0070.370.19, 0.63
Number of persons living in household
 1–51.00  1.00 
 6–101.220.74, 2.03 1.240.62, 2.23
 11–200.940.55, 1.60 1.190.57, 2.22
 >201.870.97, 3.260.0832.561.07, 5.23
Toilet disposal
 Latrine used by only one household1.00  1.00 
 Shared latrine within household yard1.260.87, 1.80 1.691.02, 2.65
 No toilet1.961.24, 3.100.0141.960.96, 3.60
Distance to river
 ≤750 m1.00  1.00 
 >750 m0.570.40, 0.820.0020.610.37, 0.93
Bathing/swimming in Kô River1.640.97, 2.780.0623.161.42, 6.25
Fishing with a net1.641.09, 2.460.0162.351.30, 3.93
Farming-related risk factors
 Contact with water from irrigation well or pond1.741.25, 2.42<0.0012.501.60, 3.76
Cultivated crop type
 Rain-fed rice§1.000.69, 1.360.995  
 Irrigated rice§1.240.86, 1.780.251  
 Market garden vegetables0.580.42, 0.820.0020.760.45, 1.20
 Rain-fed food crops1.671.19, 2.330.0031.450.93, 2.16
inline image (non-spatial variation)   0.120.03, 0.40
inline image (spatial variation)   0.040.00, 0.25
ρ (smoothing parameter)   0.540.08, 0.98
Deviance information criterion    730.0
Table 4.   Results of bivariate, non-random model and spatially-explicit random effects multivariate logistic regression model. Outcome: hookworm infection; explanatory variables: demographic and education, socioeconomic, household characteristics and agricultural parameters. Medians were used to summarize the posterior distribution of the parameters of the Bayesian model
Explanatory variablesBivariate modelBayesian multivariate model†
OR95% CIP-value‡OR95% BCI
  1. BCI, Bayesian credible interval.

  2. †Spatially-explicit model with household and location-specific random effects.

  3. P-value based on likelihood ratio test.

  4. §Not included in the multivariate model.

Demography
 Sex
  Male1.00  1.00 
  Female0.700.48, 1.020.0630.640.38, 1.01
 Age (years)
  <101.00  1.00 
  10–144.302.13, 8.70 6.052.31, 13.41
  15–244.522.31, 8.83 7.302.96, 15.74
  25–394.382.11, 9.13 6.482.38, 14.73
  ≥403.371.74, 6.52<0.0015.602.31, 11.91
Socioeconomic status
 Poorest1.00  1.00 
 Very poor0.720.41, 1.23 0.630.27, 1.27
 Poor1.070.61, 1.90 0.620.21, 1.40
 Less poor0.580.31, 1.11 0.250.08, 0.59
 Least poor1.190.68, 2.080.1560.570.20, 1.28
Number of persons living in the household
 1–51.00  1.00 
 6–100.930.51, 1.69 1.390.54, 3.03
 11–200.940.51, 1.76 1.040.36, 2.42
 >201.730.89, 3.380.1062.610.86, 6.19
Toilet disposal
 Latrine used by only one household1.00  1.00 
 Shared latrine within neighbouring household1.060.69, 1.62 0.560.29, 0.98
 No toilet1.350.82, 2.230.4961.230.51, 2.53
Agricultural zone
 1 (mixed crops)1.00  1.00 
 3 (mixed crops)2.811.14, 6.95 3.070.81, 8.42
 4 (traditional smallholder plot)2.741.10, 6.82 6.691.65, 19.05
 5 (large rice perimeter)3.211.40, 7.36 6.541.97, 16.96
 6 (traditional smallholder plot)1.070.46, 2.50 1.550.38, 4.39
 7 (traditional smallholder plot plus large rice perimeter)1.770.71, 4.39<0.0011.790.38, 4.39
Private well in yard1.601.03, 2.470.0322.321.24, 4.05
Washing clothes in River Kô1.581.06, 2.340.0241.610.91, 2.67
Cultivated crop type
 Rain-fed rice0.530.33, 0.l860.0010.520.21, 1.07
 Irrigated rice0.840.56, 1.270.4051.870.89, 3.52
 Vegetables1.601.07, 2.400.0191.730.84, 3.19
 Rain-fed food crops§0.890.61, 1.320.570  
inline image (non-spatial variation)   0.230.04, 0.99
inline image (spatial variation)   0.540.20, 1.36
ρ (smoothing parameter)   0.560.09, 0.98
Deviance information criterion    596.8

Family members living in large households (>20 persons) were more likely to be infected with S. mansoni than members from households with ≤5 persons (OR = 2.56, 95% BCI = 1.07–5.23). Household members residing at least 750 m away from the Kô River were at a lower risk of a S. mansoni infection than those living in close proximity (OR = 0.61, 95% BCI = 0.37–0.93). Water contact related risk factors were bathing/swimming in the Kô River (OR = 3.16, 95% BCI = 1.42–6.25), fishing with a net (OR = 2.35, 95% BCI = 1.30–3.93) and use of water from irrigation wells and ponds (OR = 2.50, 95% BCI = 1.60–3.76).

A hookworm infection was age-related, but sex showed only borderline significance. Other risk factors for a hookworm infection included low socioeconomic status (OR = 0.25, 95% BCI = 0.08–0.59 for ‘less poor’), living in agricultural zone 4 (OR = 6.69, 95% BCI = 1.65–19.05) or zone 5 (OR = 6.54, 95% BCI = 1.97–16.96) and the use of domestic water from a well (OR = 2.32, 95% BCI = 1.24–4.05).

Spatial correlation of Schistosoma mansoni and hookworm infections

The results of the multivariate spatial random-effects logistic regression models (Tables 3 and 4) indicate that the spatial autocorrelation of S. mansoni and hookworm infections were negligible. In fact, the minimum distance at which the spatial correlation between houses dropped below 5% was 5.6 m in the case of S. mansoni and 5.4 m in the case of hookworm.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We investigated risk factors for S. mansoni and hookworm infections, including micro-spatial heterogeneity, in farming and non-farming households in a medium-sized town of Côte d'Ivoire. Both descriptive and Bayesian spatial statistics were employed. Our findings confirm that the distribution of S. mansoni and hookworm is highly focal also in this urban setting with irrigated rice cultivation zones (Fenwick et al. 2006; Raso et al. 2006a,b). Moreover, infection prevalences of both parasites in farming communities were related to agricultural activities.

Four aspects pertaining to urban agriculture are particularly noteworthy. First, highest levels of household-aggregated S. mansoni infections were found in a zone of smallholder irrigated rice plots (zone 6). Second, agricultural risk factors for a S. mansoni infection included the use of water from irrigation wells and ponds. These two observations are underscored by the discovery of B. pfeifferi from irrigation wells, canals and ponds in agricultural zones 1, 4, 5, 6 and 7 and human water contact sites in the Kô River in October 2004 and April 2005. Laboratory investigations revealed that the snails were shedding S. mansoni cercariae (sites where B. pfeifferi were collected are marked on Figure 1). A previous study already established a link between S. mansoni infection and irrigated rice cropping systems in Côte d'Ivoire (Yapi et al. 2005). Household-aggregated S. mansoni infections related to agricultural water contact have also been reported from Brazil (Bethony et al. 2004). The observed differences in S. mansoni infections among males and females in farming households are probably associated with gender-specific water contact patterns, already documented from irrigated river valleys in Morocco (Watts et al. 1998).

Third, households located in agricultural zones of smallholder irrigated rice cultivation and a large irrigated rice perimeter (zones 4 and 5) were at higher risk of hookworm infection. Fourth, high S. mansoni-hookworm co-infections were clustered in a zone of a large rice perimeter (zone 5). Zones 3, 4 and 5 are located east of the Kô River and the clustering of hookworm and S. mansoni-hookworm co-infections in these zones cannot be explained by spatial disparities of the socioeconomic status of farming communities or by the cultivated crop types. While the eastern part of the town of Man is a plain, the western part is hilly, which probably results in marked differences in soil types. Variations in clay content and soil structure can play an important role in providing suitable habitats for hookworm eggs and larvae, as recently observed in a study in South Africa (Saathoff et al. 2005). Spatial clustering of hookworm infections within a few hundred meters has also been observed in a village and small settlement some 20 km east of the town of Man (Utzinger et al. 2003) and in an urban setting of Brazil (Brooker et al. 2006b).

A methodological shortcoming of our study was that only one stool specimen was collected from each participant. It is widely acknowledged that there is important day-to-day and intra-stool variation of S. mansoni and hookworm egg output (de Vlas & Gryseels 1992; Engels et al. 1997; Utzinger et al. 2001; Booth et al. 2003). To enhance the diagnostic sensitivity, we prepared two Kato-Katz thick smears from each stool specimen, and employed an ether-concentration method and considered the combined diagnostic results in our final analyses.

Close proximity to the Kô River was identified as a significant risk factor for a S. mansoni infection. Domestic, agricultural, recreational and other activities were carried out at the riverside by people living close-by. Thus, households located a few hundred metres from the Kô River are at a higher risk of S. mansoni than households situated further away as frequency of water contact depends on distance to infected sources of water (Brooker et al. 2001). Recent studies focussing at micro-geographical units of analysis, carried out in Kenya, showed that prevalence and intensity of S. mansoni and soil-transmitted helminth infections were distance-dependent from lakes and permanent water bodies (Handzel et al. 2003; Booth et al. 2004; Clennon et al. 2004).

The significantly higher hookworm prevalence in farming households than in non-farming households, particularly among males, might be associated with the work-related presence in green areas that provide at the same time hiding places for open defecation. Although wearing of footwear was not systematically assessed in our study, we observed that farmers on agricultural plots rarely wore boots, closed shoes or sandals. Wearing shoes was reported as ‘uncomfortable’ when working in the field because of the wet soils and penetrating water. Footwear appeared as a protective factor against soil-transmitted helminth infections in tea-growing communities in India (Traub et al. 2004), but had no effect on hookworm infections in another study from Mali (Behnke et al. 2000).

The prevalence of hookworm was considerably lower in the current population sample when compared with a household-based study conducted in a village 30 km east of Man in 2002 (Raso et al. 2004). Other school-based and community-wide investigations revealed no significant rural-urban differences of hookworm prevalence (Sinuon et al. 2003; Brooker et al. 2006a), which is in contrast to our findings. However, differences in hookworm patterns were found between rural and peri-urban agricultural settings, as well as in non-agricultural urban settings of Vietnam (van der Hoek et al. 2003), with highest prevalence occurring in peri-urban settings with intensive vegetable farming. These observations, together with the fact that the socioeconomic status did not differ between farming and non-farming communities in our study area, support agriculture-related hookworm profiles in farming communities.

Risk factors for a S. mansoni infection were governed by living conditions, including high numbers of household members and sharing a latrine with neighbours. Household crowding has been reported to be associated with S. mansoni (Gazzinelli et al. 2006) and with hookworm infection (Curtale et al. 1998; Olsen et al. 2001; Traub et al. 2004). We assume that ‘shared latrines’ are part of domestic activities associated with water usage and storage that are not necessarily limited to one household (Cairncross et al. 1996), and resulting in shared infective sites (Bethony et al. 2001). The use of water from domestic wells was another risk factor for hookworm infection. Moist shaded soil around wells provides suitable living conditions for infective hookworm larvae. Differences in defecation habits by adults and children result in distinct hookworm transmission patterns (Chan et al. 1997). In an investigation from Thailand, soil samples taken in family yards around foot-washing zones, under trees and near latrines were highly contaminated with soil-transmitted helminths (Chongsuvivatwong et al. 1999). In the present study, low education attainment was a risk factor for S. mansoni, and low socioeconomic status for hookworm infection. Note that people's socioeconomic status was assessed by a household-based asset approach. In our previous work carried out in surrounding villages in the Man area, we used the same household-based asset approach which proved to be useful to understand schoolchildren's infection status with S. mansoni, hookworm, co-infection and multiparasitism (Raso et al. 2005, 2006b).

The extremely low spatial correlation of both parasites investigated (i.e. 5–6 m) might be explained by shared transmission sites and common exposure among neighbouring households. With regard to S. mansoni, contact with irrigation wells and ponds was identified as a risk factor. Given the close proximity of irrigation wells and ponds in rice paddies and market gardens to farmers’ houses, shared transmission is likely to occur. With regard to hookworm infection, it is conceivable that shared transmission sites were private wells in house yards. A comparable study carried out in a small urban area in Brazil, also using Bayesian inference, lends support to our suggestions of shared transmission sites at the peri-domestic area for hookworm infection and proximity to infested water bodies for S. mansoni infection (Brooker et al. 2006b). A study carried out in a rural Brazilian setting found spatial correlations over a larger distance for both parasites (Brooker et al. 2006b). We speculate that in densely populated urban settings with small-scale land use patterns, transmission sites are more clustered when compared with rural settings where more open space is available. Findings from a study at a medium spatial scale (56 rural schools scattered over a 40 × 60 km surface) revealed a quasi-absent spatial correlation (∼2 km) of hookworm infections between villages (Raso et al. 2006a). Another investigation at a relatively large scale (i.e. 530 × 670 km) from Tanzania found a spatial correlation of 3.4 km for S. mansoni infections at village locations along lake shores and perennial water bodies (Clements et al. 2006). The prediction of high-risk areas for targeting control interventions is usually carried out at large spatial scales, and hence takes into account only the aggregated village rather than the household level. This issue prevents direct comparison with small-scale studies carried out at the household level. In future investigations at large-scale, it would be interesting to explore spatial effects at the household level.

Prediction maps based on the Bayesian regression models were tested for the prevalence of S. mansoni infection in Man (data not shown). Emphasis was placed on girls, aged 10–14 years, as we assumed that this group was minimally involved in agricultural activities and thus, contact with infested transmission sites was probably restricted to the Kô River. Lower infection prevalences were predicted for zones ≥700 m away from the Kô River. A limitation of our prediction is that it was based on a single environmental explanatory variable, i.e. distance to the Kô River. We suggest that human behaviour-related factors are equally or even more important to explain S. mansoni infections than environmental factors at a small scale. This outcome needs to be taken into consideration in further spatial investigations that focus on the micro-geographical distribution of S. mansoni and employ Bayesian spatial statistics.

In conclusion, the present study emphasises micro-spatial heterogeneity of S. mansoni and hookworm infections in farming communities of a typical medium-sized town of Côte d'Ivoire. Contextual determinants include agricultural, behavioural, demographic, environmental and socioeconomic factors. Our spatial Bayesian modelling approach at this scale was limited to identify risk factors, which can be explained by individual- and household-based behavioural factors that are more important at this scale than environmental parameters. Future studies should investigate whether Bayesian spatial statistics can explain clustering of human parasitic infections at such small scales in different parts of the world. As chemotherapy-based control programmes of schistosomiasis and soil-transmitted helminthiasis target large communities rather than specific households, more work remains to be carried out to further our understanding of infection patterns among special high-risk groups, such as farming communities. Chemotherapy should go hand-in-hand with sound health education, and active participation of urban farmers for prevention and control of helminth infections.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank all families for their participation. We acknowledge political and religious leaders, school directors and community youth associations for placing rooms and other infrastructures at our disposal during the cross-sectional surveys. We are grateful to D. Doua and his team (S. Tokpa, M. Kpan, C. Gueu Sadia, R. Dion, P. Blé Gosamé, A. Thian Yohan and S. Sadia) of the non-governmental organization ODAFEM in Man for their commitment in this study. We thank M. Koné from the Université de Bouaké and E. Gbede Becket for help with the socioeconomic survey. We are grateful to the laboratory technicians (A. Allangba, A. Fondjo and B. Sosthène) and the medical field staff of Man for their excellent work in the field and the laboratory. We thank M. Mabaso for statistical support. Comments from two anonymous referees helped in further improving this manuscript. This investigation received financial support from the National Centre of Competence in Research (NCCR) North-South programme entitled ‘Research partnerships for mitigating syndromes of global change’, individual project no. 4 (IP4), entitled ‘Health and well-being’, the Swiss Development Cooperation (SDC) for support granted to the Centre Suisse de Recherches Scientifiques via a project entitled ‘Contribution to the process of national reconciliation in Côte d'Ivoire’, and the Swiss National Science Foundation (SNF) through a research project to P. Vounatsou and L. Gosoniu (project no. 3252B0-102136), a fellowship to G. Raso (project no. PBBSB-109011) and a ‘SNF-Förderungsprofessur’ to J. Utzinger (project no. PP00B–102883).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Amoah P, Drechsel P, Abaidoo RC & Ntow WJ (2006) Pesticide and pathogen contamination of vegetables in Ghana's urban markets. Archives of Environmental Contamination and Toxicology 50, 16.
  • Behnke JM, De Clercq D, Sacko M, Gilbert FS, Ouattara DB & Vercruysse J (2000) The epidemiology of human hookworm infections in the southern region of Mali. Tropical Medicine and International Health 5, 343354.
  • Bethony J, Williams JT, Kloos H et al. (2001) Exposure to Schistosoma mansoni infection in a rural area in Brazil. II: household risk factors. Tropical Medicine and International Health 6, 136145.
  • Bethony J, Williams JT, Brooker S et al. (2004) Exposure to Schistosoma mansoni infection in a rural area in Brazil. Part III: household aggregation of water-contact behaviour. Tropical Medicine and International Health 9, 381389.
  • Bethony J, Brooker S, Albonico M et al. (2006) Soil-transmitted helminth infections: ascariasis, trichuriasis, and hookworm. Lancet 367, 15211532.
  • Binns JA, Maconachie RA & Tank AI (2003) Water, land and health in urban and peri-urban food production: the case of Kano, Nigeria. Land Degradation and Development 14, 431444.
  • Booth M, Vounatsou P, N'Goran EK, Tanner M & Utzinger J (2003) The influence of sampling effort and the performance of the Kato-Katz technique in diagnosing Schistosoma mansoni and hookworm co-infections in rural Côte d'Ivoire. Parasitology 127, 525531.
  • Booth M, Vennervald BJ, Kenty L et al. (2004) Micro-geographical variation in exposure to Schistosoma mansoni and malaria, and exacerbation of splenomegaly in Kenyan school-aged children. BMC Infectious Diseases 4, 13.
  • Brooker S, Miguel EA, Waswa P et al. (2001) The potential of rapid screening methods for Schistosoma mansoni in western Kenya. Annals of Tropical Medicine and Parasitology 95, 343351.
  • Brooker S, Bethony J & Hotez PJ (2004) Human hookworm infection in the 21st century. Advances in Parasitology 58, 197288.
  • Brooker S, Clements ACA & Bundy DAP (2006a) Global epidemiology, ecology and control of soil-transmitted helminth infections. Advances in Parasitology 62, 221261.
  • Brooker S, Alexander N, Geiger S et al. (2006b) Contrasting patterns in the small-scale heterogeneity of human helminth infections in urban and rural environments in Brazil. International Journal for Parasitology 36, 11431151.
  • Bryld E (2003) Potentials, problems, and policy implications for urban agriculture in developing countries. Agriculture and Human Values 20, 7986.
  • Cairncross S, Blumenthal U, Kolsky P, Moraes L & Tayeh A (1996) The public and domestic domains in the transmission of disease. Tropical Medicine and International Health 1, 2734.
  • Chan MS, Bradley M & Bundy DAP (1997) Transmission patterns and the epidemiology of hookworm infection. International Journal of Epidemiology 26, 13921400.
  • Chongsuvivatwong V, Uga S & Nagnaen W (1999) Soil contamination and infections by soil-transmitted helminths in an endemic village in southern Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 30, 6467.
  • Clements ACA, Lwambo NJ, Blair L et al. (2006) Bayesian spatial analysis and disease mapping: tools to enhance planning and implementation of a schistosomiasis control programme in Tanzania. Tropical Medicine and International Health 11, 490503.
  • Clennon JA, King CH, Muchiri EM et al. (2004) Spatial patterns of urinary schistosomiasis infection in a highly endemic area of coastal Kenya. American Journal of Tropical Medicine and Hygiene 70, 443448.
  • Curtale F, Shamy MY, Zaki A, Abdel-Fattah M & Rocchi G (1998) Different patterns of intestinal helminth infection among young workers in urban and rural areas of Alexandria Governorate, Egypt. Parassitologia 40, 251254.
  • Engels D, Sinzinkayo E, De Vlas SJ & Gryseels B (1997) Intraspecimen fecal egg count variation in Schistosoma mansoni infection. American Journal of Tropical Medicine and Hygiene 57, 571577.
  • Ensink JH, Van Der Hoek W, Mukhtar M, Tahir Z & Amerasinghe FP (2005) High risk of hookworm infection among wastewater farmers in Pakistan. Transactions of the Royal Society of Tropical Medicine and Hygiene 99, 809818.
  • Fenwick A, Keiser J & Utzinger J (2006) Epidemiology, burden and control of schistosomiasis with particular consideration to past and current treatment trends. Drugs of the Future 31, 413425.
  • Filmer D & Pritchett LH (2001) Estimating wealth effects without expenditure data – or tears: an application to educational enrollments in states of India. Demography 38, 115132.
  • Firmo JOA, Lima e Costa MF, Guerra HL & Rocha RS (1996) Urban schistosomiasis: morbidity, sociodemographic characteristics and water contact patterns. Predictive of infection. International Journal of Epidemiology 25, 12921300.
  • Fournet F, N'Guessan NA & Cadot E (2004) Gestion de l'espace et schistosomose urinaire à Daloa (Côte d'Ivoire). Bulletin de la Société de la Pathologie Exotique 97, 3336.
  • Gazzinelli A, Velasquez-Melendez G, Crawford SB, LoVerde PT, Correa-Oliveira R & Kloos H (2006) Socioeconomic determinants of schistosomiasis in a poor rural area in Brazil. Acta Tropica 99, 260271.
  • Gelfand AE & Smith AFM (1990) Sampling-based approaches to calculating marginal densities. Journal of the American Statistical Association 85, 398409.
  • Gryseels B, Polman K, Clerinx J & Kestens L (2006) Human schistosomiasis. Lancet 368, 11061118.
  • Gwatkin D, Rustein S, Johnson C, Pande R & Wagstaff A (2000) Socio-economic differences in health, nutrition, and population in the Côte d'Ivoire. In HNP/Poverty Thematic Group of the World Bank, Washington, DC.
  • Handzel T, Karanja DMS, Addiss DG et al. (2003) Geographic distribution of schistosomiasis and soil-transmitted helminths in western Kenya: implications for anthelminthic mass treatment. American Journal of Tropical Medicine and Hygiene 69, 318323.
  • Van Der Hoek W, De NV, Konradsen F et al. (2003) Current status of soil-transmitted helminths in Vietnam. Southeast Asian Journal of Tropical Medicine and Public Health 34 (Suppl. 1), 111.
  • Hotez PJ, Molyneux DH, Fenwick A, Ottesen E, Ehrlich Sachs S & Sachs JD (2006) Incorporating a rapid-impact package for neglected tropical diseases with programs for HIV/AIDS, tuberculosis, and malaria. PLoS Medicine 3, 102.
  • Huang Y & Manderson L (1992) Schistosomiasis and the social patterning of infection. Acta Tropica 51, 175194.
  • Katz N, Chaves A & Pellegrino J (1972) A simple device for quantitative stool thick-smear technique in schistosomiasis mansoni. Revista do Instituto de Medicina Tropical de São Paulo 14, 397400.
  • Lemeshow S & Robinson D (1985) Surveys to measure programme coverage and impact: a review of the methodology used by the expanded programme on immunization. World Health Statistics Quarterly 38, 6575.
  • Lopez AD, Mathers CD, Ezzati M, Jamison DT & Murray CJL (2006) Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet 367, 17471757.
  • Marti HP & Escher E (1990) SAF – Eine alternative Fixierlösung für parasitologische Stuhluntersuchungen. Schweizerische Medizinische Wochenschrift 120, 14731476.
  • Matthys B, N'Goran EK, Koné A et al. (2006a) Urban agricultural land use and characterization of mosquito larval habitats in a medium-sized town of Côte d'Ivoire. Journal of Vector Ecology 31, 319333.
  • Matthys B, Vounatsou P, Raso G et al. (2006b) Urban farming and malaria risk factors in a medium-sized town in Côte d'Ivoire. American Journal of Tropical Medicine and Hygiene 75, 12231231.
  • Olsen A, Samuelsen H & Onyango-Ouma W (2001) A study of risk factors for intestinal helminth infections using epidemiological and anthropological approaches. Journal of Biosocial Science 33, 569584.
  • Raso G, Luginbühl A, Adjoua CA et al. (2004) Multiple parasite infections and their relationship to self-reported morbidity in a community of rural Côte d'Ivoire. International Journal of Epidemiology 33, 10921102.
  • Raso G, Utzinger J, Silue KD et al. (2005) Disparities in parasitic infections, perceived ill health and access to health care among poorer and less poor schoolchildren of rural Côte d'Ivoire. Tropical Medicine and International Health 10, 4257.
  • Raso G, Vounatsou P, Gosoniu L, Tanner M, N'Goran EK & Utzinger J (2006a) Risk factors and spatial patterns of hookworm infection among schoolchildren in a rural area of western Côte d'Ivoire. International Journal for Parasitology 36, 201210.
  • Raso G, Vounatsou P, Singer BH, N'Goran EK, Tanner M & Utzinger J (2006b) An integrated approach for risk profiling and spatial prediction of Schistosoma mansoni-hookworm coinfection. Proceedings of the National Academy of Sciences of the United States of America 103, 69346939.
  • Saathoff E, Olsen A, Sharp B, Kvalsvig JD, Appleton CC & Kleinschmidt I (2005) Ecologic covariates of hookworm infection and reinfection in rural Kwazulu-natal/South Africa: a geographic information system-based study. American Journal of Tropical Medicine and Hygiene 72, 384391.
  • De Silva NR, Brooker S, Hotez PJ, Montresor A, Engels D & Savioli L (2003) Soil-transmitted helminth infections: updating the global picture. Trends in Parasitology 19, 547551.
  • Sinuon M, Anantaphruti MT & Socheat D (2003) Intestinal helminthic infections in schoolchildren in Cambodia. Southeast Asian Journal of Tropical Medicine and Public Health 34, 254258.
  • Spiegelhalter DJ, Best N, Carlin BP & Van der Linde A (2002) Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society, Series B 64, 583639.
  • Steinmann P, Keiser J, Bos R, Tanner M & Utzinger J (2006) Schistosomiasis and water resources development: systematic review, meta-analysis, and estimates of people at risk. Lancet Infectious Diseases 6, 411425.
  • Traub RJ, Robertson ID, Irwin P, Mencke N & Thompson RCA (2004) The prevalence, intensities and risk factors associated with geohelminth infection in tea-growing communities of Assam, India. Tropical Medicine and International Health 9, 688701.
  • Utzinger J & Keiser J (2006) Urbanization and tropical health – then and now. Annals of Tropical Medicine and Parasitology 100, 517533.
  • Utzinger J, Booth M, N'Goran EK, Müller I, Tanner M & Lengeler C (2001) Relative contribution of day-to-day and intra-specimen variation in faecal egg counts of Schistosoma mansoni before and after treatment with praziquantel. Parasitology 122, 537544.
  • Utzinger J, Müller I, Vounatsou P, Singer BH, N'Goran EK & Tanner M (2003) Random spatial distribution of Schistosoma mansoni and hookworm infections among school children within a single village. Journal of Parasitology 89, 686692.
  • De Vlas SJ & Gryseels B (1992) Underestimation of Schistosoma mansoni prevalences. Parasitology Today 8, 274277.
  • Watts S, Khallaayoune K, Bensefia R, Laamrani H & Gryseels B (1998) The study of human behavior and schistosomiasis transmission in an irrigated area in Morocco. Social Science and Medicine 46, 755765.
  • WHO (2002) Prevention and control of schistosomiasis and soil-transmitted helminthiasis: report of a WHO expert committee. WHO Technical Report Series 912, 157.
  • Ximenes R, Southgate B, Smith PG & Guimaraes Neto L (2003) Socioeconomic determinants of schistosomiasis in an urban area in the northeast of Brazil. Revista Panamericana de Salud Publica 14, 409421.
  • Yapi YG, Briët OJT, Diabaté S et al. (2005) Rice irrigation and schistosomiasis in savannah and forest areas of Cote d'Ivoire. Acta Tropica 93, 201211.