The effect of irrigated urban agriculture on malaria, schistosomiasis and soil-transmitted helminthiasis in different settings of Côte d'Ivoire

Malaria is responsible for more than one million deaths every year, mainly children under the age of five years living in sub-Saharan Africa. At least one billion people harbor one or several of the three main soil-transmitted helminths, namely Ascaris lumbricoides, hookworms and Trichuris trichiura, and about 207 million people are infected with schistosomes. An estimated 70,000 people die each year from amoebiasis, caused by Entamoeba histolytica. Giardiasis, caused by Giardia duodenalis, is responsible for 2.8 million annual infections. Poor rural and urban communities in developing countries bear the highest burden of the above-mentioned diseases. Their causes are multifactorial including lack of access to clean water, improved sanitation and health services, as well as inadequate treatment, protection and prevention. The highest rates of urbanization currently occur in the less developed regions of Africa and Asia, and it is predicted that the majority of the population will be living in small and medium urban centers in the near future. The maintenance of traditional livelihoods, including agriculture, is a typical feature in urban settings across Africa, especially in small towns where population densities and land pressure are lower than in big cities. Urban agriculture contributes to food security and livelihood opportunities for poor urban dwellers. However, the adaptation of disease vectors and intermediate hosts to urban ecosystems has been observed, which might further enhance the negative effects associated with persistent rural lifestyles. For example, the creation of malaria vector breeding sites and contact with contaminated water and soil in areas of irrigated agriculture may increase the transmission of vector-borne, water-related and soil-transmitted parasitic diseases. This PhD focused on the interface of agriculture and human health in two different urban settings of Côte d’Ivoire. The overarching goal was to contribute to a better understanding of the effects of irrigated urban agriculture on the transmission of malaria, schistosomiasis, soil-transmitted helminthiasis and intestinal protozoan infections. The research entailed a considerable amount of fieldwork, carried out between April 2004 and July 2005. In a first step, six zones of irrigated agricultural land use were identified both in Abidjan, the economic capital of Côte d’Ivoire (3.3 million inhabitants), and in the mediumsized town of Man (115,000 inhabitants) in the western part of the country. Next, two standardized mosquito breeding site assessments were conducted in these agricultural zones in the rainy season (September 2004) and in the dry season (February 2005). In each urban setting, a minimum of 120 farming households and additionally 30 non-farming households were randomly selected. Geographic coordinates of houses and the main agricultural plots were recorded. Name, age and sex of all household members were registered. In October
2004, interviews on agricultural land use, farming practices and water storage were
conducted with the heads of the farming households. In a cross-sectional survey done in
May/June 2005, questionnaires were administered to all households to assess the socioeconomic
status, sanitary facilities and common water contact patterns. From each study
participant, a finger prick blood sample and a stool sample were collected. Thick and thin
blood films were stained with Giemsa and examined for Plasmodium spp. under a light
microscopy. The stool samples were prepared based on the Kato-Katz technique and eggs of
Schistosoma mansoni, A. lumbricoides, hookworm and T. trichiura were recorded separately.
Cysts or trophozoites of intestinal protozoa, including Entamoeba histolytica/E. dispar and G.
duodenalis were identified using an ether-formalin concentration method. Risk factors were
identified by fitting multivariate non-random and random effects Bayesian regression models
integrating spatial correlation of infection.
Agricultural land use in Abidjan was characterized by market gardens on lagoon shores
and high-yield vegetable production. The vegetable production areas in Abidjan developed as
a consequence of land access difficulties in unexploited public and private areas near poor
settlements, mainly at lagoon shores. In Man, farming was family- and subsistence-based.
Predominant agricultural activities were traditional irrigated rice farming and vegetable
production in lowlands and along the Kô River.
The typical demographic and socio-economic profile of a vegetable producer in
Abidjan is that he is an immigrant from Burkina Faso, illiterate and lacking a professional
agricultural training. Malaria and intestinal parasitic infections were most prevalent among
the final study cohort of 370 farmers and family members from 121 farming households. We
found overall prevalences of T. trichiura, P. falciparum and hookworm of 42.6%, 24.3% and
17.3%, respectively. The parasites were heterogeneously distributed between the six market
garden zones. Prior health issues were suppressed by the farmers with the daily livelihood
struggle and reported symptoms due to intestinal parasitic infections were of less importance
when compared to malaria. In-depth focus group discussions revealed that the working
environment was discerned as unhealthy, and waste dumps and human defecation grounds
were perceived as main health risks. Farmer communities and their network are indicative for
the degree of social coherence and stability in a vegetable production area because these
zones are characterized by highly dynamic land use patterns.
In the town of Man, Anopheles larvae were present in 50.7% and 42.4% of 369 and 589
examined potential mosquito breeding sites in the rainy and in the dry season, respectively.
The most productive habitats were man-made, i.e. agricultural trenches, irrigation wells, and
flooded and recently transplanted rice plots. The overall prevalence of P. falciparum
infections in the final study cohort of 574 individuals from 112 farming households was
32.1%. Risk factors for P. falciparum in children <15 years of age included living in specific
agricultural zones (i.e. traditional irrigated rice plots, mixed crops and a large rice perimeter),
proximity to permanent man-made ponds and fish ponds, periodic stays overnight in farm
huts and low socio-economic status.
The final study cohort for S. mansoni and soil-transmitted helminthiasis comprised 586
individuals from 113 farming households. The overall prevalences of S. mansoni, hookworm,
E. histolytica/E. dispar and G. duodenalis were 51.4%, 24.7%, 20.2% and 6.3%, respectively.
Members from farming households harbored significantly more often an infection with
E. histolytica/E. dispar and G. duodenalis when compared to non-farming households.
Predictors for an S. mansoni infection included close proximity to the Kô River, contact with
irrigation wells and ponds on the agricultural plots and low educational attainment. Risk
factors for hookworm infection comprised living in agricultural zones of traditional
smallholder irrigated rice plots and a large rice perimeter, using water from domestic wells
and low socio-economic status. Infection prevalences of P. falciparum, S. mansoni and
hookworm were spatially highly heterogeneous between the agricultural zones and highest
infections occurred in a zone of a large rice perimeter. P. falciparum infection intensity and
hookworm infection prevalence were best explained by spatial random effect models. Spatial
correlation between farmers’ houses was not significant.
The findings of the present work illustrated a clear picture of the interconnections
between specific irrigated agricultural land use and agricultural activities, and malaria and
intestinal parasitic infections in different urban settings of Côte d’Ivoire. The outcomes lead
to an enhanced understanding of their epidemiology in local agro-ecological urban settings
and related contextual determinants (i.e. agricultural, behavioural, demographic, socioeconomic
and environmental factors) and allows the design of readily adapted prevention and
control interventions (e.g. tangible vector control strategies and prevention measures for
helminth infections) which actively involve farming communities in the subsequent
implementation and control management.