Mapping schistosomiasis risk landscapes and implications for disease control: A case study for low endemic areas in the Middle Paranapanema river basin, São Paulo, Brazil
Vivian Alessandra Ferreira da Silva,Milton Kampel,Rafael Silva dos Anjos,Raquel Gardini Sanches Palasio,Maria Isabel Sobral Escada,Roseli Tuan,Alyson Singleton,Caroline Kate Glidden,Andrew Chamberlin,Giulio Alessandro De Leo,Adriano Pinter dos Santos,Antônio Miguel Vieira Monteiro
DOI: https://doi.org/10.1371/journal.pntd.0012582
2024-11-05
PLoS Neglected Tropical Diseases
Abstract:Schistosomiasis, a chronic parasitic disease, remains a public health issue in tropical and subtropical regions, especially in low and moderate-income countries lacking assured access to safe water and proper sanitation. A national prevalence survey carried out by the Brazilian Ministry of Health from 2011 to 2015 found a decrease in human infection rates to 1%, with 19 out of 26 states still classified as endemic areas. There is a risk of schistosomiasis reemerging as a public health concern in low-endemic regions. This study proposes an integrated landscape-based approach to aid surveillance and control strategies for schistosomiasis in low-endemic areas. In the Middle Paranapanema river basin, specific landscapes linked to schistosomiasis were identified using a comprehensive methodology. This approach merged remote sensing, environmental, socioeconomic, epidemiological, and malacological data. A team of experts identified ten distinct landscape categories associated with varying levels of schistosomiasis transmission potential. These categories were used to train a supervised classification machine learning algorithm, resulting in a 92.5% overall accuracy and a 6.5% classification error. Evaluation revealed that 74.6% of collected snails from water collections in five key municipalities within the basin belonged to landscape types with higher potential for S. mansoni infection. Landscape connectivity metrics were also analysed. This study highlights the role of integrated landscape-based analyses in informing strategies for eliminating schistosomiasis. The methodology has produced new schistosomiasis risk maps covering the entire basin. The region's low endemicity can be partly explained by the limited connectivity among grouped landscape-units more prone to triggering schistosomiasis transmission. Nevertheless, changes in social, economic, and environmental landscapes, especially those linked to the rising pace of incomplete urbanization processes in the region, have the potential to increase risk of schistosomiasis transmission. This study will help target interventions to bring the region closer to schistosomiasis elimination. Schistosomiasis is a Neglected Tropical Disease whose transmission in Brazil is related to human contact with water contaminated by the trematode parasite Schistosoma mansoni . The national prevalence survey from 2011–2015 revealed a decline in schistosomiasis positivity rates to 1%, marking areas in the Middle Paranapanema river basin in SP-Brazil (MP) as low endemic. However, control programs face additional challenges due to the new dynamics connected with the social, environmental, and economic developments fueled by an urban-industrial urbanization model in the MP region. To address this, our study proposes a landscape-based methodological approach that categorizes and assesses schistosomiasis transmission potential in different regional landscape patterns. By combining remote sensing, environmental, socioeconomic, epidemiological, and malacological data, a multidisciplinary team identified ten distinct landscape-unit types associated with varying levels of schistosomiasis transmission potential. Using a decision-tree based machine learning classification method, we mapped schistosomiasis risk across the basin at a landscape scale. The identification of landscape features associated with schistosomiasis transmission risk will help to fine tune suitable surveillance and control strategies at both local and regional levels. Our findings indicate that landscape-units with higher transmission potentials are less interconnected across the basin. Low connectivity at the landscape scale, in part, explains the MP low endemic areas, contributing to the low endemicity in the Middle Paranapanema region. This approach offers promise for local and regional schistosomiasis management efforts aimed at eradicating the disease.
tropical medicine,parasitology