Overview of snakebite in Brazil: Possible drivers and a tool for risk mapping

Abstract

Snakebite envenoming affects close to 2.7 million people globally every year. In Brazil, snakebites are reported to the Ministry of Health surveillance system and cases receive antivenom free of charge. There is an urgent need to identify higher risk areas for antivenom distribution, and to develop prevention activities. The objective of this study is to provide an overview of the epidemiological situation of snakebite envenoming in Brazil and explore possible drivers; as well as to create a flowchart tool to support decision-makers identify higher risk areas. An ecological-type study was carried out using data by municipality (2013–2017). Study parts: 1) Create a geocoded database and perform a descriptive and cluster analysis; 2) Statistical analysis to measure the association of snakebite and possible environmental and socioeconomic drivers; 3) Develop a flowchart to support decision-makers and the application of this tool in one state (Rio Grande do Sul) as an example. An average of 27,120 snakebite cases per year were reported at the country level. Clusters of municipalities with high numbers of snakebites are mostly found in the Amazon Legal Region. The negative binomial regression model showed association with the snakebite case count: the type of major habitat, tropical or non-tropical; temperature; percentage of urbanization; precipitation; elevation; GDP per capita; a weaker relation with forest loss; and with venomous snake richness. The state where the instrument was applied reported 4,227 snakebites in the period. Most municipalities were considered as medium risk and 56/496 as high risk according to the tool created. Snakebite cases are distributed across the entire country with the highest concentration in the Legal Amazon Region. This creates a complex situation both for better understanding of the association of environmental and socioeconomic factors with snakebites and for the distribution and maintenance of antivenom to remote areas. Research into types of antivenom with a longer shelf life without the need for refrigeration is needed.

Addressing the global snakebite crisis with geo-spatial analyses – Recent advances and future direction

“…Relationships between spatial snakebite variation and human population density are more complex: usually snakebites increase with human population density in rural areas but drop off at higher densities associated with urbanization ( Chippaux, 2017 ; Ediriweera et al, 2016 ). As expected, snakebite incidence also correlates with measures of presence, activity, abundance, or diversity of snakes ( Bravo-Vega et al, 2019 ; Goldstein et al, 2021 ; Hansson et al, 2013 ; León-Núñez et al, 2020 ; Schneider et al, 2021 ; Suraweera et al, 2020 ; Yañez-Arenas et al, 2014 , 2016 ) or with variables that affect snake activity. Often snakebite incidence increases during certain seasons when snakes and farmers are both more active such as in rainy or harvest seasons ( Chippaux, 2017 ; Ediriweera et al, 2018 ; Goldstein et al, 2021 ; Hansson et al, 2010 ; Mohapatra et al, 2011 ; Molesworth et al, 2003 ; Patiño-Barbosa et al, 2019 ; Suraweera et al, 2020 ), during flooding events ( Ochoa et al, 2020 ), or at higher temperatures, lower altitudes, and higher precipitation ( Angarita-Gerlein et al, 2017 ; Chaves et al, 2015 ; Chippaux, 2017 ; Ediriweera et al, 2018 ; Ediriweera et al, 2016 ; Goldstein et al, 2021 ; Hansson et al, 2013 ; Schneider et al, 2021 ; Suraweera et al, 2020 , Table 2 ).…”

“…Potential drivers of spatial snakebite variation at intermediate scales have been quantified to some extent, using anything from simple statistics such as t-tests ( Chippaux, 2017 ; León-Núñez et al, 2020 ) to more elaborate statistical models such as generalized additive models (GAM; Ediriweera et al, 2016 ), geostatistical binomial logistic models ( Ediriweera et al, 2018 ), spatial Poisson models ( Suraweera et al, 2020 ) or bottom-up agent-based models ( Goldstein et al, 2021 ); Table 2 ). Again, hotspots tend to occur in rural, agricultural, and poor areas ( Chaves et al, 2015 ; Ediriweera et al, 2016 ; Hansson et al, 2010 , 2013 ; Leynaud and Reati, 2009 ; Schneider et al, 2021 ; Suraweera et al, 2020 ), and more bites occur in young to middle aged males or in regions with a higher male population percentage ( Chippaux, 2017 ; Ediriweera et al, 2016 ; Hansson et al, 2010 ; León-Núñez et al, 2020 ; Mohapatra et al, 2011 ; Suraweera et al, 2020 ). Relationships between spatial snakebite variation and human population density are more complex: usually snakebites increase with human population density in rural areas but drop off at higher densities associated with urbanization ( Chippaux, 2017 ; Ediriweera et al, 2016 ).…”

“…Certain crop expansion and changes in farming practices could lead to increased snakebite prevalence, while mechanization of farming practices could, conversely, reduce exposure to snakes. Human-wildlife conflict also often increases with deforestation ( Lamarque et al, 2009 ; Schneider et al, 2021 ) and in border-country to remnant forests and protected areas ( Hansson et al, 2013 ; Sharma et al, 2020 ) because animals are forced to leave their natural habitat and use anthropogenic landscapes. For species that are incapable of using anthropogenic landscapes, this might lead to decreases in suitable habitat and short term increases in human encounters as they search for new suitable habitat ( Acharya et al, 2017 ; Distefano, 2005 ).…”

Snakebites in Rural Areas of Brazil by Race: Indigenous the Most Exposed Group

“…To save lives, a large number of ampoules of antivenom needs to be distributed regularly across the entire country. The Ministry of Health in Brazil has been distributing around 230,000 ampoules of antivenom to approximately 2000 hospitals free of charge annually [16], which is likely one of the reasons for the low mortality rates associated with snakebites in Brazil found in this and prior studies [9,16,[28][29][30]41]. Brazil's case-fatality rate for snakebites of around 0.4%, found in our study and the previous study by Bochner [9], is particularly low if compared with studies in other countries, such as the reported case-fatality rate of 3% in a recent district-level study in Cameroon [42], and an estimated case-fatality rate of 3.2% for in-hospital cases of snakebite envenoming in India [43].…”

“…The major habitat type in Brazil is tropical and subtropical moist broadleaf forests (TSMBF) [35]. This habitat type includes the Legal Amazon Region and some areas of the Atlantic Forest in the coastal area, where close to 70% of the snakebites occurred [16].…”