Como o desmatamento influencia a ocorrência de leishmaniose visceral? Uma abordagem contrafactual

• Autor(a) principal: Santos, Cleber Vinicius Brito dos
• Data de Publicação: 2020
• Outros Autores: [e-mail]
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/19348
• topic: Causal inference; Vector-borne diseases; Neglected tropical diseases; Emerging infectious diseases; Bayesian inference; Potential outcomes; One health; Inferência causal; Doenças transmitidas por vetores; Doenças tropicais negligenciadas; Doenças emergentes e reermergentes; Inferência Bayesiana; Desfechos potenciais; Saúde única; CIENCIAS DA SAUDE::SAUDE COLETIVA

Visceral leishmaniasis (VL) is a systemic vector-borne disease. In Brazil, it caused by the protozoan Leishmania infantum and is transmitted by sandflies of the genus Lutzomyia, with dogs as the principal source of infection in urban areas. Historically, VL was known as a rural endemic disease, since the 80's it has become endemic and epidemic in large Brazilian cities. Since then, many factors were hypothesised as driving VL expansion, as socioeconomic, climatic and environmental variables. More specifically, concerning deforestation and human-made actions in the environment, most studies tend to be qualitative in nature or use traditional reductionist approaches, ignoring the complexities that are inherent of vector-borne zoonotic infectious diseases. The present study aimed to investigate the effect of deforestation in the occurrence of Lutzomyia longipalpis, canine visceral leishmaniasis (CVL), and human visceral leishmaniasis (HVL), taking as a motivating example the São Paulo state (Brazil). To this end, we chose a counterfactual approach to estimate the effects (overall, direct and indirect) of deforestation in the occurrence of vector/CVL/HVL. We did it in two steps; first, we estimated the parameters through a double Metropolis-Hastings algorithm and, finally, we estimated the causal effects through a Gibbs sampler, using the autognet package in R. We observe that deforested cities show 2.63, 2.07, and 3.18 higher odds of vector/CVL/HVL occurrence, respectively, when compared to non-deforested municipalities. We also see a significant influence of vector, CVL, and HVL presence in the neighbours in its appearance in previous naive cities, 6.67, 4.26, 4,27 respectively. Lastly, under hypothetical changes in deforestation's prevalence from 50% to 0% in the whole state, is expected a decrease in its prevalence of the vector, LVC and LVH of 11%, 6.67% and 29.87% respectively. Deforestation in the occurrence of infectious diseases and, more specifically, VL importance, is two-folded: (i) changing's the ecosystem equilibrium and community structure, allowing its vector to reproduce and colonise; (ii) promoting a close contact through the VL cycle components. In such a way, for correct control of VL and infectious disease as a whole, it is essential an eco-friendly development with viable solutions for trade-offs between agriculture, urbanization and conservation.