Poisson model and its generalizations applied to dengue data, Brazil

• Autor(a) principal: Freitas, Jucarlos Rufino de
• Data de Publicação: 2020
• Outros Autores: Oliveira, Marília Gabriela Ferreira de Miranda, Cunha Filho, Moacyr, Silva, Frank Sinatra Gomes da, Vasconcelos, Josimar Mendes de
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Research, Society and Development
• Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/8874
• topic: DENV; Linear predictor; Dissemination; Modeling.; DENV; Predictor lineal; Diseminación; Modelado.; DENV; Preditor linear; Disseminação; Modelagem.

Objective: analyze and compare the weekly behavior of dengue cases in the five most populous municipalities in the Pernambucan mesoregions, namely Caruaru, Palmares, Recife, Petrolina and Serra Talhada. Method: the weekly epidemiological records of dengue were used, from 2009 to 2018, made available through the Citizen Information Service (SIC). Probability models were applied, more precisely, the Poisson models and their generalizations. Results: the Negative Binomial model stood out in relation to the Quasi-Poisson Model, reducing the dispersion parameters with more precision due to the nature of the overdispersed data. In addition, the analyzes indicated that precipitation and temperature were significant factors that affected the number of cases in some municipalities. Conclusion: modeling has made it a useful tool for local authorities to plan decision-making and intervention in the most propitious periods of proliferation.