Generalized linear models applied to the analysis of the effectiveness of the Sterile Insect Technique

Aedes aegypti mosquito, popularly known as the dengue mosquito, represents a serious public health problem, because it is the vector responsible for several serious diseases such as Chikungunya fever, Zika virus and Dengue, the latter of which has a higher incidence in the Brazilian population. That mosquito has a high capacity for adaptation in the most diverse types of environments, mainly in places with high population density and lack of basic sanitation, factors that favor its proliferation. The control methods used by the epidemiological surveillance agencies are divided into three modalities: mechanical control, chemical control and biological control. Among the biological control methods, the Sterile Insect Technique (SIT) technique stands out, which consists of introducing sterile males into the mosquito population. This article presents reproduction analyzes of a population of natural mosquitoes compared to a population of mosquitoes with the insertion of males sterilized by ionizing radiation, seeking to observe the impact caused by this control method. The statistical analyzes showed the Negative Binomial regression model presented a better fit to the daily egg count data, when this model was compared to the Poisson model. In addition, the introduction of irradiated males decreases the average oviposition rate by approximately 30%. (author)