Determinantes sociais e ambientais na incidência de dengue em Sergipe: um modelo de regressão

Abstract

In the last few years, the care costs of supplementary health companies have been suffering an exponential growth, which leads Dengue is a viral infectious disease prevalent in tropical developing countries, transmitted by the Aedes aegypti vector. In this sense, the objective of the present study was to develop a regression model, in order to determine the degree of influence of social and environmental factors on the incidence of dengue in Sergipe from 2001 to 2012. In a descriptive study, it was observed that the year of The highest and lowest incidence was the year 2008 (17,950 notifications) and 2004 (133 notifications), respectively. During the study period, the months of February, March, April, May and June showed a higher prevalence, giving indications of a seasonal behavior of the disease. In addition, about eight variables were selected for the construction of the model: essential services, water supply, direct garbage collection, piped water supply, total garbage collection, sanitary sewage, sewage connected to the collection network and total precipitation. By preliminary correlation analysis, it was found that total precipitation, essential services and sanitary sewage could influence the incidence of dengue, but the presence of multicollinearity was observed. Thus, it was necessary to apply the main components method, resulting in 3 main components CP1 (Essential services, water supply, direct garbage collection, piped water supply, total garbage collection); CP2 (sewage, sanitary sewage connected to the collection network); CP3 (total precipitation). These were adjusted to a multiple linear regression model where it was observed that deficiency in sewage and essential services (water, sewage and garbage collection) imply an increase in the incidence of dengue. On the other hand, the increase in precipitation increases the number of cases of the disease. The result of the analysis of variance shows that the obtained equation is significantly adherent. Moreover, the analysis of the regression assumptions allows us to conclude that the model is adequate. Thus, we conclude that the study of social and climatic indicators can be extremely important to subsidize public health actions in the control of dengue.