Modelos de regressão espacial ajustados a dados de arboviroses (Aedes aegypti) do Estado da Paraíba: Influência de fatores socioeconômicos

Abstract

Introduction: Among the emerging and reemerging infectious diseases, arboviruses transmitted by Aedes aegypti (Diptera, Culicidae), such as dengue, chikungunya and zika, are considered important challenges for Public Health, especially in countries with socio-environmental conditions conducive to development and for the proliferation of this vector. Objective: To analyze the spatial distribution of the incidence coefficients of arboviruses in the state of Paraíba between 2009 and 2018, assessing the existence of geographic dependence and its relationship with socioeconomic and stomatological factors. Methodology: This is an ecological study, having the state’s 223 municipalities as its unit of analysis. The global and local Moran Index was used to assess the spatial autocorrelation of dengue and its association with socioeconomic variables and the Rapid Index Survey for Aedes aegypti (LIRAa). In order to verify the best explanatory model for the association of variables in the socioeconomic and environmental context, four alternatives of regression models were applied: classical regression, spatial autoregressive, spatial error and Durbin’s spatial model. The analyzes were performed using the statistical software R CORE TEAM version 3.5.3. Results: Analysis of the distribution of cases of arboviruses showed an increasing number of municipalities in Paraíba with the presence of the vector mosquito. The highest coefficients were those recorded in the Hinterlands of Paraíba, in Patos (17,134.9 cases / 100 thousand inhab.), in Princesa Isabel (10,158.0 cases / 100 thousand inhab.), and in São Bento (7,097.8 cases / 100 thousand inhab.). In the Borborema region, the highest average was in Monteiro (11,704.1 cases / 100 thousand inhab.); and in Agreste region, in Remígio (3,335.4 cases / 100 thousand inhab.). The maps of significance of the local Moran index made it possible to identify clusters of municipalities with similar characteristics, presenting areas of greater vulnerability around the region of Agreste of Paraíba with clusters for low income, Gini index and illiteracy, with results showing a pattern of dependency high-high type. However, in this same region, a low-low HDI standard. Unemployment showed dissimilarity in all regions, only the Agreste region with a small cluster of municipalities (low-low). The model that best fitted the data was Durbin’s spatial model, as it presented the lowest value of the Akaike Information Criterion. According to the model, the collection of solid waste showed a significant contribution to the incidence of dengue in the state, while the general sewage network and the municipalities that have other forms of collection pointed to an inverse correlation. In the analysis of the distribution of the municipalities of Paraíba, among the four years in which the LIRAa was carried out, the highest averages of the Building Infestation Index were those recorded in the Hinterlands of Paraíba, in Riacho dos Cavalos (x ̅6.8) and Malta (x ̅7.5); in the Borborema region, in Juazeirinho (x ̅9.5) and Picuí (x ̅8.3); and in Agreste region, in Alagoa Nova (x ̅11.2), in Fagundes (x ̅10.2) and in Cacimba de Dentro (x ̅9.3). Conclusion: There was no evidence of an association between economic differences and dengue cases, which leads to the conclusion that the disease has no distinct social stratification. The form of disposal of effluents seems to have a negative association with the incidence of the disease, since places with better sanitation conditions have lower incidences, since, due to irregular rainfall in the region, domestic effluents can become an important resource for the vector. From the point of view of the State Health Secretariats, the model obtained is succinct and viable, as it allows, with little information, to establish important statistical results that can be used by the Epidemiological Surveillance Services of the State Health Secretariats.