Modelagem de equações estruturais: uma ferramenta para descrição dos determinantes dos contextos de vulnerabilidade ao HIV

Abstract

The infection with the human immunodeficiency virus (HIV) and its manifestation though the acquired immune deficiency syndrome (AIDS) became an epidemiologic milestone in the ways health processes are understood the disease of the mankind, since the discovery of the first case up until the multiple faces the disease has turned into. With different forms and intensities, every one is vulnerable to HIV; however, the young population has become potentially more vulnerable to the virus, considering their multiple interactions in the context they interact. As a consequence, this worked sought to describe the way in which the dependency relations among the constructs in the context of vulnerability to the HIV of that group, through the use of structural equations modeling (SEM), identifying the elements that make them vulnerable, examining the relations of simultaneous dependence among those components and then presenting the individual share of contribution of the vulnerability determinants within the context embraced by the young individuals. Therefore, it was conducted a descriptive inquisitive study, in the municipal public schools in João Pessoa, PB, which comprised 417 students from the junior high level. A survey was applied within the school grounds, in which the variables constituted the pattern for measuring the youngsters vulnerability and they were analyzed via SEM. The confirmatory factor analysis (CFA) proved the (b = 0,43), behavioral patterns, knowledge (b = 0,44) and beliefs (b = 0,45) as determinant for individual vulnerabilities and cultural reference numbers and gender relations explaining the social vulnerability (b = 0,525; 0,46 e 0,32, respectively). These patterns asserted that about 99% of the individual and social constructs of the vulnerability (R2 = 0,998), revealing the synchrony between the epidemiology and the statistical methods. Thus, the SEM proved itself efficient in solving problems regarding the health endogenous constructs, producing valid models.