Análise dos determinantes socioeconômicos do diabetes autorreferido no Brasil: uma abordagem econométrica

Abstract

This study analyzes the socioeconomic determinants associated with self-reported diabetes mellitus in Brazil, using microdata from the 2019 National Health Survey (Pesquisa Nacional de Saúde – PNS). It is based on the hypothesis that diabetes is not randomly distributed in the population, but follows systematic gradients across age and socioeconomic position. The empirical strategy combines the estimation of a binary logit model, the calculation of the Erreygers-corrected Concentration Index, and decomposition techniques, including the nonlinear Oaxaca–Blinder decomposition. The results indicate a strong concentration of diabetes among older individuals and a more moderate socioeconomic gradient when ranked by income. Age, hypertension, and obesity are strongly associated with a higher probability of diagnosis, while higher levels of education exhibit a relevant protective effect, even after controlling for income and lifestyle factors. The decomposition analyses suggest that educational differences in diabetes prevalence cannot be fully attributed to observable characteristics included in the model, although the decomposition does not allow precise statistical identification of the relative contribution of explained and unexplained components. Similarly, in the case of racial differences, while disparities in diabetes prevalence are observed across groups, the decomposition suggests that these differences are partly associated with the unequal distribution of socioeconomic characteristics and access to health services, and should therefore be interpreted with caution. Overall, the findings reinforce the interpretation of diabetes as an outcome shaped by social processes accumulated over the life course. The study concludes that effectively addressing diabetes in Brazil requires intersectoral public policies that combine improved access to healthcare with broader efforts to reduce structural inequalities in education.