Anais do 11 º Congresso Brasileiro de Epidemiologia

Aim: To identify spatial clusters of type 2 diabetes mellitus (DM2) among participants of Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), analyzing individual and neighborhood socioeconomic characteristics of these areas. Methods: This was cross-sectional study conducted with 4,335 participants of ELSA-Brasil (2,486 from the city of Belo Horizonte; 1,849 from the city of Salvador) whose data were collected from 2008 to 2010. The neighborhood socioeconomic characteristics were obtained from Instituto Brasileiro de Geografia e Estatística (IBGE 2010) national census. We conducted a spatial data analysis to detect the DM2 clusters using the SaTScan to pontual data. Moreover, we analyzed the individual and neighborhood socioeconomic characteristics of participants living inside and outside of DM2 identified disease clusters using hierarchical logistic regression model. Results: We identified a cluster with high chance of DM2 in Northeast region of Belo Horizonte and North region of Salvador. In the city of Salvador, we also identified a cluster with low chance of DM2, located in South region. In the both cities, participants living inside the higher cluster were independently more likely to have brown or black skin color, low schooling, and manual work activity, and the lower income area. On the other hand, participants inside the lower cluster were less likely to have black skin color, high schooling, and the higher income area. Conclusion: Individual and neighborhood socioeconomic characteristics influenced the geographic distribution of participants with DM2. Clusters with higher and lower chances of the disease occurrence characterized as, respectively, poor and wealthy locals.