Implementation of the pact for health in Brazil: what changes did it bring to child mortality?

This study aimed to verify the association of social inequalities with infant mortality rate so as to infer what has changed in Brazil with the implementation of the Pact for Life. This was an epidemiological ecological study of the time series type with spatial correlation carried out through the Mortality Information System, considering the years from 1996 to 2016. For the development and analysis of the data, the 161 Intermediate Regions of Urban Articulation were considered. The infant mortality coefficient in the years studied was the dependent variable, and as independent variables, the ones selected were the Human Development Index, Gini Index, percentage of poor people and the coverage of the Family Grant Program. Descriptive data were analyzed in the Statistical Package for the Social Sciences. Mortality data were spatialized to determine bivariate spatial autocorrelations using the Geoda software. It was identified in the results that there was a decrease in the infant mortality rate in the decades; spatial autocorrelation showed high coefficients in the Northeast in the first decade, and higher coefficients in the North and Midwest in the second decade. In the bivariate analysis of the infant mortality coefficient with the Human Development Index, greater autocorrelation was observed in the Southeast, South and Midwest regions in the two decades; with the Gini index, the first decade showed autocorrelation in the North and Northeast, and in the second decade, there was autocorrelation in the North, Northeast, and Midwest. When assessed with poverty, autocorrelation was observed in the North and Northeast; and with coverage of the Family Grant Program, autocorrelation was concentrated in the Northeast. Even before and after the release of the Pact for Life, social inequalities were directly related to infant mortality. The Pact, which had one of its indicators the reduction in infant mortality, was effective when evaluated in isolation, however, despite the decrease in this problem.