Epidemiology of infant mortality in Rio Grande do Sul, Brazil. The influence of agricultural production

Infant mortality rates (IMRs) for the 24 districts of the State of Rio Grande do Sul, Brazil, for 1972, have been correlated to 12 variables representative of the state's agrarian structure, as well as five other socioeconomic variables, through Pearson's correlation coefficient and multiple linear regression. The IMR was found to be significantly larger in the districts characterized by large properties, cattle raising and wage earning, rather than in the districts of small properties, agriculture and large rural population. The IMR was not found to be correlated to the variables representing education, housing, medical care or availability of banks, but there was a direct association with the sanitation variable. A multiple linear regression equation was then obtained, in which the agrarian variables accounted for 70% of the variation in the IMR, while the non-agrarian ones accounted for 13%. These findings support the hypothesis that the agrarian structure of Rio Grande do Sul may influence the observed distribution of the IMR in the state, while other variables specifically related to the preservation and restoration of health do not seem to play an important role.