Geoprocessamento para identificar padrões do perfil de nascimentos na região do Vale do Paraíba

Abstract

PURPOSE: to apply geoprocessing techniques for the spatial birth profile analysis of each municipality. METHODS: ecological and exploratory study, using data from the Health Information System about born alive babies in 2004, and using geoprocessing techniques. The spatial autocorrelations of the variables: cesarean section, mother's schooling, low birth weight, Apgar score at five minutes, prematurity, number of medical appointments and adolescent mothers, besides the map with the index of human development were estimated. For the detection of spatial events aggregates, Moran's IM statistics, through the program Terra View 3.13 (developed by INPE and available to the public) was used. Spatial maps with those variables were built, and Pearson's correlation coefficients, estimated. RESULTS: results have shown that the rate of born alive babies, from mothers with school level over primary school and from cesarean sections, presented a spatial pattern visually identifiable and significant spatial self-correlation. Low birth weight, prematurity, Apgar score, number of pre-natal appointments and adolescent mothers have presented a random spatial pattern, showing that, in this analysis scale, those markers have not discriminated the risk groups, despite their unquestionable predictive value for children's morbidity-mortality at individual level. There has been a positive correlation between cesarean section and schooling, and between cesarean section and human development index; and a negative correlation between adolescent mothers and human development index, with statistical significance (p<0.05). CONCLUSIONS: this methodology has allowed us to identify spatial clusters for the variables cesarean section and mother's schooling, besides deepening our knowledge on birth profile in the municipalities, presenting good potential on how to direct actions for specific areas.