Os partos cesáreos e sua influência na mortalidade neonatal nos espaços regionalizados do estado da Paraíba de 2009 a 2017

Abstract

Infant mortality is considered an event of extreme relevance in public health and demography. One of the important factors associated with neonatal infant mortality is cesarean delivery, which in Brazil reaches levels around 57%, one of the highest in the world. This study aimed to investigate the influence of cesarean deliveries in neonatal mortality in the regionalized divisions of the State of Paraíba from 2009 to 2017. It was selected as the set of maternal and child variables common to the SIM and SINASC micro databases: gender and race / child color, age of mother , mother's level of education, number of children born alive, number of children born dead, type of pregnancy, duration of pregnancy, birth weight and type of delivery. The problem of information not filled in or ignored of the variables in the Neonatal Death Statements was treated with a combination of procedures based on the application of the Deterministic Linkage and Multiple Imputation techniques. The estimation of the coverage of neonatal deaths and the Unknown information índex of the variables (IIDV) for the three regionalized divisions of the State: Urban, Adjacent Rural and Adjacent Intermediate were obtained. For application of the Multilevel Binary Logistic Model, two levels were considered: each neonatal death record (level 1) and municipal characterization (level 2). Three models were performed: in the first, the variable Type of Delivery was considered as a dependent variable; in the second and third models the variable Type of delivery was considered as independent. In the light of the micro-data of 5,149 neonatal deaths in Paraíba from 2009 to 2017, higher proportions of neonatal deaths of children born via cesarean delivery were found in Urban municipalities. The variables Type of delivery, Mother's level of education and number of children dead stood out as differentiating factors for neonatal deaths in Urban municipalities in relation to the Adjacent Rural and Adjacent Intermediate municipalities. Linkage and Multiple Imputation solved the problem of variables with missing data and prevented observations from the microdata bank from being removed which would cause biased interpretations. The estimation of the coverage of neonatal deaths and the IIDV dimensioned the level of under-registration. The modeling results revealed that the chance of birth via cesarean section among children who died within 28 days increases as the mother's age and schooling increases and with the presence of multiple pregnancies, and the chance decreases as she increases the number of children born dead. The results reinforced the urgent need for more effective public and private policies to reduce the rate of elective cesarean sections to acceptable levels (10% - 15%) mainly in Urban municipalities for a consequent reduction in neonatal mortality levels.