Temporal evolution of the risk factors associated with low birth weight rates in Brazilian capitals (1996-2011)

Objective: To analyze the trend of low birth weight (LBW) and its determinants in Brazilian state capitals between 1996 and 2011. We intended to determine which variables are associated with LBW during the period studied. Methods: This is a cross-sectional study that used data from the National Information System of Live Births from 26 state capitals and Brasilia (the federal capital), divided into five geographical regions. The Average Annual Percentage of Change (AAPC) was used to assess the possible time trend in the low birth weight rates for considering all regions together and each region separately, according to each variable, and the Poisson regression was calculated in order to demonstrate time trends in low birth weight and the impact of variables (age and educational maternal level, antenatal visits, type of delivery, and gestational age) during the period. All variables were analyzed together using the Poisson regression as well. Results: From the total of 11,200,255 live births used in this study, there was a significant reduction in the number of live births, especially in the more developed regions. The low birth weight rate was 8 %, and it was stable during the period. Considering regional trends, the rate was higher in the Southeast and South regions, and significantly higher in the North, Northeast, and Central West regions. Improvements in maternal education and antenatal care coverage reduced the risk for low birth weight in all regions. Also, there was an increase in caesarean sections in all regions, with a small impact on low birth weight rates. Conclusions: Improvements in education and health care reduced the risk for low birth weight in all Brazilian regions during the period of study. Trends in low birth weight rates and the associated factors differ from region to region, showing different stages of demographic, epidemiological and developmental transition in Brazil. The present study was approved by the Research Ethics Committee at the Hospital de Clínicas de Porto Alegre (Protocol 120323)