A spatio-temporal analysis of very low-birth-weight and low-birth-weight prevalence in Brazil from 2010 to 2023

Background: Given the strong link between low-birth-weight (LBW) and neonatal mortality, reducing neonatal mortality to at least 12 per 1,000 live births is one of the Sustainable Development Goals (SDGs) for 2030. Objective: The objective of this ecological study is to perform a spatio-temporal analysis of the prevalence of LBW and very low-birth-weight (VLBW) in Brazil between 2010 and 2023. Methods: This is an ecological study using data from 510 immediate regions in Brazil obtained from official health information systems. A Bayesian spatio-temporal model using integrated nested Laplace approximations (INLA) was employed, where spatial and temporal effects respectively used a conditional autoregressive dependence structure and a random walk of order two. Results: A total of 38,772,489 birth records from 2010 to 2013 were identified, being the average VLBW and LBW rates were estimated at 1.03% and 7.10%, respectively. An increase in the temporal trend of VLBW and LBW relative risks on a national scale was observed in the period, being a more expressive increase starting in 2020. The highest rates were observed in the Brazilian Southeast and South regions, compared to other regions. Regions with a higher number of physicians per inhabitant tend to have a higher number of cases of VLBW per 100 live births. Higher percentages of illiteracy are associated with lower VLBW rates, while higher gross domestic product (GDP) per capita is associated with higher VLBW rates in the immediate regions. Conclusions: The prevalence of LBW and VLBW has increased in Brazil and, in line with the LBW paradox, is higher in less vulnerable regions.