The inner state differences of preterm birth rates in Brazil: a time series study

Background: Preterm birth is a serious public health problem, as it is linked to high rates of neonatal and child morbidity and mortality. The prevalence of premature births has increased worldwide, with regional differences. The objective of this study was to analyze the trend of preterm births in the state of Parana, Brazil, according to Macro-regional and Regional Health Offices (RHOs). Methods: This is an ecological time series study using preterm births records from the national live birth registry system of Brazil's National Health Service - Live Birth Information System (Sinasc), for residents of the state of Parana, Brazil, between 2000 and 2013. The preterm birth rates was calculated on a yearly basis and grouped into three-year periods (2000-2002, 2003-2005, 2006-2008, 2009-2011) and one two-year period (2012-2013), according to gestational age and mother's Regional Health Office of residence. The polynomial regression model was used for trend analysis. Results: The predominance of preterm birth rate increased from 6.8 % in 2000 to 10.5 % in 2013, with an average increase of 0.20 % per year (r(2) = 0.89), and a greater share of moderate preterm births (32 to < 37 weeks), which increased from 5.8 % to 9 %. The same pattern was observed for all Macro-regional Health Offices, with highlight to the Northern Macro-Regional Office, which showed the highest average rate of prematurity and average annual growth during that period (7.55 % and 0.35 %, respectively). The trend analysis of preterm birth rates according to RHO showed a growing trend for almost all RHOs -except for the 7th RHO where a declining trend was observed (-0.95 a year); and in the 20th, 21st and 22nd RHOs which remained unchanged. In the last three-year of the study period (2011-2013), no RHO showed preterm birth rates below 7.3 % or prevalence of moderate preterm birth below 9.4 %. Conclusions: The results show an increase in preterm births with differences among Macro-regional and RHOs, which indicate the need to improve actions during the prenatal period according to the specificities of each region.