Critical evaluation of national vital statistics: the case of preterm birth trends in Portugal

• Autor(a) principal: Correia, S
• Data de Publicação: 2015
• Outros Autores: Rodrigues, T, Montenegro, N, Barros, H
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10216/114893
• topic: Preterm birth; Perinatal datasets

INTRODUCTION: Using vital statistics, the Portuguese National Health Plan predicts that 14% of live births will be preterm in 2016. The prediction was based on a preterm birth rise from 5.9% in 2000 to 8.8% in 2009. However, the same source showed an actual decline from 2010 onwards. To assess the plausibility of national preterm birth trends, we aimed to compare the evolution of preterm birth and low birthweight rates between vital statistics and a hospital database. MATERIAL AND METHODS: A time-trend analysis (2004-2011) of preterm birth (<37 gestational weeks) and low birthweight (<2500 g) rates was conducted using data on singleton births from the national birth certificates (n = 801,783) and an electronic maternity unit database (n = 21,392). Annual prevalence estimates, ratios of preterm birth:low birthweight and adjusted prevalence ratios were estimated to compare data sources. RESULTS: Although the national prevalence of preterm birth increased from 2004 (5.4%), particularly between 2006 and 2009 (highest rate was 7.5% in 2007), and decreased after 2009 (5.7% in 2011), the prevalence at the maternity unit remained constant. Between 2006 and 2009, preterm birth was almost 1.4 times higher in the national statistics (using the national or the catchment region samples) than in the maternity unit, but no differences were found for low birthweight. CONCLUSION: Portuguese preterm birth prevalence seems biased between 2006 and 2009, suggesting that early term babies were misclassified as preterm. As civil registration systems are important to support public health decisions, monitoring strategies should be taken to assure good quality data.