Project 462123

Neonatal mortality has declined substantially during the last decades, especially in infants born preterm, which is primarily due to the highly specialized care provided in neonatal intensive care units (NICU). In Canada, 8% of infants are born preterm and up to 10% of babies are admitted to NICU. Surviving preterm infants face higher risk of neurodevelopmental disorders later in life, and these risks increase with decreasing gestational age at birth. The purpose of this project is to employ cutting edge epidemiological approaches in the analyses of uniquely rich population data sources available in British Columbia (BC), Canada, and Sweden. This project aims to fill the knowledge gaps regarding babies' risk factors associated with increased morbidity, including the reasons for preterm delivery and postnatal interventions, which influence neurodevelopmental disorders in childhood and early adulthood. Most previous studies were hospital-based and included a relatively small number of infants that were followed up throughout the childhood and adolescence. We propose to create a virtual cohort of approximately 3.7 million singleton infants with a follow-up of up to 22 years. We will use artificial intelligence (machine learning techniques) to create prediction models that will identify individual risk factors that place preterm infants at higher risk of death or illness later in life. We will also use special analytical designs to investigate differences between boys and girls and between siblings. These results will be compared between two high-income countries with different preterm birth rates, Canada (8.3%) and Sweden (5.6%). The between-country comparisons will help us study how various causes of preterm birth affect long-term prognosis. We anticipate that this will be the largest population-based study to date to provide a robust evidence about the effects of newborn conditions on health and developmental trajectories of children and adolescents born preterm.