| Description: |
Background: Preterm birth is a leading cause of neonatal morbidity and mortality worldwide. Despite medical advancements, the United States (U.S.) has seen a rise in preterm birth rates. Few studies have examined the underlying causes of this trend. The objectives of this thesis were to examine temporal trends of preterm birth rates in the U.S using time series methods. Methods: A time series analysis was conducted using U.S. National Vital Statistics System Natality data from 2016 to 2023, restricted to singleton live births among nulliparous individuals. Monthly preterm birth rates (per 1,000 live births) were calculated, including subcategories: extremely, very, and moderate-to-late preterm. Subgroup analyses were performed for maternal, paternal, and delivery characteristics. To assess the impact of COVID-19, an interrupted time series analysis was applied, with the COVID-19 pandemic onset (March 11, 2020) and the Delta variant emergence (June 1, 2021) as intervention points. Results: A total of 9,169,343 live births were included. The preterm birth rate increased from 92.5 to 102.5 between 2016 and 2023. Time series analysis revealed strong seasonality, with peak rates in June (median=101.238, interquartile range (IQR): [97.186, 102.779]) and troughs in September (median=89.135, IQR: [87.023, 91.198]). Subgroup analyses showed disproportionately higher monthly preterm birth rate increases among older mothers (Slope: 0.248/month, p |