Spatial and temporal patterns in preterm birth in the United States.

Area of Science:

• Reproductive Health
• Epidemiology
• Biostatistics

Background:

• Etiologies of preterm birth remain largely unknown despite extensive research.
• This study aims to generate novel research hypotheses by analyzing spatial and temporal patterns of preterm birth.
• Utilized a large-scale dataset encompassing 145 million US births from 1971-2011.

Observation:

• Examined state-level trends in early (<34 weeks) and late (34-36 weeks) preterm birth rates.
• Conducted K-means cluster analyses to identify county-level gestational age distribution patterns over time.
• Observed periodicity in county-level trends for early preterm births.

Findings:

• Identified significant associations between trends in early and late preterm birth rates.
• Found significant associations between trends in early preterm births among white and African American populations.
• Cluster analyses revealed periods of both heterogeneity and homogeneity in gestational age distribution trends across US counties.

Implications:

• Geographic and temporal patterns suggest complex, shared influences on US preterm birth rates.
• Observed periodicity in preterm birth trends may offer insights into underlying causal factors.
• Findings provide a foundation for developing new, data-driven hypotheses for future preterm birth research.