Area-level risk factors for adverse birth outcomes: trends in urban and rural settings

Abstract

Background: Significant and persistent racial and income disparities in birth outcomes exist in the US. The analyses in this manuscript examine whether adverse birth outcome time trends and associations between area-level variables and adverse birth outcomes differ by urban-rural status.

Methods: Alabama births records were merged with ZIP code-level census measures of race, poverty, and rurality. B-splines were used to determine long-term preterm birth (PTB) and low birth weight (LBW) trends by rurality. Logistic regression models were used to examine differences in the relationships between ZIP code-level percent poverty or percent African-American with either PTB or LBW. Interactions with rurality were examined.

Results: Population dense areas had higher adverse birth outcome rates compared to other regions. For LBW, the disparity between population dense and other regions increased during the 1991-2005 time period, and the magnitude of the disparity was maintained through 2010. Overall PTB and LBW rates have decreased since 2006, except within isolated rural regions. The addition of individual-level socioeconomic or race risk factors greatly attenuated these geographical disparities, but isolated rural regions maintained increased odds of adverse birth outcomes. ZIP code-level percent poverty and percent African American both had significant relationships with adverse birth outcomes. Poverty associations remained significant in the most population-dense regions when models were adjusted for individual-level risk factors.

Conclusions: Population dense urban areas have heightened rates of adverse birth outcomes. High-poverty African American areas have higher odds of adverse birth outcomes in urban versus rural regions. These results suggest there are urban-specific social or environmental factors increasing risk for adverse birth outcomes in underserved communities. On the other hand, trends in PTBs and LBWs suggest interventions that have decreased adverse birth outcomes elsewhere may not be reaching isolated rural areas.

Figure 1

Spatial Distributions of (A) Rural–urban Commuting Area Code categories and (B) Population Density Tertiles Across Alabama.

Figure 2

B-Splines Representing Long-Term Alabama Adverse Birth Outcome Trends for (A) PTB by Rural–urban Commuting Area Code category (B) PTB by Population Density Tertile (C) LBW by Rural–urban Commuting Area Code category and (D) LBW by Population Density Tertile.