Upward or Downward Mobility: Longitudinal residential trajectories and risk for preterm birth

Background: Many studies have been conducted on the potential effect of neighborhood deprivation on preterm birth and how this effect varies throughout the life-course. Most of these have been cross-sectional. We investigated the preterm birth risk associated with moving to a lower or higher deprivation neighborhood compared to staying in the same neighborhood through a longitudinal study design. We hypothesized that a decrease in deprivation exposure would be associated with reduced preterm birth risk while an increase in deprivation would result in increased risk.

Methods:A retrospective cohort was constructed through linking sibling birth records to 170,865 mothers living in the Atlanta area between 1994 and 2007. The residential addresses were geo-coded and linked to Census measures of neighborhood deprivation, creating a partial adult exposure history preceding the second birth. Fixed-effects regression modeling was performed to draw comparisons between women who shared a neighborhood at the baseline measurement.

Results: Moving to a lower deprivation neighborhood was associated with a decrease in preterm birth risk (OR=0.89, 95% confidence interval 0.84, 0.94) while moving to a higher deprivation neighborhood was associated with an increase in preterm birth risk (OR=1.27, 95%CI 1.21, 1.34). The association strengths were reduced but remained significant when adjusted for maternal race, education, and age.

Conclusions: Longitudinal measures of residential mobility and consequent changes in deprivation lend support to a causal association between neighborhood deprivation and preterm birth. Residential mobility trajectories are a tool to improve our understanding of neighborhood effects throughout the life-course.